IBM WebSphere eXtreme Scale & XC10: An introduction to Elastic Caching

Seema KumarWebSphere Product Management

2

Agenda• What is Elastic Cache?• WebSphere eXtreme Scale v7.1• WebSphere DataPower XC10 Appliance• Common Usage Patterns

3

4

Very powerful cache

�Scales from simple in-process topologies to powerful distributed topologies.

Platform for building powerful XTP/Data Grid applications

Form of in-memory data grid

�Manage application state

�Scales to 1000’s of servers

�Sometimes referred to as Distributed Application State Management.

A flexible framework for realizing high performance, scalable and data-intensive

applicationsNew York San Francisco

London Shanghai

3 Use Cases

What is Elastic Cache?

5

Modern Application Infrastructure Topology

Web Server Tier Back-end SystemsDatabase Tier

App Server Tier Elastic Data Grid

IBM HTTP ServerWebSphere

Application ServerDB2 UDB

DataPower XC10 for simple data oriented scenarios:• HTTP Session Replication• Elastic Dynacache• Web Side Cache

eXtreme Scale for maximum flexibility covering data and application oriented scenarios

1

2

6

When do I need an Elastic Grid?

7Redundant copies of data at different versions

Traditional Cache Operation

■ Cache capacity determined by individual JVM Size.■ Size of each cache = M■ # JVMs = N■ Total cache = M

■ Invalidation load per server increases as cluster grows.

■ Cold start servers hit the database. AApplication

AApplication

AApplication

AApplication Invalidation Chatter

Database

ApplicationInvalidation load

increases with cluster size

New Serverwith cold cache

High load onDatabase

8

Application E

D’

WebSphere eXtreme Scale Cache Operation

■ Cache capacity determined by total cluster size■ Size of each cache = M■ # JVMs = N■ Total Cache = M x N

■ No invalidation chatter

■ Linearly scalable

■ Less load on database and no cold start spikes

Application D C’

Application C B’

Application B A’

Application A

Cache4x Larger

Cache5x Larger

E’

9

Innovative Elastic Caching Solutions“Data Oriented”

“Application Oriented”

Session management

Elastic DynaCache

Web side cache

Event Processing

In-memory OLTP

Worldwide cache

Petabyte analytics

Data buffer

In-memory SOA

eXtreme Scale

DataPower XC10 Appliance

• Ultimate flexibility across a broad range of caching scenarios

• In-memory capabilities for application oriented scenarios

• Drop-in cache solution optimized and hardened for data oriented scenarios

• High density, low footprint improves datacenter efficiency

Elastic caching for linear scalabilityHigh availability data replication

Simplified management, monitoring and administration

10

10

IBM WebSphere eXtreme Scale• Proven mature product:

– Fourth major release of product with V7.1 in 2010– Used at some of the largest web sites in the world

• Lightweight runtime footprint (15MB jar)

• Integrates with all versions of WebSphere Application Server and almost any Java-based application container or Java Virtual Machine

• Only requires JavaTM SE run-time environment

– Exploits WAS-ND environment when available

• Meets the needs of a variety of application environments:

– Java SE, Java EE (WAS, WebLogic, Tomcat…), as well as for .Net applications via REST APIs*

• Proven multi-data center capabilities

• Proven low-latency access to data

11

11

IBM WebSphere DataPower XC10 Appliance V2

• Scale out with ease– Large, 240GB elastic cache allows you to

scale more economically while providing high Quality of Service

– Scales elastically without application downtime– Linear, predictable scaling at predictable cost

• Easy drop in use for common scenarios

– Support for data-oriented caching scenarios without rip & replace

– Unbinds cache from application server memory constraints

• Fault tolerance– Lower risk of data loss while providing

continuous availability

• Flexible and simple user management

– Simple solution for real world management and monitoring

12

Easy drop-in use for common scenarios

• Simple Cache

• HTTP Session Cache

• WAS Dynamic Cache Service (“Dynacache”) support

Little or no code changes required!

13

Fault Tolerance

• Create an appliance “collective”for high availability

• Lowers risk of data loss through automatic replication of data

• Continuous Availability!

• Self Healing - Failures are automatically detected

• Easy to configure

14

14

Common Elastic Caching Scenarios

HTTP Session ReplicationSide CacheData Access LayerData Grid

15

15

HTTP Session Replication

• Many enterprise applications today require HTTP session persistence.

• A grid of JVMs can be established with the sole purpose of storing HTTP session (or any java) objects.

• Isolating the application runtime from grid runtime, thereby, freeing up the JVM heap for application use.

• Provide linear scalability to accommodate growth (in # of sessions or size of session objects).

• Providing replication and management of session objects within the grid.

• Can even store session objects across datacenters.

