387 Planning a System Landscape of the SAP HANA Platform (1)

© 2015 SAP SE. All rights reserved. 1

0387: Planning an SAP HANA System LandscapeRon Silberstein, SAP HANA Product ManagementMay 2015


Disclaimer

This presentation outlines our general product direction and should not be relied on inmaking a purchase decision. This presentation is not subject to your licenseagreement or any other agreement with SAP. SAP has no obligation to pursue anycourse of business outlined in this presentation or to develop or release anyfunctionality mentioned in this presentation. This presentation and SAP's strategy andpossible future developments are subject to change and may be changed by SAP atany time for any reason without notice. This document is provided without a warrantyof any kind, either express or implied, including but not limited to, the impliedwarranties of merchantability, fitness for a particular purpose, or non-infringement.SAP assumes no responsibility for errors or omissions in this document, except ifsuch damages were caused by SAP intentionally or grossly negligent.



RETURN ON INVESTMENT


List one or more practices that can be obtainedspecific to your topic area.

BEST PRACTICES


Agenda: Planning an SAP HANA System Landscape – 1-

First Things First: Plan to Plan

Preparation

• Sizing SAP HANA systems

• Evaluate Hardware Deployment Options

• Cloud and Hybrid Scenarios

Evaluate SAP HANA System Landscape Deployment Options

• Multitenant Database Containers / MCOD, MCOS

• Virtualization

• NW AS ABAP and NW AS Java on SAP HANA Hardware



Develop Strategy for Ensuring Business Continuity

• Overview, Persistence, Redundancy, Failover, etc

• Backup/Recovery

• HA/DR with Storage Replication or System Replication

Consider Extended System Landscape Implications

• Overview of Key Components

• Application Lifecycle Management / Transport

• Data Management Options

• Big Data

More Information




Preparation

• SAP HANA System Sizing





• Virtualization

• NW AS ABAP and NW AS Java on SAP HANA hardware


SAP HANA Project Implementation: Applications, UseCases, Implications

First things First: Gain a comprehensive understandingof your organization’s strategy in regards to SAP HANA

Take a comprehensive, forward-looking assessment of pain points, opportunities,and likely uses of SAP HANA in the short-term, mid-term, and long-term

Business Case(s): Ensure that a sound rationale exists for each project, with cleargoals, objectives, success criteria, appropriate investment, staffing, executivesponsorship, etc.

Project / likely project overview:– What projects (applications, use cases, etc) are planned for the next 6 months?

For one year’s duration? 2-3 years? 5 years+?– Understand where dependencies may exist between different projects,

applications, etc, and account for this in planningMap short-term, mid-term, and long-term project strategy to hardware /deployment planning, capacity planning

– Foresee future needs as much as possible and integrate the requirements theybring into a forward-looking, comprehensive plan


Supportsany Device Any Apps

Any App ServerAny Apps

Any App ServerSAP Business Suiteand BW ABAP App ServerSAP Business Suiteand BW ABAP App Server

JSONR Open ConnectivityMDXSQL

Other AppsLocationsReal-timeHADOOPMachineUnstructuredTransaction

SAP HANA PlatformSQL, SQLScript, JavaScriptSQL, SQLScript, JavaScript

Integration Services/Security/ Governance/LCM/Landscape ManagementIntegration Services/Security/ Governance/LCM/Landscape Management

SpatialSpatial

Business FunctionLibrary


Search/GraphSearch/Graph Text MiningText Mining

PredictiveAnalysis Library


DatabaseServicesDatabaseServices

Stored Procedure& Data Models


Planning EnginePlanning Engine Rules EngineRules Engine

Application & UIServices


SAP HANA Platform

SAP HANA Platform converges Database, Data Processing, Application Platform capabilities & provideslibraries for Predictive, Planning, Text, Spatial, Graph and Business Analytics to enable business to operate

in real-time. SAP HANA Platform is available as an On Premise Appliance or via Cloud offerings: SAPHANA One on AWS, SAP HANA Enterprise Cloud (HEC).





Hana One

HEC/HCP

Analytics(Visualize, Predict, Engage)





Preparation






• Virtualization



The Sizing Process : Three-Party Collaboration Model

ContributionsCertified benchmarks scalablehardwareDifferent configurations together withtechnology partnersPerformance studiesCustom load tests in collaboration withcustomersService level agreementsResponsible for final sizing

Expectations from benchmarkingand sizing

Optimal performanceSuggestion for hardware configuration

ContributionsResponse time requirementsThroughput requirementsProvides business data input

ContributionsDevelopment and provision ofbenchmark toolkitsRegression testing for new releasesStandard sizing guidelines as part ofquality assurance processSizing verification processes

Hardware Vendors Customer

SAP

Sizing RecommendationCPU (SAPS)Memory (GB)Database space (GB)Disk I/O operations per secFrontend bandwidth mbps

Customer

Three parties collaborate in the benchmarking and sizing processHardware configuration

Feedback andexperience

Technicalrequirements

Businessrequirements

Feedback andexperience


SAP HANA Sizing

SAP HANASizing

1.Initial Sizing

2.Migrating to SAP

HANA asDatabase

3.Sidecar

Scenarios

4.SAP HANAon VMware

Sizing

Customer

SAP HANA Quick Sizer - http://service.sap.com/hanaqs


How To Do An Initial Sizing For Suite On HANA?

1.Initial Sizing

StandaloneHANA

SAPBusiness

Suitepowered by

HANA

SAP HANAEnterprise

Search

IndustrySolutions

powered bySAP HANA

SAPNetWeaver

BWpowered bySAP HANA

Othersizing

guidelinesusing

HANA asdatabase

HANAQuickSizerand SAP

Note1514966

UsingHANA

versionQuick Sizer

UsingHANAQuick

Sizer andGuidelines

UsingHANA

versionQuickSizer

(1793345)

Customer


Systems Migrated to SAP HANA: Key Points

SAPHANASizing

2.Migrating to SAP HANA

Applications “ProductiveSizing”

StandaloneHANA

SAP BusinessSuite (SoH)powered bySAP HANA

IndustrySolutions

powered bySAP HANA

SAPNetweaver

BW poweredby SAPHANA

Apply SAPNote

1514966

Apply SAPNote 1872170

Apply SAPNote 1736976

SAP Note 1872170:• Memory: Use SAP Note 1872170• CPU: Check database CPU requirements• Disk Size: Check size of current database• The application servers can be re-used if

the OS/Hardware platform is supported

Customer

• Is suitable for sizing of allBusiness Suite products (ERP,CRM, SCM, SRM, etc…)

• Can be used for sizing of allproducts running on SAPNetWeaver w/ exception of BW

• The script calculates the size ofthe data in memory. It does notestimates the work space. (Workspace is considered as equal tothe data size).

SAP Note 1736976:• Sizing Report for BW on HANA




Preparation

• Size your SAP HANA system





• Virtualization



SAP HANA Platform

SAP HANA Technology Partners SAP HANA Hardware Partners

Starting withHANA SPS08:

Linux on IBM Power CPUspossible start of pilot phase after SPS09


SAP HANA Evolution StoryA continuous journey forward…

Four years ago, SAP embarked on a HANA journey using the appliance delivery model– SAP HANA Appliance: Pre-engineered in-memory appliance that combines pre-installed HANA software components with

pre-configured, SAP-certified hardware delivered by SAP’s leading partners

In 2013, SAP introduced HANA tailored data center integration delivery model (SAP HANA TDI)– SAP HANA TDI: An alternative delivery model providing more choice to customers by allowing them to leverage their

existing hardware components and operational processes to build their own HANA deployment infrastructure

2011-2012 Feb 2014 Nov 2014

SAP HANA Appliance GEN 1CPU based on Intel Xeon EX

Nehalem and EX E7 v1(Westmere) processors

Oct 2013

SAP HANA TDI Phase 1Shared Enterprise

Storage

SAP HANA Appliance GEN 2CPU based on Intel Xeon EX E7 v2

(Ivy Bridge ) processors

SAP HANA TDI Phase 2, Phase 3TDI Phase 2: Shared EnterpriseNetworkingTDI Phase 3: Entry-level HANA E5servers

Coming Soon: SAPHANA Appliance GEN 3CPU based on Intel Xeon

EX E7 v3 (Haswell)processors

Coming Soon: SAPHANA Appliance GEN 3CPU based on Intel Xeon

EX E7 v3 (Haswell)processors

Today:Hundreds of certified

SAP HANAconfigurations ready tomeet every customeruse case and budget

needs!


SAP HANA Appliance Approach

The appliance delivery of SAP HANA provides an easy, turn-key approach for customers

Standardized and highly optimized

Preconfigured hardware setup and preinstalled software package

Fully supported by SAP

HANAServer

HANAServer

Storage

SAPHANAServer

SAP HANA Appliance:Certified, preconfigured

hardware with all requiredsoftware pre-installed

(including firmware, storagesoftware, OS, and HANA

in-memory platformcomponents)

OSOS

Server Database Network

Provision Setup Server

Easy, plug-indeployment


SAP HANA ApplianceRich partner ecosystem; Wide range of scalability and configuration options

HW Layout Notes

In-memory DB Size(# of CPUs and RAM)

Additionally forSAP Business Suite

on HANA

Westmere EX IvyBridge EX Westmere EX IvyBridge EX

Single-node Server(Scale-up) - For datamarts or

accelerators- Support for high-

availabilty and disasterrecovery

From2 CPUs/128GB

To8 CPUs/1TB

From2 CPUs/128GB

To8 CPUs/2TB

Up to4TB Up to 12 TB

Multi-nodeCluster

(Scale-out)- To be used when single

server is not enough(e.g. BW)

- Usually, 2 to 16 serversper cluster

- Largest certifiedconfiguration: 56 servers

- Largest testedconfiguration: 100+servers

- Support for HA/DR

Each server:

4 CPUs/512GBor

8 CPUs/1TB

Each server:

2 CPUs/512GBor

4 CPUs/1TBor

8 CPUs/2TB n.a. n.a.

SAP’s rapidly growing partner ecosystem provides a wide range of certified HANA hardware in avariety of scalability & configuration options:

For an up-to-date list of HANA appliances on Intel EX E7 Westmere / Ivy Bridge) processors,check this SAP Integration and Certification Center (ICC) site: SAP Certified ApplianceHardware for SAP HANA.


PROsSAP HANA appliance provides customers with faster time-to-value by deploying pre-installedand pre-configured, all-in-one box solutions

Workload optimized systems built by SAP partners following reference architecture and strict KPIs defined by SAPto ensure that optimal performance criteria are met

A wide range of configurations and scalability options available providing customers with many options to chosefrom

o Each partner solution differs in design architecture, has vendor-specific memory, built-in storage,networking and redundancy components, and has a unique system management approach

Fully supported by SAP

SAP HANA Appliance: PROs and CONs

CONs

Customers have less flexibility with hardware choices and may incur higher than necessarycosts when using SAP HANA Appliance approach

Using appliances means customers may not be able to leverage existing storage and network hardwarecomponents , people and processes, resulting in higher than necessary costs for deploying SAP HANA


SAP HANA Tailored Data Center Integration (TDI)Build your own solution using existing/preferred hardware

SAP HANA TDI is an alternative deployment approach.

EnterpriseStorage

HANAServe

r

HANAServe

r

SAPHANAServer

Enterprise Network

Network

Storage

Server (CPU,RAM)

Compared to the HANA appliance, HANA TDI offers increased flexibility and TCO savings byallowing customers to leverage their existing HW components (e.g. storage, networking) andoperation processes.

SAP HANA TDI: Use your own preferred hardware

Customers can choose their preferred hardware vendors and infrastructurecomponents (compute server, storage, networking) from a menu ofsupported SAP HANA hardware.

Key business benefits:o Lower hardware cost - Customers can leverage existing hardware in

their data centerso Easier integration of SAP HANA into customers data center – Ability

to reuse existing operational processes and skills greatly facilitatesHANA integration into data center, also reduces total HANA TCO

“Operational expenses dominate total cost of ownership, withoperation costs accounting for 79% of the equipment TCO.” (GartnerGroup)“Most industries spend less than 15 percent of their IT budgets oninnovation, meaning that the lion’s share goes to maintenance andupkeep of IT operations.” (PriceWaterCoopers)


SAP HANA TDI provides more flexibility, saving IT budget and existing investmentCustomers can re-use existing hardware in the data center, specifically:

Shared enterprise storage (TDI Phase 1) and network components (TDI Phase 2)A choice of compute servers: Enterprise class Intel EX E7 Westmere/Ivy Bridge servers, or entry-levelSAP HANA E5 commodity servers (New with SPS09, TDI Phase 3)

Mitigate risks by enabling existing IT management processes

Customer responsibilities increase with SAP HANA TDIIndividual support agreement(s) with HW, OS partner(s) requiredOnly hardware is delivered, software installation needs to be done by customer (incl. OS installation)Customer is responsible for the OS provider’s service contractWorking with 2-3 vendors increases solution implementation time resulting in longer time-to-valueCustomer is responsible for solution validation and safe guarding

While with appliance approach, SAP and HW partner(s) validate and provide on-going maintenance for the solution -with TDI approach, customer coordinates with partners end-solution validation and is responsible for on-goingsolution safe guarding and support

SAP HANA TDI: PROs and CONs

CONs

PROs


SAP HANA TDI Phases

SAP HANA TDI was delivered in 3 phases, with each phase further opening upHANA hardware infrastructure:

SAP HANA TDI Phase 1: Shared Enterprise Storage was first introduced in 2013 to allowcustomers to leverage their existing enterprise storage for SAP HANA deployments. As of today, most of the majorenterprise storage vendors have certified their solutions for SAP HANA, providing customers with a variety ofchoices for designing their HANA storage landscape.

– With the HANA TDI shared storage approach, customers can combine any supported HANA compute server(either from HANA on Intel Xeon E7 Appliances site or from HANA on Intel Xeon E5 Entry-level Systems site),with the storage solution shown on the Certified HANA Enterprise Storage site, to maximize their IT landscapeefficiency.

SAP HANA TDI Phase 2: Shared Enterprise Networking shortly followed in 2014 to definerequirements, reference architecture and best practices for SAP HANA networking. HANA TDI enables customersto leverage the existing networking infrastructure and network components in their data center, such as routers,bridges, and switches for HANA cluster inter-node and cross-site communication.

SAP HANA TDI Phase 3: Introduction of entry-level HANA E5 systems (announced withHANA SPS09) provides the more price-sensitive customers with a new choice for HANA compute nodes based onIntel Xeon E5 commodity hardware.

– Check the following site: HANA on Intel Xeon E5 Entry-level Systems for an up-to-date list of supportedhardware vendors and configurations.


SAP HANA Hardware and Scale Up / Scale Out

Single Server2 CPU 128GB to 8 CPU 1TB(Special layout for Suite on HANAfor up to 4 TB per host)

Single HANA deployments fordata marts or accelerators

Support for high availabilityand disaster recovery

Scale Out Cluster2 to n servers per cluster

Each server is either 4 CPU/ 512GB or 8 CPU/1TB

Largest certified configuration: 56 servers

Largest tested configuration: 100+ servers

Support for high availability and disastertolerance


Scalability and Scale-Up, Scale-Out

High-availability is one aspect of scalability

Scalability is the ability of a computer hardware/software totake full advantage of its changed context in a re-scaledsituation, e.g. by adding or reducing resources

Vertical Scalability (Scale-up)– Adding resources within the same computing unit, e.g.

#CPUs, #DRAMs

Horizontal Scalability (Scale-out)– Adding multiple computing units and making them perform

well together as one logical computing unit

For successful scalability, all layers of a HW/SWstack have to scale with fast performance


SAP HANA Scale-Up Concept – 1-

Basic Concept: Single-node hardware server with expansion capacity

Benefits, when compared to scale-out:

No overhead of network communication between hosts (performance)

Potential for very efficient use of available resources (especially main memory)

Cost benefits may exist depending on hardware partner

Support for virtualization

Constraints when compared to scale-out:

Hardware of same size required for HA

Less total hardware capacity than with multi-node


SAP HANA Scale-Up Concept – 2 -

Single-node SAP HANA systems with scale-up are typically deployed in the followingscenarios:

Moderate amount of data and/or concurrent operations expected

SAP Business Suite systems

Non-production systems, such as QA, test, development, sandbox, etc.

Virtualized systems

Custom data marts

Relatively small SAP NetWeaver Business Warehouse systems


Performance/ Workload and Vertical Scalability in HANA(scale-up)

Parallelism in HANAHANA is heavily optimized for state-of-the-art HW architecturesHANA uses massive intra-plan and intra-operator parallelism for query executionHANA tries to use all available resources for maximal parallelizationChallenge: NUMA-effects on large multi-socket systemsRemark: Virtualization of CPU resources reinforces NUMA-effects

“Job Scheduler” - misnomer, actually real-time execution “engine”Central place in HANA to execute plan-operator “jobs” (tasks)Re-adjust dynamically the concurrency level avoiding “over-parallelization”Can consider additional .ini configuration data to limit the concurrency level


SAP HANA Scale-Out Concept – 1-

Basic Concept: Multi-node system comprised of several server nodes acting together

Benefits, compared to scale-up:

Extensive scalability to handle large amount of data and/or concurrent operations

Table distribution automated for SAP NetWeaver Business Warehouse

A small number of standby nodes is sufficient for HA feature (fail-over) of an SAPHANA multi-node cluster


SAP HANA Scale-Out Concept – 2 -

Constraints, compared to scale-up

• Table distribution/partitioning required (currently automated forSAP NetWeaver Business Warehouse on SAP HANA only)

• Additional rack and storage system may be required when aserver node is added (depending on hardware partnerconfiguration)

Scale-out SAP HANA systems typically deployed in scenarios:

• Large amount of data and/or concurrent operations expected

• SAP NetWeaver Business Warehouse system with extensiveactive data volume

SAP Business Suite:

• Restricted availability for production environments, limited tocertain 3+1 configurations


Performance/ Workload & Horizontal Scalability in HANA(scale-out)

GeneralScale-out means expanding to multiple servers rather than a single, bigger server in one database systemMultiple servers (nodes) are switched together to one logical, but physically distributed database systemDistributed systems can overcome hardware limitations of one single server by distributing load betweenmultiple serversScale-out provides more hardware flexibility and less initial hardware costs than scale-upBut, scale-out requires more knowledge about data, application and hardware than scale-up

Scaling FactorIn theory, in contrast to scale-up, with scale-out you can scale infinitely (there are no hardware limits)But, the communication costs between servers decreases the scaling factor of distributed systemsThe shared-nothing architecture helps to minimize communication costs between serversBut, shared-nothing imposes to understand how data are used in a distributed system


SAP HANA Multi-Node Architecture

Maintains landscape information

Holds data and executes all operations

Collects performance data about HANA

Text analysis pre-processor

Repository for HANA Studio updates

Enables remote start/stop

Manages SW updates for HANA

SAP HANA Appliance

(Master) Node

Single-nodeconfiguration Multi-node cluster

configuration

Shared storage for fail-over and recovery

(Master) Name Server

SAP HANA Studio Repository

(Master) Index Server

Statistics Server

Preprocessor

SAP Host Agent

Software Update Manager

Worker Node 1

Index Server

Preprocessor

SAP Host Agent

Worker Node n

Index Server

Preprocessor

SAP Host Agent

Name Server Name Server


HANA’s Shared-Nothing Architecture

DataVolume

LogVolume

Master node

Worker node

Worker node(s)

DataVolume

LogVolume

HDB net

Topology

Each node runs on its own local data(shared-nothing)

Standby node(s) without own persistence

Shared file system for node fail-over andrecovery (HW partner implementation)

Nodes communicate via internal HDB netprotocol

All nodes belonging to the same HANAsystem must have the same HW setup

Auto fail-over


SAP HANA Scale-out – Pro & Con: Perf/Workload Aspects

ProIncreasing available hardware resources (memory + cpu)Load distribution across available resourcesParallel processing, e.g. aggregation and faster loadsFaster query processing (select), e.g. partition pruning

ConCommunication overhead (runtime) = inter-node communicationAdmin overhead (design & operations) = active data distribution


BW workload characteristicsMostly OLAP workload in rather “static” distribution environmentOLAP load benefits from parallel processing on distributedpartitions

Master NodeHandles OLTP load and DDL statements: ABAP system tables,meta data, operational tables and all row tables stored

Worker Nodeshandle OLAP load exclusively: BW data (master data +cubes/DSOs/PSAs) distributed evenly across all workersCube, DSO and PSA tables are partitioned dependent on the tableplacement rules

Useful informationglobal.ini [table placement] method= 2 (keeps row-store tables onthe master node)note #1908075 (BW scale-out configuration recommendations)

Symmetric scale-out across the system

Master-Node

OLTPtables

Worker-NodeOLAPtables

Worker-Node

OLAPtables

Standby-Node

BW ABAP

App-Server

…

SA PHANA Scale-out SolutionsBW on HANA


Symmetric hardware across the system

HANA Scale-out SolutionsSuite on HANA

Suite workload characteristicsMixed OLTP/OLAP workloads across all nodes (2PC, Cross-joins)Table access patterns dynamic and differ from customer tocustomer

“Dynamic” Table distributionBuffered SQL statements in the statement cache are parsed andprioritized based on DB statisticsDetermine disjoint table groupsTable groups are placed on same node while balancing memoryand CPU

Useful informationnote #1899817 (Suite scale-out configuration recommendations)Note #1781986 (Suite on distributed HANA database)

Master-Node

Tablegroups

Worker-Node

Standby-Node

Suite ABAP

App-Server

Tablegroups

Tablegroups

Worker-Node


Example: Suite Scale-out Customer PoC


Takeaways: Scale-up / Scale Out Performance Aspects

Scale-upScale-up first !Most common and easiest way for databases to scaleNo changes in the database or application requiredHANA parallelism strongly supports latest multi-cpu, multi-core HW architecturesHANA NUMA-aware scheduling and memory allocation is under development

Scale-outRequires good knowledge about data and workload to achieve performance KPIsScale-out works good in static distribution environments, like OLAPScale-out in mixed OLAP/OLTP environments can be challengingOptimizing scale-out performance is an iterative task




Preparation

• Size your SAP HANA system





• Virtualization



Definition: Public and Private Cloud and Managed Service

http://www.idc.com/prodserv/FourPillars/Cloud/downloads/239772.pdf

Some categories

• IaaS – PaaS – SaaS

• Public – private

• Dedicated – shared

• Managed Cloud – self-run

• On-premise – at Service Provider


SAP HANADeployment

singleinstance

scale-outcluster

bare metal

virtualized

appliance

tailored

on-premise

public cloud

private(managed)cloud

3rd partycloud

SAP HANA – Variety of deployment options

(Formerly only) On-PremiseBare metal single ServerScale-Out / HA & DR clusterVirtualized with VMware / LPARTailored setup (storage, network, compute)

Cloud + Managed Service + Hosting

HANA developer edition / Cloud Appliance LibrarySAP HANA BYOL / SAP HANA OneHANA Cloud PlatformHANA Enterprise Cloud

etc

Combine nearly every option w/ ever other option


SAP HANA – Variety of deployment options


SAP Cloud Powered by SAP HANAOverview Product Portfolio

SAP HANA

SAP HANA Enterprise Cloud SAP HANA Cloud Platform Line-of-Business Apps

(On Premise) Private Cloud (Hosted) Public CloudManaged-Cloud-as-a-Service Platform-as-a-Service Software-as-a-Service

Customer Systems

SAP Business Suite

SAP Business Warehouse

SAP HANA Datamart …

Build Extend Runapplications

Custom infrastructure andmaintenance

New Apps Collaboration

PeopleSAP Jam

Soccer

Health

ConsumerBusinessAriba

Commerce Platform: Hybris

Integration - leads to new & innovative business processes

Concur

Startups

Suite

People Customer

Finance Supplier

S/4 HANAS/4HANA


SAP HANA Cloud: Further Reading

SAP HANA Cloud Platform http://hcp.sap.comSAP HANA Enterprise Cloud http://www.sap.com/entcloudSAP HANA on AWS:http://www.saphana.com/community/blogs/blog/2014/02/19/...SAP HANA One http://www.saphana.com/docs/DOC-2482SAP HANA Developer Edition: http://scn.sap.com/docs/DOC-28294

SAP HANA Deployment Options:http://scn.sap.com/community/hana-in-memory/deployment-optionsSSAP HANA Virtualized Central Note https://service.sap.com/sap/support/notes/1788665




Preparation






• Virtualization



Standard SAP HANA Deployment Scenario

• One SAP HANA DBMS, one database, one application,one schema

• Simple, straightforward scenario

• Supported with no restrictions

• Key benefit: maximum resource allocation to singleapplication/scenario with no resource contention withothers

• Key tradeoff: TCO


Multiple Applications on One SAP HANA systemMultiple Components One Database (MCOD)

One SAP HANA DBMS, one database,several applications, several schemas

• Key benefit: May have TCO advantages

• Key tradeoffs:• Contention for resources may negatively impact performance• Additive sizing approach required• DB recovery available for entire DB (not available per schema)

• Supported for non-production with no restrictions

• Supported for production with restrictions: see note 1661202 (whitelist of applications / scenarios) and note 1826100 (white list relevantwhen running SAP Business Suite on SAP HANA)


Several Databases on One SAP HANA SystemMultiple Components One System (MCOS)

More than one SAP HANA DBMS (with one DB in each),1-n applications, 1-n schemas

• Key benefit: May have TCO advantages

• Key tradeoffs:• Contention for resources may negatively impact performance• Additive sizing approach required

• Supported for non-production with restrictions• Performance issue can only be reported to SAP if they still occur

when all other DBs stopped

• Not supported for production

• Current status outlined in SAP note 1681092


SAP HANA virtualized: Use Cases

Use Cases for virtualized SAP HANA deployments:

• For customers already standardizing on virtualizationtechnology, SAP HANA offers the customer TCO reductionsand additional options for planning and managing theirsystems landscapes.