Session GridWeb App Layer

16

16 DataPower XC10 Overview

Common set of Patterns - HTTP Session data cache

• No new code required • Extension of legacy session management caching mechanism in WebSphere Application

Server• Extensions to WebSphere Application Server administrative console to support WebSphere

DataPower XC10 session management caching and WebSphere eXtreme Scale session management caching

• WebSphere Application Server connects seamlessly to the WebSphere DataPower XC10 appliance or WebSphere eXtreme Scale

– Client code must be installed on WebSphere Application Server systems

• Easily configure WebSphere applications to store HTTP session data to a data cache on the WebSphere DataPower XC10 appliance through the WebSphere Application Sever administrative console

• Replaces other session replication mechanisms (memory-to-memory replication)

• Removes the need for Database traditionally used for persistence

• Enables HTTP session failover between WebSphere Application Server cells

17

Active/Active Datacenter HTTP Session Failover

Web Server

PlantsByWebSphereCluster

memory to memoryreplication

Datacenter1

xPlantsByWebSphereCluster

Deployment Manager

PlantsByWebSphereCluster

memory to memoryreplication

Datacenter2

xPlantsByWebSphereCluster

Deployment Manager

WebSphere eXtreme Scale

OR

WebSphere eXtreme Scale

18

18

Side Cache

• Client first checks the grid before using the data access layer to connect to a back end data store.

• If an object is not returned from the grid (a cache “miss”), the client uses the data access layer as usual to retrieve the data.

• The result is put into the grid to enable faster access the next time.

• The back end remains the system of record, and usually only a small amount of the data is cached in the grid.

• An object is stored only once in the cache, even if multiple clients use it. Thus, more memory is available for caching, more data can be cached, which increases the cache hit rate.

• Improve performance and offload unnecessary workload on backend systems.

19

19 DataPower XC10 Overview

Elastic Caching used as a Side Cache

• Used to store data for fast, lower-cost access than a database• Uses ObjectMap APIs from WebSphere eXtreme Scale• Client can run in a standard Java EE compliant server environment or in

any Java Virtual Machine compliant with Java SE V1.4 or beyond• Dynamic Cache Service Provider for WebSphere Application Server (ND

6.1 / ND 7.0)– Page fragment cache– Web services cache– Command cache

• Support L2 Cache – Open JPA/Hibernate (WXS only)• Side Cache for Datapower XI50 thru REST Gateway (XC10 only)• Side Cache for WPS/WESB• Side Cache for Message Broker

20

Objects in the Elastic Data Grid – Key, Value pairsKey Value pair examples:

• integer key, string value• (hex) integer key, compound value• compound key, compound value

21

21

Data Access Layer – (WXS only)

• The grid can be used as a special data access layer where it is configured to use a loader to get data from the back-end system.

• The loader can be configured to asynchronously write back the changes to the back-end system

• Even when the back end is down, the application can write changes. They are buffered in the grid to be written back when the back-end system comes online again.

• Reduced dependency on back end data store

22

22

Client Usage: Online BankingRetail Banking

35xreducedresponsetimes

22 Milliononline banking users

Next-generation Online Banking– Before: 700ms to login with 2 backend calls

– After: 20ms to login with profile cache access

– $6M/yr cost savings in MIPs reduction

– 3 datacenters: If one datacenter goes down it fails over seamlessly to the others

– 8Gb of data transfer per hour between DC’s

$500kreducedcosts per month

20xreduction in “FCIs”

Provide seamless cache infrastructure across applications

Deliver high performance & consistent response times

Ensure high availability of critical online applications

23

23

Entertainment 10xreducedresponsetimes

7 Billionrequests per day

23

Support transaction-intensive services

Deliver consistent & predictable response times

Take action on growing volumes of business events

Scale with simplicity and lower cost

Fantasy Sports Web Infrastructure– Before: 60ms response time against

database– After: WXS improved to 6ms response

time – 450k concurrent users– 80k requests per second up to 1M in

2011– 6 weeks from concept to production

Client Usage: Worldwide Fantasy Sports Web Site

24

24

Resources

• Fully functional J2SE trial downloadhttp://www.ibm.com/developerworks/downloads/ws/wsdg/learn.html

• Wiki documentationhttp://www.ibm.com/developerworks/wikis/display/objectgrid/Getting+started

• User's Guide to WebSphere eXtreme Scale http://www.redbooks.ibm.com/abstracts/sg247683.html

• Data Sheetftp://public.dhe.ibm.com/common/ssi/pm/sp/n/wsd14088usen/WSD14088USEN.PDF

• XC10 Web Sitehttp://www-01.ibm.com/software/webservers/appserv/xc10/

• Getting Started Wiki for XC10http://www.ibm.com/developerworks/wikis/display/extremescale/2010/06/25/IBM+WebSphere+DataPower+XC10+appliance+now+available

http://www.ibm.com/developerworks/spaces/xtp

WebSphere Extreme Transaction Processing for Developers Space will discuss various topics for developing and deploying XTP applications and will point out emerging trends, benefits, challenges, and features associated with it.

http://www.youtube.com/user/ibmextremescale#p/a

Weekly video podcasts covering customers questions and forum posts on the IBM WebSphere eXtreme Scale product.

Additional resources

25

Questions?

26

Java is the registered trademark of Oracle Corp

27

27

28

Backup

29

Elastic Caching for WebSphere Portal