• Ease of HW replacement / Avoidance of re-certification ofOS & SAP installations

• Separation of IT Ownership (HW and SW layer)• OS independent monitoring• Low-cost HA capabilities in Dev & Test environments

• Private and Public Cloud offerings also lower entry barriere.g. for startups by starting their business small and laterscale along their needs in regards to user and data volume.

• Positive impact on capital expenditures

• Current status on virtualization is outlined in SAP note1788665

More on virtualization later


Summary: MCOD / MCOS on one SAP HANA hardware unit

“MCOS”Multiple Components on oneSystem, multi-SID

1 x Appliancen x HANA DBn x DB scheman x Applications

E.g. DEV and QA system onone hardware. See SAPnote 1681092.

„Classical“ scenarioAppliance approach foroptimal performance

1 x Appliance1 x HANA DB1 x DB schema1 x Application(e.g. ERP, CRM or BW)Bare-metal or virtualized

SAP HANA

<HDB>

AS ABAPSID: ABC

Schema ABC

AS ABAPSID: ABC

SAP HANA SAP HANA

Schema ABC

AS ABAPSID: XYZ

Schema XYZ

<HDB1> <HDB2>

Productive Systems Non-Productive SystemsVirtualization (on premise)Virtualization technologyseparates multiple OS imageseach containing one HANA DB

n x Virtualized Appliancesn x HANA DBn x DB scheman x Applications

See SAP note 1788665

AS ABAPSID: ABC

SAP HANA SAP HANA

Schema ABC

AS ABAPSID: XYZ

Schema XYZ

<HDB> <HDB>

“MCOD”Multiple Components on oneDatabase

1 x Appliance1 x HANA DBn x DB scheman x Applications

Prod. usage for white listedscenarios allowed, e.g. SAPERP together with SAP FraudManagement. See SAP notes1661202 and 1826100.

AS ABAPSID: ABC

ApplicationSID: XYZ

SAP HANA

Schema ABC

<HDB>

Schema XYZ

White-Listed ScenariosV

irtua

lizat

ion

Virt

ualiz

atio

n

Bar

eM

etal

Bar

eM

etal


MCOD / MCOS on one SAP HANA hardware

For the current status, please check the following SAP notes

1661202 – Support for multiple applications on SAP HANA

1681092 – BW on SAP HANA - landscape deployment planning

1666670 – Multiple SAP HANA DBs on one appliance

1826100 – Multiple applications SAP Business Suite powered by SAP HANA

Whitepaper about Deployment options – SAP HANA System Landscape Guide

1849151 – MCOD on SAP HANA for ABAP and JAVA AS database schemas


Multitenancy - Introduction

SAP HANA multitenant database containers establishesa foundation for providing multitenancy in SAP HANAMultitenancy refers to a principle in software architecture where asingle instance of the software runs on a server, serving multiple tenants. A tenantis a group of users sharing the same view on a software they use. With amultitenant architecture, a software application is designed to provide everytenant a dedicated share of the instance including its data, configuration, usermanagement, tenant individual functionality and non-functional properties.Multitenancy contrasts with multi-instance architectures where separate softwareinstances operate on behalf of different tenants.

From: http://en.wikipedia.org/wiki/Multitenancy


SAP HANA multitenant database containersConcept and Terminology

A single database container is also called a tenantdatabaseRun multiple tenant databases on one SAP HANAsystemRun/support multiple applications/scenarios onone SAP HANA system in productionStrong Separation of data and usersBackup and restore available by tenant DBResource management by tenant

CPU, Memory

Move/copy tenant DBs/applications to differenthosts/systemsIntegration with existing data center operationsprocedures

Application

SAP HANA System

Tenant DB

Application

Tenant DB

System DB


SAP HANA multitenant database containers

New administration layer containing a System database• Landscape topology information• System-wide parameter settings• Focal point for complete backup of all databases• Resource management for all tenant DBs (CPU, memory, etc)

0 to n tenant databases identified by their names• Tenant database related parameter settings• Individual backup/restore of tenant database• Clear separation of application data and user management

One database software version for a SAP HANA system(all tenant databases)One HA/DR setting for a SAP HANA system: all tenantsare included in a HA/DR scenario

AS ABAPConnect to:HAN.DB’A’

SAP HANA

SID: HAN

HAN.DB A

AnyApplicationConnect to:HAN.<port>

HAN.DB B

HAN.SystemDB


First Focus: SAP HANA multitenant database containers

SPS09/10Cloud Scenarios

SAP HANA Cloud PlatformSAP HANA Enterprise Cloud

On-Premise ScenariosReplace most MCOS deployments (Multiplecomponents one system)Featuring several tenant databasesAddress common MCOD scenarios (e.g. ERP-CRM-BW, QA/DEV, Data Marts)Cross scenario support: Fast federation betweentenant databases (read only with SPS09)

App X

SAP HANA

SID: HAN

HAN.DB 1

App Y

HAN.DB 2

HAN.SystemDB


SAP HANA System

Scale-out scenario with multitenant database containers

Tenant databases canspread over multiplenodes (hosts) inscale-out systems

Example:If host 2 goes down,the standby hostbecomes active. Thetenant DBs normallyrunning on host 2 willbecome active on thestandby host

Tenant DB A.3

Tenant DB B.1

System DB(standby)System DB

Tenant DB C

Tenant DB B.2

Tenant DB A.2

System DB(standby)

Tenant DB D

Tenant DB A.1

HOST 1 HOST 3HOST 2 Standby (HOST 4)

System DB(standby)


Cross-DB queries w/ multitenant database containers

SAP HANA System

Tenant DB B

Tenant DB A

Scan

Scan

Join

ScanScan

Tenant DB C

Scan

HOST 1 HOST 2

Cross-databasequeries (federation)are supported inSQL engine andCalculation engine.

SPS09: Read-only


Migration of a single DB -> multitenant database system

SAP HANA single database system can bemigrated to a multitenant database system. Thisstep is irrevocable.

System database will be generatedSingle DB will be converted into a tenant DBautomaticallyNo changes to application/customer dataMigration does not occur automatically with SPS09upgrade- Must be explicitly triggered- Single DB is SPS09 default, MDC is optional


Summary: SAP HANA multitenant database containers

A new option for the SAP HANA platformReduces TCOEnables tenant operation on database levelOffers integrated administration, monitoringOffers powerful resource managementOffers strong isolationOffers optimized cross-database operation withinthe systemSupports flexible landscape managementSupports cloud scenariosSupports on-premise scenarios

See SAP note 2096000 for restrictions




Preparation






• Virtualization



Software-level Server Virtualization: SAP HANA onVMware vSphere

VMware Virtual Machines hide the physical hardwareand emulate a virtual machine for each OS running ontop of it.

Offers excellent flexibility and ease of management.

VM1 VM2

storage

hardware

hypervisor

VM

OS

HANA

Certified SAP HANA appliance orSAP HANA TDI verified hardware

Running SAP HANA virtualized can offer agility, HW consolidationand ease system provisioning.

Especially to those customer who already standardizing onVMware such a scenario may offer further TCO reductions andadditional options for planning and managing of multiple systemslandscapes.


Software-level Server Virtualization: SAP HANA on VMwarevSphereVMware vSphere Features Supported with SAP HANA

• VMware vSphere 5.5 support forSAP HANA in production alsocovers the following VMwarevSphere products / capabilities:

• The use of additional non-SAP HANAVMs on SAP HANA server

• Use of Snapshots and Cloningcapabilities

• The use of VMware vMotion inconjunction with DRS rules

• Use of VMware HA capabilities

See SAP Note 1788665 and1995460 for list of constraintswhich do apply.

vSphere 5.5

RHEL

HANA

SLES

HANA

SLES

HANA

(Prod)

250GB

vSphere 5.5

SLES

HANA

(Prod)

HA

vMotion

250GB

500GB

SLES

ETL

SLES

ETL

SAP HANAcertified servers

SAP HANATDI certified storage


Software-level Server Virtualization: SAP HANA on VMwarevSphereSummary of Benefits and Limitations

HANA on VMware Benefits:With server virtualization virtual machines are nottied to any particular physical server or host, theycan easily be moved from one physical server toanother. This improves business agility, facilitatesdisaster recovery and replication to remote sites, andallows for much greater hardware resourcesutilization resulting in significant TCO savings andhigh ROI.Running multiple workloads on the same hostappliance (system consolidation)Ease of HW replacement / Zero downtime preventivemaintenanceLow-cost HA capabilities in Dev & Test environmentsFast Provisioning and Decommissioning of entireVM, including SAP HANA databaseSnapshots and sharing of VM possibleSeparation of IT Ownership (HW and SW layer)Supported on all certified SAP HANA hardware(appliance, TDI)

HANA on VMware Restrictions:• As of SP09, support for scale-up

scenarios only and max VM sizes of up to1TB of RAM and 64vCPUs (scale-out /multi-node database installations are notsupported; HANA VMs greater than 1TBare also not supported).

• Limited to 2 and 4 socket certified SAPHANA appliance hardware (large 8socket appliances are not supported)

• CPU & memory overprovisioning is notallowed (dedicated resources required foreach VM)

• Has higher performance penalty thanhardware partitioning technologies(majority of tests showed12%performance degradation compared tobare metal)


MCOS

General supportfor single or multipleSAP HANA virtual

machines incombinationwith MCOS

for non-production

Multi VM

ControlledAvailability

for multiple SAPHANA virtual

machines on singleSAP HANA certifiedserver in production

1x HANA +other

General Supportfor single SAP HANAvirtual machine on a

dedicated SAP HANAcertified server

in production (withoutoverprovisioning and with

resource priorityconfigured over other

VMs)

Single VM

General Supportfor single SAP HANAvirtual machine on a

dedicated SAP HANAcertified serverin production

phostLPARESXi /LPAR

phostLPARESXi/LPAR

SAP HANA Virtualized – The Big PictureSupported Deployment Options for SAP HANA Virtualized

Scale-out

No Supportfor SAP HANA scale-out

configurations in virtualizedenvironment, eitherproduction or non-

production until furthertesting

had been finalized.

SAP Note 1681092SAP Note 2024433 **SAP Note 2063057 **

SAP Note 1995460SAP Note 2063057 **

host

ESXi / LPAR

VM1

SLES

SAP HANA

host

ESXi / LPAR

VM1

RHEL

HANA

VM2

Win *

ABAP

host

ESXi / LPAR

VM1

RHEL

HANA1

VM2

SLES

ABAP

HANA2

VM3

SLES

HANA3

HANA4

phost

ESXi/LPAR

VM1

SLES

VM2

SLES

VM3

SLES

VM4

RHEL

HANA2

HANA3

HANA1

MCOS

No Supportfor multiple SAPHANA database

installations on oneSystem / OSin production

* Windows guest OS currently not supported withHitachi LPAR for SAP workloads

** Access to SAP Note is restricted to participantsof Controlled Availability


SAP HANA Virtualized: Current Status Supported Hypervisors

VMwarevSphere / ESXi

VMwarevSphere / ESXi

HitachiLPAR

HitachiLPAR

Single VM

General Supportfor single SAP HANAvirtual machine on asingle certified SAPHANA host server

in production

Multi VM

Controlled Availabilityfor multiple SAP HANAvirtual machines on asingle certified SAPHANA host server in

production

Single/Multi VM

Controlled Availabilityfor single or multiple

SAP HANA virtual machineson a single certified SAP

HANA host serverin production

OtherHypervisors

OtherHypervisors

Scale-out

Not Supporteduntil further testinghad been finalized.

Single/Multi VM

Not Supporteduntil further testinghad been finalized.

SAP HANA PlatformSupported Hypervisors

SAP Note 2063057SAP Note 2024433SAP Note 1995460


SAP HANA VirtualizedCurrent Supported Hypervisors

Currently, the only SAP supported virtualization solutions for running SAP HANAvirtualized are

VMware vSphere 5.1 and SAP HANA SPS 05 (or later releases) for non-production use cases.VMware vSphere 5.5 and SAP HANA SPS 07 (or later releases) for production and non-production use cases.VMware vSphere 6.0 support by SAP HANA planned for 2015.

The following general conditions & constraints for running SAP HANA virtualized:Limited to scale-up scenario only (scale-out / multi-node database installations are notsupported).Limited to 2 and 4 socket certified SAP HANA appliance hardware (large 8 socketappliances are not supported)CPU & memory overprovisioning must not be usedSAP HANA installation was either done by an SAP HANA certified engineer on SAP HANAcertified hardware and successfully verified with the SAP HANA hardware configuration checktool (SAP HANA Tailored Datacenter Integration option), or system had been delivered pre-configured as certified SAP HANA appliance, with hypervisor installed by SAP HANA hardwarepartner.

See SAP Note 1788665 – SAP HANA Support for Virtualized Environments


SAP HANA VirtualizedSAP HANA on VMware vSphere in production

SAP has released SAP HANA on VMware vSphere 5.5 for general availability, allowing to golive with SAP HANA on VMware vSphere 5.5, provided the following conditions have beenmet:

Single SAP HANA virtual machine on a dedicated 2 or 4-socket SAP HANA certified serverMultiple SAP HANA virtual machines on a single physical servero No SAP HANA multi-node / scale-out deployment configurationso No 8-socket hardware configurations

Both, SAP HANA appliance and SAP HANA Tailored Datacenter Integration (TDI) deliverymethods are supported for SAP HANA on VMware vSphere.o The maximum size of a virtual SAP HANA instance is limited by the maximum size of a virtual

machine on VMware vSphere 5.5 release, which is 64 vCPUs and 1 TB of memory (limited byVMware, not SAP HANA).

o No CPU and/or Memory overcommittingo VMware Vmotion (hot move) or VMware-HA are supported

See SAP HANA Guideline for Being Virtualized with VMware vSpherehttp://www.saphana.com/docs/DOC-4192


SAP HANA Virtualized: Technology Roadmap

2014Support for SAP HANA onVMwarein non-production scenariosSupport for single-VM SAPHANA on VMware inproduction and non-production scenariosControlled Availability formulti-VM scenarios inproduction

2015, Roadmap, etcSupport for scale-outscenariosSupport of larger VMs (4 TB)Support for 8 socket HWSupport of additionalhypervisors

Single VMproduction

support

Complementdeployment

options

Multi VM supportGA

HANA scale out

Single HANAVM on vSphere

cluster

Multiple HANAVMs on vSphere

cluster

Extend platformsupport

8 sockethardware

vSphere 6Large VM

support (4TB)

Add variety

Support ofadditional

hypervisors

H1/2014 H2/2014 2015+This is the current state of planning and may be changed by SAP at any time.SAP HANA on VMware vSphere on SCN


SAP HANA VirtualizedComparison SAP HANA virtualized vs. native, based on VMware vSphere 5.5

Use Cases:Mission Critical / High-PerformanceScenariosAbsolute Performance Testing (E2Eelapse time)Scale-out / HANA Host Auto-FailoverSAP Central System (Business Suite)

Users> ~500 namedusers(BusinessSuite)

PerformancePerformanceCritical

Technical> 64 vCPU *> 1 TB memory*

FinancialVMs > 512 GBRAM *

Use Cases:Sandbox / Trial Systems / Developmentand Test SystemsRelative Performance Tests (old vs. newversion on VM)High-Available / Disaster RecoveryTolerant System Setup

Suite)

Users1:1 (server :user)< ~500 namedusers(BusinessSuite)

PerformanceNon-PerformanceCritical

Technical 64 vCPU * 1 TB memory *

FinancialVMs 512 GBRAM *

During performance analysis themajority of tests stayed within12% performance degradationcompared to bare metal.

However, there are around 100low-level performance tests in thetest suite exercising variousHANA kernel components thatexhibit a performancedegradation of more than 12%.

This indicates that there areparticular scenarios which mightnot be suited for HANA onVMware.

* Relates to VMware vSphere 5.5 releaseWhat use cases are a good fit for SAP HANA virtualized:


SAP HANA multitenant database containers & Virtualization

Multitenant Database Containers vsVirtualization

Multitenant Database ContainersLower TCO, single software stackCentral configuration & administration (database level)Direct database resource managementOptimized federation (performance benefits)Performance advantages (no virtualization overhead)Licensed via SAP HANA

VirtualizationStrong isolationSeparate SAP HANA revisions optionStandard federation (SDA)Additional virtualization license (e.g. VMWARE)


SAP HANA on VMware Sizing Guidelines

on the HANA system. The tools and resources used for this effort areexactly the same for both virtual and native environments

• Once the initial sizing numbers for memory, CPU and SAPS have beendetermined, they can be used to define the size of a SAP HANA virtualmachine

• Follow the sizing guidelines in the “Best Practices andRecommendations for Scale-up Deployments of SAP HANA on VMwarevSphere Deployment Guide”

Each SAP HANA instance / virtual machine is sized according to the existing SAPHANA sizing guidelines, powered by SAP HANA Application sizing and VMwarerecommendation.

4.SAP HANA onVMware Sizing

Using QuickSizer,and SAP Note

1995460, 1788665

SAP HANA on VMware Sizing Process:

• SAP Quick Sizer tool provides the memory, CPU and SAPS requirements for the application/workload running

Customer


VirtualizationTechnologies

Hardware partitioning technologies• Hitachi LPAR• Fujitsu pPAR• HP nPAR

Software-level hypervisorvirtualization technologies• Vmware vSphere• Others (planned)

Options which may increase infrastructure efficiency by leveragingsupported partitioning & virtualization technologies (as of SPS09)

1. HP nPAR HW partitioningHard partitions on with electrical isolation on CS900 servers,that behave and perform like completely separate servers

2. Hitachi LPAR firmware-level partitioningServer hardware resources are divided into multiple partitions,which appear as independent “bare metal” servers

3. Fujitsu pPAR physical partitioningfor SAP HANA certified Fujitsu servers

4. VMware vSphere VM containersESXi based, software layer virtualizationfor almost all SAP HANA certified x86 servers

SAP HANA Supported Partitioning & VirtualizationTechnologies


Comparison MDC / LPAR / VirtualizationMCOS SAP HANA

Multi-tenant DBcontainers

HW basedpartitioning

SW levelvirtualization

Productionsupport

No Yes Yes Yes

RelativePerformanceOverhead

Low Low Medium “High” (relative toMDC)

HW resourcemanagement

None SAP HANAinternal

Firmware Hypervisor

WorkloadIsolation

Application level Application Level OS level OS level

Shared SAPHANA binaries

No Yes No No

HA support No Yes Yes Yes

HW vendorindependent

Yes Yes No Yes

Max. instancesize

Unlimited Unlimited e.g. Max HitachiLPAR v2 size is1.5TB

e.g. Max VmwarevSphere 5.5 sizeis 1TB




Preparation






• Virtualization



Joined SAP HANA and SAP NetWeaverABAP Application Server and HANA Database on one hardware

SAP HANA and SAP NetWeaver AS ABAP deployed on one server is a multi-component, resource and cost optimized deployment approach

SAP HANAServer

SAP HANASystem

SAP NW ASABAP System

SAP HANA and SAP NetWeaver ASdeployed on one server

Hardware resources isolatedSeparate hardware

Cost optimized approachShared Memory and CPU resources

SAP HANAServer

SAP HANASystem

SAP NW ASABAP Server


Separateddeployment approach


Joined SAP HANA and SAP NetWeaverABAP Application Server and HANA Database on one hardware

SAP HANA and SAP NetWeaver AS ABAP deployed on one system isavailable since December 16, 2013.

SAP HANAServer

SAP HANASystem


AvailabilityFor all productive and non-productive SAP HANA SPS7 single nodeinstallations. All products based on SAP NetWeaver AS ABAP 7.4 aresupported.Requirements

Additive sizing: Additional memory resources for the SAPNetWeaver AS ABAP system needs to be available on the SAPHANA server. For more information, see memory sizing based onSAP Release Note - 1953429Separate SID‘s for both systems required

SAP HANA software installationThe exam “SAP Certified Technology Specialist (Edition 2013) – SAPHANA Installation” (E_HANAINS131) needs to be successfully passed fora person to perform SAP HANA software installations. For moreinformation, see SAP Training and Certification Shop


Joined SAP HANA and SAP NetWeaverHigh Availability setup based on System Replication

SAP HANA Server

SAP HANASystem

(Primary)


ABAP SID<ERS##>

ABAP SID<DVEBMGS##>

ABAP SID<ASCS##>

Shared FileSystem

Data Center 1

SAP HANA Server

SAP HANASystem

(Secondary)


ABAP SID<ERS##>

ABAP SID<DVEBMGS##>

ABAP SID<ASCS##>

Data Center 2

SAP HANASystem Replication





• Backup/Recovery






• Big Data

More Information





• Backup/Recovery






• Big Data

More Information


Continuousavailability

SAP HANA Continuous AvailabilityCustomer Expectation: Planned & Unplanned

Planned downtime

Unp

lann

eddo

wnt

ime

SAP HANA consumption

Extended SAP backend deployments

Hardware failure / Malfunctionincluding NetworksSoftware Malfunction / securitythreat / updateNatural / Man-made disastersFailure of compliance &operationUnplanned outages

SAP HANA Revisions & SPSsPatches for Data Services and SLTMaintenance Events for OS & HardwareCustom development & enhancementsPlanned outages…….

Data CenterReadiness


SAP HANA Data Center ReadinessQuick Overview (incl. SAP HANA SPS09)

Security & AuditingComprehensivesecurity framework

Fine-granularauthorizationsEncryptionCompliance (SoD,audit logging, ...)Secure hardware /software setup

IDM and GRCintegration3rd party viastandard /documentedinterfaces

HighAvailability

In case ofhardware orsystem failurethe standbysystem takesover in thesame datacenterSeveral options:

storage-basedshadowdatabasesInternal orexternal clustermanager

Design & SetupSeveraldeploymentoptions

Multi-TenantDatabaseContainerNetWeaverCentral instanceon HANA server

Virtualizationfor productionusageTailored DataCenterIntegrationDynamic Tiering

Backup &Recovery

Data & LogBackup

Point-In-TimeRecovery

3rd-partybackup toolsupport

Netbackup,Tivoli, Simpana,DataProtector,Networker,Sesam…

StorageSnapshots

Point-In-TimeRecovery

Data Center Readiness

SAP HANA

Continuous Improvement of Simplification & Flexibility

DisasterRecovery

Failover to adifferent HANAinstance inanother, evenfar distant datacenterAutomatic andmanualprocedurespossibleSeveral options:

storage-basedshadowdatabasesExternal clustermanager

Starting Page: Features of SAP HANA: Data Center - Enterprise Readiness and HA/DR


High Availability – Disaster Recovery: Concepts

Business Continuity

High Availability

per Data Center

Disaster recovery

between Data Centers

SAP HANA Host Auto-Failover(Scale-Out with Standby)

SAP HANA Storage Replication

SAP HANA System ReplicationPerformance OptimizedCost Optimized

SAP HANA System ReplicationPerformance OptimizedCost Optimized


SAP HANA High Availability: Scale-Out with Host Auto-Failover

Scale-out clusters address two requirementsScale to a setups, bigger than one hostOffer an easy HA option by putting one or more hosts asspare/standby

Host Auto-Failover is offered by the Name ServiceThe resulting cluster is managed by this name serviceinside of HANA.He regularly checks on the cluster members to be stillactive.In case of problems he initiates a fully automated take-overto the standby hardware.Together with the switched of mounts/disks also the identityof the failing cluster member is moved to the standbyhardware.

Starting with shared storage, HANA Scale-Out todaycan use SAN storage with Fiber Channel adapters

Storage Connector API ensures the possibility ofremounting necessary file systems to standby hostsMore details with: SAP Note 1900823 - Storage ConnectorAPI Please check its attachments for white papers etc.

Sha

red

Sto

rage

SA

NS

tora

ge

Sto

rage

Con

nect

orA

PI

Server 1

Server 2

Server 3

Server 4

Server 5

Server 6

Standby Server

Server 1

Standby Server SA

NSt

orag

e

Sto

rage

Con

nect

orA

PI

Minimalistic setup for only HA:

Nameserver

Nameserver

Nameserver

Nameserver

Nameserver

Nameserver

Nameserver

Nameserver

Nameserver


SAP HANA Architecture / Components / Scale-Out

SAP HANA Appliance

Software Update Manager

SAP Host Agent

SAP HANA Studio Repository

SAP HANA Database Node 2 Node n

…Name Server

Index Server

Statistics Server*

Preprocessor

Index Server

Preprocessor

Index Server

Preprocessor

Single host configuration

Multi-node cluster configuration

Maintains landscape information

Holds data and executes all operations

Collects performance data about HANA

Text analysis pre-processor

Repository for HANA Studio updates

Enables remote start/stop

Manages SW updates for HANA

Shared storage for fail-over and recovery

SAP Host Agent SAP Host Agent

Name Server Name Server

XS engine XS engine XS engineXS engine


In-Memory

SAP HANA Database Landscape: Including Standby Node

Persistence Layer

LOGDISK

DATADISK

LOGDISK

DATADISK

LOGDISK

DATADISK

LOGDISK

DATADISK

LOGDISK

DATADISK

*Standby Host:

Name Server (active)

Index Server (standby)

Distributed HANAdatabase even on asingle host with sharednothing concept

Standby without ownpersistence


HANA High Availability: Host Auto-Failover (standby)

Different implementation of High Availability by HW partners

Using storage solution inside Using internal disk

NameServer

IndexServer

StandbyNameServerIndexServer

NameServer

IndexServer

DataDisks

LogDisks

DataDisks

LogDisks

DataDisks

LogDisks

GPFS

GPFS


SAP HANA High Availability: Minimal Setup for Host Auto-Failover

Minimal setup for a Host Auto-Failover(Scale-Out):

2 Servers including one Standby

External storage or similar technologynecessary which ensures the data provisioningto second node via external data location

This setup aims for High Availability notperformance scaling or size.

Note:Some use cases (e.g. SAP BW powered byHANA) might have different requirementsor recommendations for minimal setups(e.g. BW has a defined setup for SAP HANAScale-Out – SAP note 1736976 attachedPDF).

MasterNameServer

IndexServer

DataDisks

LogDisks

active standby

IndexServer

NameServer


SAP HANA High Availability: Client Management with Scale-Out

Clients:During installation the clients get initial information about how tocontact to HANA database – often only one host is offeredTo prevent single point of failure, more host should be offered incase of Scale-OutThe list is only necessary to establish a first connect to HANA cluster– afterwards the client gets the full topology from the databaseName Server anywayThe complete list of hostnames including the standby host should bestored

User store:Contains the list of host names like “hana1;hana2;hana3” etc. nextto user and encrypted password informationAll tools based on this database interface named sqldbc (SAP Appl.Server, hdbsql, ODBC, python, etc.) can use this user store.

Algorithm:Round robin process is used to find this first contact point

SQL clients:SAP Appl. Server

hdbsql

User Storehana1;hana2;hana3

round robinhana1 hana2 hana3

HANA Scale-Out

DataDisks

LogDisks

hana1NameServer

Indexserver

hana2NameServer

Indexserver

hana3standby

NameServer

Indexserver


SAP HANA High AvailabilityNews with SAP HANA SPS09 and beyond

Scale-OutNetwork Requirement Paper (http://www.saphana.com/docs/DOC-4805)o A lot of additional information about how Scale-Out works internallyTable Re-distribution offers Object Pinning e.g. also to hosts in Scale-outo Objects like schemas, tables or table groups

HA/DR Provider Framework – Communication channel to the worldHA/DR Provider for Host Auto-Failover and SAP HANA System Replicationo Internal decisions transmitted to external stakeholders (e.g. external cluster manager handling virtual

IP addresses)

SAP HANA SPS10 (current planning)Integration of SAP HANA Dynamic Tiering into Scale-Out and Host Auto-Failover operation

Planned beyondExtension of Integration with Dynamic Tiering into Scale-Out

This is the current state of planning and may be changed by SAP at any time.





• Backup/Recovery






• Big Data

More Information


SAP HANA database Data backupsContain the current payload of the datavolumesAny pages that are changed during the databackup written to different locations in thedata volumes (shadow page concept)Manual (SAP HANA studio, SQLcommands), or scheduled (DBA Cockpit)

Log backupsContain the content of closed log segmentsAutomatic (asynchronous) whenever a logsegment is full or the timeout for log backuphas elapsed

Log Area(disk)

Data Area(disk)

Memory

Savepoint COMMIT

Data Backups Log Backups

SAP HANA Backup and RecoveryMemory Disk Backup


SAP HANA Backup and RecoveryTerminology

Log Area

Data Area

DataVolume

LogVolume

Log Volume

LogSegment

Log volume contains logsegmentso Number of pre-formatted log

segments is configurableo Log segments are closed when

they are full, or the log backuptimeout has elapsed

o After a log segment has beensuccessfully backed up, it isreleased for overwriting

DataData area = all data volumes1 data volume per service with persisted data(per node)

Redo logLog area = all log volumes1 log volume per service with persisted data(per node)


Shared Backup Directory

SAP HANA Backup/RecoveryData backup: Single-node and scale-out systems

SAP HANA automatically handles thesynchronization of backups for all nodes

no special user interaction requiredAll services that persist data are backed upo e.g. index servers, master name server)

Global data backup savepoint for all theseserviceso Synchronized across all nodes and serviceso Transactions are paused very brieflyo Savepoint is kept until the backup is finished

for all services. If a page is changed duringthe backup, it’s written to a different location(shadow page concept)

Data marked in the savepoint is read fromdata volumes and written to backup fileso One backup file per serviceo Parallelization

Backup File

NameServer

IndexServer

Savepoint

NameServer

IndexServer

Savepoint

MasterNameServer

IndexServer

Savepoint

Parallelization

Savepoint

Synchronizedbackupsavepoint


SAP HANA Backup and RecoveryDestinations for backups (I)

Backups to the file systemFor both data and log backupsE.g. to an NFS shareFor information on file systems:SAP Note 1820529Data backupstriggered/scheduled using SAPHANA studio, SQL commands,or DBA Cockpit, log backupswritten automatically (unlessdisabled)

SAP HANADatabase

BackupStorage,e.g. NFS

Create backup

hdbsql

SAP HANAstudio


SAP HANA Backup and RecoveryDestinations for backups (II)

Backups to 3rd party backup serverFor both data and log backups“Backint for SAP HANA” API can beimplemented by a 3rd party backup agent(certification required)Provides functions for backup, recovery,query, delete3rd party backup agent runs on the SAPHANA server, communicates with 3rdparty backup serverBackups are transferred via pipeDirect integration with SAP HANA:o Data backups to Backint can be

triggered/scheduled using SAP HANA studio,SQL commands, or DBA Cockpit

o Log backups are automatically written toBackint (if configured)

SAP HANADatabase

3rd PartyBackupServer

3rd PartyBackup Agent

hdbsql

SAP HANAstudio

Create backup


SAP HANA Backup and RecoveryBackint Certification

In December 2012 SAP released the certification process for “Backint for SAP HANA”. Certification is aninstallation prerequisite for backup tools using the “Backint for SAP HANA” interface.

SAP Note 1730932 (“Using backup tools with Backint”)Release announcement

Certified tools (as of 2014-June)

Online listing of certified tools: http://www.sap.com/partners/directories/SearchSolution.epx”SAP-Defined Integration Scenarios” = "HANA-BRINT”

Information for tool vendors: http://scn.sap.com/docs/DOC-34483

Vendor Certified Backup Tool Support ProcessSymantec NetBackup 7.5 SAP Note 1913568

IBM Tivoli Storage Manager for Enterprise 6.4 SAP Note 1913500

Commvault Simpana 10.0 SAP Note 1957450

HP Data Protector 8.0 SAP Note 1970558

EMC Data Domain SAP Note 1970559

EMC Networker 8.2 SAP Note 1999166

SEP Sesam 4.4 SAP Note 2024234

Dell Quest Netvault Backup - Planned -


SAP HANA BackupBackup in SAP HANA Studio


SAP HANA Backup and RecoveryDestinations for backups (III)

1. Using SAP HANA studio, preparethe database for the storagesnapshot. Technically, thiscreates an internal data snapshot

2. Using the storage tool, create astorage snapshot of the SAPHANA data area

3. In SAP HANA studio, confirm thestorage snapshot as successful.An entry including the externalbackup ID is written to the backupcatalog

SAP HANA Database

ExternalStorage

StorageTool

Data Area (Disk)Data snapshot

Prepare database

Create storagesnapshot

Confirm storagesnapshot

Storage snapshots as backupsSAP HANA also supports the creation of storage snapshots, which can later beused for recovery

hdbsql

SAP HANAstudio


SAP HANA Backup and RecoveryOptions for backup: Comparison

File system Backint Storage snapshot

Advantages Consistency checks on block level Consistency checks on block levelEase of use – no explicit backup filesmanagement, integrated into StudioData center integrationAdditional features, e.g. encryptionor de-duplicationBackups immediately available forrecovery

FastNegligible network loadFirst storage partners offerintegration in their tools

Disadvantages Additional storage requiredFile system fill level needs to bemonitoredAdditional time needed to makebackups available for recoveryNetwork loadIn case of recoveries, backup filesmust be returned to staging area

Network load3rd party backup tool necessary

No consistency checks on blocklevel

Size Payload only Payload only ~ Size data area, but usuallycompressed/de-duplicated bystorage

Duration IO-bound (reading from datavolume, writing to target)Network-bound (writing to filesystem)

IO-bound (reading from datavolume)Network-bound (writing to backupserver)

Negligible (logical pointers arereplicated)


Backup and RecoveryDatabase Copies

SAP HANA database copy from PROD to QA or DEV allows to change thetopology in case of a Scale-out setup on PROD side:

Backups which are produced on scale-out landscapes with n hosts can be recovered toone QA, DEV or sandbox systems.Purpose is to offer a possibility for a light system copy without the full performance scopelike PRODAbility to work on that copy limited by performance and restricted by tables/partition sizes

N 1N M

PROD

QA, DEVor Sandbox

Node 1Index Server n

Index Server 2

Index Server 1

Node nIndex ServerNode 2

Index ServerNode 1Index Server

Database inside changes


SAP HANA Backup and RecoveryDatabase Copy with SAP HANA native backup files

Using data and log backups – source and target databases may have differentnumber of hosts

Source databasewith n nodes(e.g. PROD)

Target databasewith 2 nodes

(e.g. QA)

Node n

Index ServerNode 2

Index ServerNode 1

Index Server 1

Node 2

Index ServerNode 1

Index Server 1

Index Server 2

Data backup+ log backups

(optional)


SAP HANA Backup and RecoveryDatabase Copy in combination with Storage Snapshots

Using snapshot and log backups – source and target databases must have samenumber of hosts

Source databasewith n nodes(e.g. PROD)

Node n

Index ServerNode 2

Index ServerNode 1

Index Server 1

Snapshot +log backups

(optional)

Target databasewith n nodes

(e.g. DEV)

Node n

Index ServerNode 2

Index ServerNode 1

Index Server 1


Backup and RecoveryInternal Snapshots in SAP HANA

SAP note: 1703435

Restriction: One internal Snapshot only right nowConflicts with Backup Snapshot which is needed during backup execution time.If an internal snapshot already exists when backup is started, the backup will not beexecuted and an error presented.

Roadmap: multiple named internal Snapshots are planned


SAP HANA Backup and RecoveryMore information

DocumentationSAP HANA Administration Guide,SAP HANA Technical Operations Manual

Overview presentationBackup/recovery overview presentation

Best practices2091951: Best practice: SAP HANA Backup and Restore

Important SAP Notes1642148: FAQ: SAP HANA database backup and recovery1730932: Using backup tools with Backint1869119: Check backup integrityFor further notes on backup/recovery, see HAN-DB-BAC

Backint for SAP HANA certificationCertification announcement and description


SAP HANA Backup & RecoveryNews with SAP HANA SPS09 and Beyond (may be subject to change)

Backup & Recovery3rd party backup tools (Backint)o Database copy using 3rd party backup toolso Improved handling of log backupso Improved tape handling on 3rd party backend systemsEnhanced scale-out supporto Remove host/service without necessity to write a new data

backupo UI support in SAP HANA Studio for removing servicesSupport for Multi-tenant Database Containers (MDC)with B&R of SAP HANASupport for SAP HANA Dynamic Tiering setups withB&R of SAP HANANew alerts for B&R operationso Log backup taking too longo Storage snapshot preparedo Automatic log backup disabled

SAP HANA SPS10 (current planning)Delta backups (incremental/differential)SAP HANA Cockpit: web-based backup operations3rd party backup tools: tenant copy via Backint (forMDC systems)

Planned beyondAdditional Recovery Optionso Restart-able recoveryo Partial recovery (service oriented)Additional backup options – e.g.o Support for backup operations on secondary system in

system replication scenarioso Offline log backupWeb-based administration toolso Extended functionality in HANA CockpitBackint 2.0 API and certificationo extended scope e.g. Redhat supportAdditional options for backup lifecycle management indiscussion e.g.o Integrity check for the backup catalogo Backup staging using 3rd party backup toolso Option for manual log backupo Configuration file backupo Backup compressiono Backup throttling






• Backup/Recovery






• Big Data

More Information


HA & DR Concepts in general

RPO RTO

operation resumed…operation resumed…

time

Sync orbackup

…system operational…system operational

design & prepare detect recover perf. ramp

KPIs:• Recovery Point Objective (RPO) = worst-case data-loss• Recovery Time Objective (RTO) = time to recover from outage

*synchronous solution

Solution Used for Cost RPO RTO Perf. rampBackup & Recovery HA & DR $ high high medSAP HANA Host Auto-Failover HA $ 0 med longSAP HANA Storage Replication w/ QA, Dev. DR $$ 0* med longSAP HANA System Replication HA & DR $$$ 0* low shortSAP HANA System Replication w/ QA, Dev. HA & DR $**/$$ 0* med long

** single host installations


SAP HANA Disaster RecoveryDifferent ideas of solutions

1. SAP HANA Storage Replication of SAP HANA disk areas controlled by storage technology• First synchronous implementation (available, SAP note 1755396)• Afterwards asynchronous implementation planned and in preparation with HW partners

2. SAP HANA System Replication (initial solution):DATA and LOG content is continuously transferred to secondary site under control of SAP HANAdatabase

• Fast switch-over times because secondary site can preload DATA• First synchronous implementation available since SAP HANA SPS05• Asynchronous implementation offered with SAP HANA SPS06

3. SAP HANA System Replication (extended solution):DATA content is only initially transferred to secondary site, afterwards continuous LOG transferand LOG replay on secondary site

• LOG is provided to secondary site on transactional basis (COMMIT) controlled by SAP HANAdatabase (including initial DATA transfer)

• Fastest switch-over times, sec. site preloaded and rolled forward on COMMIT basis• Synchronous and asynchronous implementation planned for SAP HANA SPS11


Data Center 2Data Center 1

SAP HANA Disaster Recovery: Storage ReplicationCluster across Data Centers

OS: Mounts

DataVolumes

LogVolume

OS: DNS, hostnames

Primary

NameServer

Indexserver

NameServer

Indexserver

NameServer

Indexserver

Secondary(inactive)

NameServer

Indexserver

NameServer

Indexserver

NameServer

Indexserver

HA

Sol

utio

nP

artn

er

Sto

rage

Mirr

orin

g

Clients Application Servers

HA

Sol

utio

nP

artn

er

DataVolumes

LogVolume

DataVolumes

LogVolume

DataVolumes

LogVolume



SAP HANA Disaster Recovery: Storage ReplicationCluster across Data Centers with QA & Dev. on 2nd site

OS: Mounts

DataVolumes

LogVolume

OS: DNS, hostnames

Primary

NameServer

Indexserver

NameServer

Indexserver

NameServer

Indexserver

SecondaryProd. (inactive), QA&DEV (active)

NameServer

Indexserver

NameServer

Indexserver

NameServer

Indexserver

HA

Sol

utio

nP

artn

er

Sto

rage

Mirr

orin

g

Clients Application Servers

HA

Sol

utio

nP

artn

er

DataVolumes

LogVolume

DataVolumes

LogVolume

DataVolumes

LogVolume

DataVolumes

LogVolume

DataVolumes

LogVolume


SAP HANA Disaster Recovery: System ReplicationCluster across Data Centers with DB controlled transfer


OS: Mounts

DataVolumes

LogVolume

OS: DNS, hostnames, virt. IPs

Primary(active)

NameServer

Indexserver

NameServer

Indexserver

NameServer

Indexserver

Secondary(active, data pre-loaded)NameServer

Indexserver

NameServer

Indexserver

NameServer

Indexserver

HA

Sol

utio

nP

artn

erClients Application Servers

HA

Sol

utio

nP

artn

er

DataVolumes

LogVolume

DataVolumes

LogVolume

DataVolumes

LogVolume

Transferby

HANAdatabase

kernel


SAP HANA Disaster Recovery: System ReplicationCluster across Data Centers with QA & Dev on 2nd site


OS: Mounts

DataVolumes

LogVolumes


Primary(active)

NameServer

Indexserver

NameServer

Indexserver

NameServer

Indexserver

Secondary(active,)

NameServer

Indexserver

NameServer

Indexserver

NameServer

Indexserver

HA

Sol

utio

nP

artn


HA

Sol

utio

nP

artn

er

DataVolumes

LogVolumes

DataVolumes

LogVolumes

DataVolumes

LogVolumes

Transferby

HANAdatabase

kernel

DataVolumes

LogVolume

DataVolumes

LogVolume

PRD QA/DEV

QA/DEVrunning

PRDshadow

operation

QA/DEVrunning

PRDshadow

operation


SAP HANA High Availability: System ReplicationMinimal setup in one Data Center for fast takeovers

Data Center 1


Primary(active)

Name Server

Index server

Secondary(active, data pre-loaded)

Name Server

Index server

HA

Sol

utio

nP

artn


HA

Sol

utio

nP

artn

er

Transferby

HANAdatabase

kernelInternalDisks

InternalDisks

DataDisks

LogDisks

DataDisks

LogDisks


SAP HANA in Data Centers:Availability of solutions

High Availability per Data Center

Host Auto-Failover (Scale-Out with Standby)Available today from several HW partners

System ReplicationAdaptations from most HW partners on the way

High Availability across Data Centers – Disaster Recovery

Storage Replication: Hardware validation successfully finished with partners, SAP note1755396Further HW partners planned to followMirroring solutions depend on HW partner technologyFurther detailed information about the solutions offered by HW partners.

System Replication: since HANA SPS5, (End 2012)Partly HW partner related, especially external cluster management (network)Similar outside implementation like Storage Replication

Step-by-Step Implementation Guide: https://scn.sap.com/docs/DOC-47702


Worldwide Data Center SetupsMulti Tier System Replication – Cascading Systems

Production Local shadowwith data preload

Remote system/shadowwith or without preload(mixed usage together withnon-prod. operation)

Data CenterData Center

Sync

Async

Tier 1 Tier 2 Tier 3


SAP HANA Disaster RecoveryNews with SAP HANA SPS09 and Beyond

System Replication extensionsSupport for SAP HANA Multitenant DatabaseContainers setupso Replication of whole systemImproved take-over performanceo Prevent reload of ROWstore during take-overo ROWstore kept aliveOptimized Delta shipmento In Multi-Tier environments: Tier 3 rebuildo Part-time extraction and operative usage of Multi-tier

memberso Log & Data Compression (LZ4) on transfer between

sitesOnline Add Host & Remove HostHA/DR Provider FrameworkMonitoring & Alertingo Historization for System Tables of System Replicationo Explicit alerts for SAP HANA System Replication

• Closed Replication Connection• Parameter check

o Fail Detection Scripts

SAP HANA SPS10 (current planning)System Replication extensiono Pure Log-based transfer

• Reduced take-over times• Reduced transfer traffic• Build the foundation for active/active operations• Pilot program after SPS10 planned

Planned beyondSystem Replication extensiono Active/Active Operation (r/o reporting on Sec.)o Backup on shadow instanceo More asymmetric options (n m)o More 1:n relationships for shadow instanceso Time travel via internal snapshots on shadow

instance to handle logical errorso Time delay option between sitesLog Shippingo Based on backup files (initial data, sub sequential log,

steady roll forward)



SAP HANA System ReplicationNew in SPS09: SAP HANA Multitenant Database Containers

SAP HANA Multitenant DatabaseContainers

SAP HANA System Replication can be used toreplicate the whole systemThe replication process treats the completecollection of tenant containers as oneHA&DR is the intention of this first supportReplication of a single tenants to an individuallocation not possible

Further information with SAP Note 2092793

PrimaryPrimary

MDC

SystemDB

TenantDB1

TenantDB2

TenantDB n

SecondarySecondary

MDC

SystemDB

TenantDB1

TenantDB2

TenantDB n

Delta-Data

Log





• Backup/Recovery






• Big Data

More Information


The SAP HANA appliance software from an deployment point of view:SAP HANA <edition>:

– SAP HANA database– SAP HANA client– SAP HANA studio GUI | P2 repository– SAP Host AgentAdditional components installed– Machine readable product description

(LM structure files)– SAP HANA Platform LM tools– SAP Solution Manager Diagnostics Agent– SAPCAR– Operating system configuration

SAP Host Agent

SAP HANA studio

SAP HANA appliance software

SAP HANA installation and configurationstack of HW and SW components

SAPCAR

SAP HANA clients

Machine readableproduct description (LMstructure files)

SAP HANAdatabase

Linux

Server Management Tools (HW vendorspecific)

Storage Subsystem (HW vendor specific)

SAP HANAPlatform LM tools

Linux

Server Mgmt. Tools

Storage Subsystem

SAP Host Agent

SAP HANA clients

SAP HANAdatabase

SAP SMD Agent SAP SMD Agent


Supportsany Device Any Apps

Any App ServerAny Apps

Any App ServerSAP Business Suiteand BW ABAP App ServerSAP Business Suiteand BW ABAP App Server

JSONR Open ConnectivityMDXSQL

Other AppsLocationsReal-timeHADOOPMachineUnstructuredTransaction

SAP HANA PlatformSQL, SQLScript, JavaScriptSQL, SQLScript, JavaScript

Integration Services/Security/ Governance/LCM/Landscape ManagementIntegration Services/Security/ Governance/LCM/Landscape Management

SpatialSpatial



Search/GraphSearch/Graph Text MiningText Mining



DatabaseServicesDatabaseServices



Planning EnginePlanning Engine Rules EngineRules Engine



SAP HANA Platform







Hana One

HEC/HCP




SAP HANA System Landscape: Connectivity overview

SAP HANA

SAP BusinessObjects BI Suite

For analyticalscenarios using BI

suite products

SAP HANA UI forInformation

AccessFor search and textcapabilities inside

SAP HANA

SAP HANAInformationComposer

For easy data uploadinto SAP HANA

SAP BusinessWarehouse

Powered by SAP HANAas primary persistence ofSAP NW 7.3 AS ABAP

SAP BusinessSuite

SAP HANA as primarypersistence also for SAPBusiness Suite systems

SAP HANAApps

Applications runningnatively on / against

SAP HANA Database

R runtime

For direct integrationwith R-runtime

libraries

SAP SolutionManager

For Monitoring andAdministration (DBACockpit, LVM, CTS+)

HANA StudioFor administration,

modeling andlifecycle management

of the SAP HANAsystem

SAP ServiceMarketplace

(SMP)For downloading SP-stacks and patches

SAP Support(OSS)

For remote supportof the SAP HANA

system

SAPHANA

Database

SAPHANA

Database

Host AgentSMD AgentHost AgentSMD Agent

……

Side-by-Side

Primary Persistence Platform / Runtime

Direct ExtractorConnection

(DXC)Data acquisition from

SAP Business ContentDataSource Extractors

e.g. SAP ERP

SAP ReplicationServer (SRS)

For real-timereplication

e.g. Non-SAP

SAP LandscapeTransformation

(SLT)For real-timereplication

e.g. SAP ERP

SAP DataServices

For ETL basedloading

e.g. Non-SAP

XSXS


SAP HANA - Smart Data Access – 1-Data virtualization for on-premise and hybrid cloud environments

BenefitsEnables access to remote dataaccess just like “local” tableSmart query processing includingquery decomposition withpredicate push-down, functionalcompensationSupports data location agnosticdevelopmentNo special syntax to accessheterogeneous data sources

Heterogeneous data sourcesOracle, MS SQL, Teradata, DB2,NetezzaHadoop –Hive (Hortonworks, Cloudera,MapR, etc.) Spark etcSAP HANA (BWoH, SoH)SAP Sybase ASE and IQSAP Sybase ESP, SQLA


SAP HANA - Smart Data Access – 2-Data virtualization for on-premise and hybrid cloud environments

SAP HANA Federation

• Virtual tables in SAP HANA map tophysical tables in other DBs

• Virtual tables modeled in views inSAP HANA

• Access data in other DBs “on the fly”

• External data can be combined withSAP HANA-resident data

Core Value Proposition SAP HANA

• Options to manage performance,data duplication, data integration,and TCO tradeoffs


SAP HANA Predictive with the R Open Source Language

SAP HANA Predictive Analytics with R

• Utilize R open source Language with SAPHANA

• R installation resides on separate server

Core Value Proposition SAP HANA

• Flexibility to create applications using R,combine data with SAP HANA

• Derive high value from advanced analyticsscenarios


SAP HANA Extended Application Services (XS)

What: Small footprint application server / webserver / basis for an application developmentplatform inside SAP HANA

Rationale: Enable application development anddeployment while minimizing architectural“layers”

Create apps that have an http-based UI (browser, mobileapps)Apps run on platform that provides eXtended applicationservices in a simple manner – evolution of architecture-> simplified system architecture = low TCOTight integration w/ SAP HANA DBScope: all kinds of appsLightweight small web-based applicationsComplex enterprise business applications


SAP HANA Extended Application Services (XS) – Today

Front-end Technologieshttp/sHTML5 / SAPUI5Client-side JavaScript

Control Flow TechnologiesODataServer-Side JavaScriptXMLA

Data Processing TechnologiesSQL / SqlScriptCalculation Engine FunctionsApplication Function Library (AFL)

Presentation logic

Control flow logic

Data

Client: Browser or Mobile

SAP HANAXS

Calculation logic


Content lifecycle management in SAP HANAManaging “content” in SAP HANA

SAP HANA content defined:Not part of the core SAP HANA DB installation itselfIs delivered by SAP as part of SAP HANA optimized solutionsIs created in SAP HANA-based development projects (partner, customer)Sometimes called “objects” or “artifacts”

Content comprises all kinds of objects, for example:Schemas and table definitionsAttribute views, analytic views and calculation viewsProcedures and privilegesSQLScript, JavaScript and HTMLRoles and permissions


Content lifecycle management in SAP HANAThe repository

The repository and lifecycle management ofobjects

Native feature of SAP HANA providing “backend”functionalityfor content lifecycle managementUsed to manage various types of design time objects(Content)During deployment/activation the design time objectsbecomeruntime objects (Catalog)

Key functions provided by the repository:Object versioningNamespace conceptSupport for server-based developmentSupport for transport (classical SAP sense and beyond)

SAP HANA StudioModeling Perspective

run time

designtime

SAP HANAcontent


Roadmap: XS Evolution = XS + Cloud + Decoupled + Open

HANA XS2

JavaScript(XS JS)

HANA Database

Java(Tomcat)

C++

SQL CNP

HTTP

JavaScript(Node.js)

R R R

Browser

R HTTP

R

Application Router

XS E is the evolution of HANA XStowards a Cloud architecture whilekeeping best in class SAP HANAsupport

XS E is based on a micro servicesarchitecture, it is decoupled fromHANA DB and enablesindependent scalability

XS E provides multiple runtimes(for JavaScript, Java and C++) andallows Cloud and On Premisedeployments

XS2 Runtime Platform (Cloud Foundry / On Premise)

Note: May be subject to change!


Roadmap: XS E Development Model – A look ahead

ContinuousIntegration

LocalDevelopmentEnvironment

User

App Runtime

App Service

App Runtime

App Service

Key User

Git

App Runtime

App Service

Developer

Git

R

GitR

Clone/Push

Clone/Push

Tool Runtime

Dev Env

Git

Deploy

R

Deploy

R

App Runtime

App Service

Developer

Deploy

R

HANA Database

ExtensionScenario

Web DevelopmentScenario

ProductionScenario

Local DevelopmentScenario

Clone

R






• Backup/Recovery






• Big Data

More Information


An Application’s Lifecycle in SAP HANA

Your ContentProducts or delivery units

Based on changes or complete entitiesUsing CTS+ or SAP HANA native

Your ApplicationConfiguration contentEnabled for mass operationand cloud automation

Your ApplicationBundle object changes via automatic recording

Lock objects individually or for teamsRelease changes when ready for transport

Your Product StructureDefine product structure incl. delivery unit andpackage assignmentView and analyze dependencies for DUs

Your ProductValidate and assemble your product

automatically to ensure consistency and shipefficiently

Create patches and support packages for yourapplication

Your ProductDownload from SMPInstall / update


SAP HANA Application Lifecycle Management

Easy to use

Can be configured basedon your preferences

Can be launchedimmediately after SAPHANA installation:http://<server>:80<instance>/sap/hana/xs/lm

Requires roleassignment


SAP HANA contentexclusivelyused by ABAPfor SAP HANA

Native SAP HANAcontent or as partof a solution(BI, Mobile, …)

Native SAP HANAcontent

SAP HANASource

Transport scenarios for SAP HANA content

SAP HANA Application LifecycleManagement

SAP HANA stand-alone transport managementNo need for ABAP-footprintLightweight and easy-to-use transport tool

SAP HANATargetUse case Transport

Management

Enhanced CTS (CTS+)Transported as any other non-ABAPcontentUses existing CTS transport landscapeSAP process tools (ChaRM, QGM)

HANA Transport ContainerTransported with standard ABAP transportsIntegrated in existing CTS transportlandscapeSAP process tools (ChaRM, QGM)


Native SAP HANA ContentTransport Landscape

Test ProductionDevelopment

HND

Content

HNQ

Content

TransportRoute

HNP

HALM

Content

TransportRoute

HALM1. Requestcontent

Export

ImportActivate

ImportActivateExport2. Content

Provided2. ContentProvided

ApplicationLandscape

1. Requestcontent


SAP HANAStudio

SAP HANAStudio

TestDevelopment

Transport via Change and Transport System (CTS+)Transport Landscape

HNQ HNP


CTSSystem

Transport Transport

ImportImport

TransportRequest

TransportRequest

TransportRequest

HND HNQ HNP

Production

HNDRepository

Objects

HALM

Attach

Repository

Objects

Repository

Objects


SAP HANA Transport ContainerTransport Process


Production

Transport

PRD

Non-ABAP

Import&

Acti-vate

DeliveryUnit

TR

HTC

ABAP

DEV

TR

Non-ABAP

Add

HTC

ABAP

DeliveryUnit

Test

QAS

Non-ABAP

DeliveryUnit

TR

HTC

ABAP

Import&

Acti-vate

Transport

SAP HANAStudio

SAP HANAStudio

Development





• Backup/Recovery






• Big Data

More Information


e

SAP HANA platformProcessing Engine

Application Function Lib. & Data Models

Integration Services

SAP HANA PLATFORMReal-time transactions + end-to-end analytics

OperationalAnalytics

Big DataWarehousing

Predictive, Spatial &Text Analytics

REAL-TIME ANALYTICS

Sense &Respond

Planning &Optimization

ConsumerEngagement

REAL-TIME APPLICATIONS

SAP ESP

SAP ASE

ReplicationServer

SAP SQLAnywhere

SAP IQ

SAP DataServices

Extended Application Services

SAP Data Management PortfolioEnd-to End Data Management & App Platform for Real-Time Business

DatabaseServices

SAP HANAdynamic tiering


Introducing SAP HANA dynamic tiering

• Manage data cost effectively, yet with desired performance based on SLAs• Handle very large data sets – terabytes to petabytes• Update and query all data seamlessly via HANA tables• Application defines which data is “hot”, and which data is “warm”• Native Big Data solution to handle a large percentage of enterprise data needs without

Hadoop

Table

TableHANADatabaseEngine

HANA DynamicTiering Engine

Extendedtable

(warm data)

Fast data movement and optimizedpush down query processing

All data resides in extended store

HANA Database Service


Data Qualities and Data Temperatures

Data in the databaseDifferent data temperatures

Maximum access performanceHot data - always in memoryReduced access performance:Warm data - not (always) in memory

All part of the database’s dataimage

Data moved out of the databaseDifferent data qualities

Available for read accessNear-line storageNot accessible without IT processTraditional archive

Data is stored and managedoutside of the applicationdatabase

SAP HANA Database

Hot

Warm

Data for daily reporting,other high-priority data

Other data required tooperate the application

NLSData that is (normally) not

updated, infrequently accessed

Traditional ArchiveData that‘s kept for legal reasons

or similar

Externalize


SAP HANA Database

Hot data

SAP HANA dynamic tieringMap data priorities to data management

Warm data

Primaryimage inmemory

Durability

Cache /Processing

PrimaryImage ondisk

Dynamic Tiering

All in onedatabase

Hot StoreClassic HANA tables

Primary data image in memoryDB algorithms optimized for in-memorydataPersistence on disk to guarantee durability

Warm StoreExtended Tables

Primary data image on diskData processing using algorithmsoptimized for disk-based dataMain memory used for caching andprocessing.

Hot Store Warm Store

RAM


SAP HANA dynamic tiering: The overall system layout

SAP HANA with dynamic tiering consists of two types of hosts:

• Regular worker hosts (running the classical HANA processes:indexserver, nameserver, daemon, xsserver,…)

• HANA hosts can be single-node or scale-out; applianceor TDI

• “ES hosts” (running nameserver, daemon, and esserver)

• esserver is the database process of the warm store

Hot Store

Fast data movement and optimized push downquery processing

SAP HANA System with dynamic tiering service

Workerhost(*)

Workerhost

Workerhost

ClientApplication

Connect

ES host(controller)

Further EShosts

ColumnTable

RowTable

ExtendedTable

Warm Store

Common Storage System(*) Standby hosts not shown

• One single SAP HANA database:one SID, one instance number

• All client communication happensthrough index server / XS server


Database Catalog

SAP HANA dynamic tiering: HANA Extended Tables

HANA Database

WarmStoreData

HANA extendedtable schema is partof HANA database

catalog

HANA extendedtable data resides in

warm store

HANA extendedtable is a first class

database objectwith full ACIDcompliance

HotStore

Table Definition

Data

Table Definition

Classical HANAcolumn/row table

Extended table(warm table)


Each HANA tenant DB is associated with exactly one extended store:

SAP HANA Dynamic Tiering: SAP HANA MDC Support

HANA Cluster

Computenode

HANA Database

Extended Store

HANA Database

Extended Store

HANA Database

Extended Store

Computenode

Computenode

Computenode


SAP HANA database

Database Catalog

Extended Tables in HANA BWUse Case: Staging and Corporate Memory

Object Classification in BWData Sources and write-optimized DSOs can have theproperty “Extended Table”

Generated Tables are of type“Extended”All BW standard operationssupported – no changesOnly minor temporary RAMrequired in HANA

InfoCubes and Regular orAdvanced DSOs

Generate standard column tableHot StoreWarm store

BW System

Write-optimizedDSO

Corporate MemoryData

Source

Staging Area

TableSchema

Data

PSA TableTableSchema

Data

Active Table

InfoCube

Data Mart

TableSchema

Data

Fact Table


SAP HANA dynamic tiering for Big Data

SAP HANA with Dynamic Tiering provides native Big Data solution

• Cutting edge, in-memoryplatform

• Transact/analyze in real-time

• Native predictive, text, andspatial algorithms

Hot data

SAP HANA

Petascale, warmstructured data

HANA extendedtables

• Petascale extension to HANA withdisk backed, columnar databasetechnology

• Expand HANA capacity withwarm/cool structured data in HANAwarm store

• Tight integration between HANA hotstore and HANA warm store foroptimal performance

Hot data

SAP HANA

Petascale, warmstructured data

HANA extendedtables


SAP HANA dynamic tiering roadmap

FUTURE

• HANA ES host scale-out and auto-failover (HA)

• Disaster Recovery (SAP HANA system replication)

• Hybrid extended tables with rule based automaticdata movement / aging

• Communication protocol optimization between hotand warm store

• Further unification of DDL and DML for HANAextended tables

• Further optimizations for HANA Calculation Engine

• Further extension of unique HANA capabilities towarm store

PLANNED

• SAP HANA dynamic tiering available to beused by any HANA application

• Common installer

• Unified administration and monitoring usingHANA Cockpit

• Extended Storage (ES) engine is part ofHANA topology

• Single authentication model

• Single licensing model

• Combined error log / trace handling

• Fully integrated backup/recovery






• Backup/Recovery






• Big Data

More Information


Modernize IT With Flexible Platform for Big DataSAP HANA platform + Hadoop / NoSQL + Analytics +Applications

Hadoop /NoSQL Data

Lake

GRAPHENGINE

PREDICTIVE

ENGINE

TEXTENGINE

SPATIALPROCES

SING

ANALYTICS

ENGINE

Logs TextOLTP Social MachineGeoERPSensor

CONSUME

COMPUTE

MANAGE

SOURCE

ACQUIRE

Reporting &Dashboards

HighPerformanceApplications

Application

Development

Environment

Transformations &Cleansing

StreamProcessing Virtual Tables

Smart DataIntegration

Smart Data Quality

Smart Data AccessSmart Data StreamingMapRedu

ce

Data Exploration& Visualization

Ad-hoc & OLAPAnalytics

PredictiveAnalysis

BusinessPlanning &Forecasting

HIVE

YARN

HDFS

STREAMPROCES

SING

User DefinedFunctions

Store &forward

Mobile applications and BIMobile applications and BI

DataExcha

nge

MPParchitecture

DynamicTiering

Aged data inDisk

101010010101101001110

In-Memory ColumnStorage

Data model &data

Parallelprocessing

Calculationengine

Fast computing

Series DataStorage

Highperformance

analytics

Store time-seriesdata


SAP PartnershipsOpen Hadoop / NoSQL Strategy


SAP and Hadoop / NoSQL IntegrationOpen Strategy

MapReduce / YARN / AWS Elastic MapReduceDistributed Processing Framework

HiveSQL QuerySpark

(in-memory)

HDFSHadoop Distributed File System

Hadoop / NoSQL

Adapters

SAP DataServices

SAP HANA Platform

SAPLumira

SAP BIPlatform

(universe)

SAPPredictiveAnalytics

SAP Analytics

PigScripting

ODBCDriver

ODBCDriver

Datastaxconnector

ImpalaMPP SQL

Query

MongoDB

CassandraNoSQL DB

GreenplumDB

SmartData

Integration

VirtualUser

DefinedOperators

RFCHadoop

webHCatWedHDFS

SmartEvent

Processing

SmartData

Access

ODBCDriver AdapterAdapter

SAP EIM


SAP HANA & Hadoop integration

HANA & Hadoop Integration(SPS09)• SQL on Hadoop via SDA (Virtual tables)

– Hive (SPS07) or Spark

• Execution of MR-jobs via HANA (VirtualFunctions)

• Access to HDFS (via vUDF)

• Integration for storage & processing

Next Steps (SPS10)• Spark SQL via SDA

• Optimization (e.g. specific Spark RDD)

• Integration: e.g. Admin


SAP HANA Smart Data Integration & Smart Data QualityReplication, Batch Integration, and Data Virtualization

CapabilitiesReal-time replication & CDC on select sourcesBulk integration (metadata / data)Data virtualization via Smart Data AccessReal-time data cleansing and transformationData enrichment with geospatial informationSAP HANA Studio to define data transformation flowsSupport for on-premise and cloud sourcesOpen SDK and built-in adapters including HIVE

BenefitsSimplified landscape: 1 environment to provision dataReal-time: lower latency with in-memory performanceOpen & extensible: supports data of any shape or size

Built-In Adapters Custom Adapters

Transformations

SAP HANA

MetadataMetadata AdapterFramework

AdapterFramework

ODataDB2, OracleSQL Server

Smart Data IntegrationSmart Data Quality


SAP HANAVirtual User Defined Function

CapabilitiesUser defined function for data virtualizationDirect access to HDFS via RFC Hadoop function

(webHCat WedHDFS) without need for package,mapper, and reducer specification

Invoke custom Map Reduce jobs; store as JARfile that be called by SQLAd-hoc query capabilities and processing of

unstructured data

BenefitsProvides flexibility, supporting use cases beyond

Hive via SAP HANA smart data access

SAP HANAvUDF

Operator

RFCHadoop

Hadoop

Map Reduce

HDFS


SAP HANA Smart Data StreamingReal-time Event Streams

Capabilities

Capture, filter, analyze and act onmillions of events per second in real-time

Capture high value data in SAP HANAand direct other data into Hadoop(adapter for HDFS or MapReduce jobinto Hive)

Stream live information to operationaldashboards

Perform continuous queries usingdeclarative (CCL) or model-drivenapproaches

Benefits

Real-time insight from streaming

Incomingstreams

Stream(push)

SAP HANA

Streaming

Service


Real-time Applications, Interactive Analysis

Tachyon

SCMERP CRM Text Geospatial Sensor SocialMedia Logs

DataSourc

e

Distributed File

Persistence

In-MemoryPersistence

In-MemoryProcessing

SAPHANAsmartdata

access

Data Access

SQL Java

Scala Python

Other

SQL.NET

Javascript

MDX Other

NodeJS

In-memory

Columnar Data

Predictive Text /NLP

Geospatial

Planning/ Rules

SAPHANA

Spark

SQL/Shar

k

SparkStream

ing

MLlib GraphX(graph)

HDFS / Any Hadoop

FaultTolerant

DFS Mgmt

SAP HANA and Apache SparkEnterprise Fabric for Big Data

Integration between SAP HANA and Spark is via SAP HANASmart Data AccessDone with Spark SQLRequires Shark ODBC driver and unixODBC Driver Manager


SAP BusinessObjects BI / SAP Lumira & Hadoop / NoSQLCombined With SAP HANA

Data Integration

Log Files

Text DataSources

StructuredData

Sources

SAP HANA Platform

SAPBusinessObjects

BI

SAP Sources Non-SAP

BI Universe

Available as of SAP Data Services 4.1 (Hive & HDFS)

SAP HANA smart data access (Hive)Available as of SAP HANA SPS6

Hive, Amazon EMR, Impala available as of BI4.0 FP3*

Hadoop

* BI 4.0 FP3 for single-source universe

BI 4.0 FP5 for multi-source universe

SAPLumira

Desktop

Hive 0.1, Amazon EMR 0,8

EMR, Hive 0.13, Impala,support planned for 1.21

SAPLumiraCloud

Hive , EMR

Hana CloudIntegration


SAP BusinessObjects BI / SAP Lumira & Hadoop / NoSQLAgnostic View

Log Files

Text DataSources

StructuredData

Sources

SAPBusinessObjects

BI

SAP Sources Non-SAP

BI Universe

Hive, Amazon EMR, Impala available as of BI 4.0 FP3*

Hadoop

* BI 4.0 FP3 for single-source universe

BI 4.0 FP5 for multi-source universe

SAPLumira

Desktop

Hive 0.1, EMR 0,8

EMR, Hive 0.13, Impala,support planned for 1.21

SAPLumiraCloud

Hive , EMR

Hana CloudIntegration


SAP Predictive Analytics & Hadoop / NoSQL

SAP Predictive Analytics

Hadoop / NoSQL

SPARK

HDFS

HIVEGreenplum

DB

CapabilitiesUnified UI for business analysts and

data scientistsPackaged business applicationsExtensive predictive library plus R,

Hadoop, and No SQL integration(Hive, HDFS, SPARK, andGreenplum)Cloud ready

BenefitsImproved forecasts from analysis of

Big DataSupport for business users & data

scientists





• Backup/Recovery






• Big Data

More Information


0387


Further Information

Whitepaper: SAP HANA System Landscape Guide

http://www.saphana.com/docs/DOC-4385

SAP Public Web

SAP HANA

SAP HANA Online Help


STAY INFORMED

Follow the ASUGNews team:

Tom Wailgum: @twailgum

Chris Kanaracus: @chriskanaracus

Craig Powers: @Powers_ASUG


No part of this publication may be reproduced or transmitted in any form or for any purposewithout the express permission of SAP SE. The information contained herein may bechanged without prior notice.

Some software products marketed by SAP SE and its distributors contain proprietarysoftware components of other software vendors.

Microsoft, Windows, Excel, Outlook, and PowerPoint are registered trademarks of MicrosoftCorporation.IBM, DB2, DB2 Universal Database, System i, System i5, System p, System p5, System x,System z, System z10, System z9, z10, z9, iSeries, pSeries, xSeries, zSeries, eServer,z/VM, z/OS, i5/OS, S/390, OS/390, OS/400, AS/400, S/390 Parallel Enterprise Server,PowerVM, Power Architecture, POWER6+, POWER6, POWER5+, POWER5, POWER,OpenPower, PowerPC, BatchPipes, BladeCenter, System Storage, GPFS, HACMP,RETAIN, DB2 Connect, RACF, Redbooks, OS/2, Parallel Sysplex, MVS/ESA, AIX,Intelligent Miner, WebSphere, Netfinity, Tivoli and Informix are trademarks or registeredtrademarks of IBM Corporation.

Linux is the registered trademark of Linus Torvalds in the U.S. and other countries.

Adobe, the Adobe logo, Acrobat, PostScript, and Reader are either trademarks orregistered trademarks of Adobe Systems Incorporated in the United States and/or othercountries.Oracle is a registered trademark of Oracle Corporation.

UNIX, X/Open, OSF/1, and Motif are registered trademarks of the Open Group.

Citrix, ICA, Program Neighborhood, MetaFrame, WinFrame, VideoFrame, and MultiWin aretrademarks or registered trademarks of Citrix Systems, Inc.HTML, XML, XHTML and W3C are trademarks or registered trademarks of W3C®, WorldWide Web Consortium, Massachusetts Institute of Technology.

Java is a registered trademark of Sun Microsystems, Inc.

JavaScript is a registered trademark of Sun Microsystems, Inc., used under license fortechnology invented and implemented by Netscape.SAP, R/3, SAP NetWeaver, Duet, PartnerEdge, ByDesign, SAP BusinessObjects Explorer,StreamWork, and other SAP products and services mentioned herein as well as theirrespective logos are trademarks or registered trademarks of SAP SE in Germany and othercountries.

© 2015 SAP SE. All rights reserved

Business Objects and the Business Objects logo, BusinessObjects, Crystal Reports, CrystalDecisions, Web Intelligence, Xcelsius, and other Business Objects products and servicesmentioned herein as well as their respective logos are trademarks or registered trademarksof Business Objects Software Ltd. Business Objects is an SAP company.

Sybase and Adaptive Server, iAnywhere, Sybase 365, SQL Anywhere, and other Sybaseproducts and services mentioned herein as well as their respective logos are trademarks orregistered trademarks of Sybase, Inc. Sybase is an SAP company.

All other product and service names mentioned are the trademarks of their respectivecompanies. Data contained in this document serves informational purposes only. Nationalproduct specifications may vary.The information in this document is proprietary to SAP. No part of this document may bereproduced, copied, or transmitted in any form or for any purpose without the express priorwritten permission of SAP SE.

This document is a preliminary version and not subject to your license agreement or anyother agreement with SAP. This document contains only intended strategies, developments,and functionalities of the SAP® product and is not intended to be binding upon SAP to anyparticular course of business, product strategy, and/or development. Please note that thisdocument is subject to change and may be changed by SAP at any time without notice.

SAP assumes no responsibility for errors or omissions in this document. SAP does notwarrant the accuracy or completeness of the information, text, graphics, links, or other itemscontained within this material. This document is provided without a warranty of any kind,either express or implied, including but not limited to the implied warranties ofmerchantability, fitness for a particular purpose, or non-infringement.SAP shall have no liability for damages of any kind including without limitation direct,special, indirect, or consequential damages that may result from the use of these materials.This limitation shall not apply in cases of intent or gross negligence.

The statutory liability for personal injury and defective products is not affected. SAP has nocontrol over the information that you may access through the use of hot links contained inthese materials and does not endorse your use of third-party Web pages nor provide anywarranty whatsoever relating to third-party Web pages.

Documents

387 Planning a System Landscape of the SAP HANA Platform (1)