Sven Oehme – STG Industry Technology Lead [email protected] · GPFS: Information Lifecycle Management ILM Support basic Technology - Storage pool – group of LUNs - Fileset - define

x

© Copyright IBM Corporation 2008

Sven Oehme – STG Industry Technology Lead

[email protected]

2 © Copyright IBM Corporation 2008

IBM Systems & Technology Group

agenda

Conclusion5

The NAS Layer4

General Parallel File System (GPFS)3

The Scale-out File Services (SoFS) approach2

NAS today1



Source: IDC Disk Systems Forecast, May 2007

External Disk Shipment Forecast

173239 306

462784

13002190

3538

$126,00$72,20

$43,90$31,20

$19,55$13,47

$8,35$5,50

$3,63$2,37

$1,54$0,99

21798 17256 13433 14414 15327 17511 18287 19459 20687 21802 23012

56999199

2420314943

23961

0,1

1

10

100

1000

10000

100000

2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011

PB

$/GB

M$



Customer’s Feedback on Current NAS

Customers have increasing demand for file services. The growth of so-called “unstructured data” outpaces that of databases by far.

- But current NAS solutions do not scale. File systems are limited to a few Terabyte and a few Million Objects

- So customers have to add box by box and manage them individually

- No way to apply Policys across this independent data islands

Parallel access to data gets more and more common, especially in the digital media space

We see requirements on data access rates and response times to individual files that have been unique to HPC environments before

Making file servers clustered is easy in theory but hard to build and maintain, there is too much complexity

Migration, integration or removal of storage for file services is a nightmare so a lot of legacy systems still exist and need to be maintained

Backup windows are a big issue and get worse while the amount of data continuously increase

Integration of Information Lifecycle Management (ILM) functions into the file system becomes more important as the amount of TB's explode



agenda

Conclusion5

The NAS Layer4



NAS today1



SoFS Quick Facts SoFS combines a Clustered Filesystem with a Global Namespace. It‘s deployed centrally,

managed centrally, backed up centrally and grown centrally.

SoFS scales horizontally not vertically. When you need more capacity, you just add more disks. When you need more performance you add nodes and/or disks. There are almost no architectural limits for future growth.

NAS boxes growth vertically. You can add controllers and storage up to the maximum configuration of the box. Beyond that you have to buy a new one.

SoFS provides integrated Information Lifecycle Management

- Different tiers of storage, e.g. FC, SATA and tape

- Policy driven placement and migration of files over their entire Lifetime

SoFS offers synchronous Block Level replication for data and Metadata

SoFS offers multi-site configurations via the cross-cluster mount functionality.

SoFS allows to build Highly Available Systems including Disaster resistant Systems across multiple sites

➔SoFS solves the root cause of the lack of scalability and performance in todays NAS systems instead of patch around it as most other Scalable NAS or Global Namespace Solutions do



SOFS overview: A ‘Scale Out NAS’ design, which provides: ‘Commodity’ Hardware based design

Product is comprised of x86 servers and inexpensive disks. Also supports numerous storage devices via SVC (disk) and TSM (disk, tape, optical).

Scaling is done with an Army of Ants and not with a single Big Elephant

Based on GPFS (General Parallel File System)

A proven platform over 10 years old which has numerous customer installs of multi-thousand node clusters and multi-PB filesystems. Designed for extreme performance, scalability, and availability – features distributed metadata, wide striping of data, byte range locking, parallel I/O, variable block size, extensive tuning parameters. Reference www.top500.org

Integrates Clustered CIFS and Clustered NFS via CTDB, which provides:

Transparent, non-disruptive failover of CIFS and NFS with no client side changes

CIFS performance of over 700MB/sec from one node to one client and NFS performance over 800MB/sec

Integrated ACL’s between Unix, Linux and Windows, supports NFSv4 and NTFS access

Unparalleled aggregate performance scaling, Intelligent load balancing, simultaneous access to a single file by heterogeneous clients, dynamic thin provisioning.

Provides Scalable Tiered Storage (ILM), via a powerful metadata policy engine - migration without path changes.

Rule sets for initial placement and migration of individual files spanning move, copy, delete, backup, restore, replicate, snapshot, etc. Immutability flag planned for future release. Features no stubs on disk to disk migrations.

Recent benchmarks show processing 1 Billion files in less than 15 minutes

Provides Storage Tiering/Pooling for ALL Storage Technologies (SSD, SAS, FC, SATA, iSCSI)



SOFS overview: A ‘Scale Out NAS’ design (cont) Integrated Backup / Restore and DR Capabilities even for multi-PB Systems

Low Level interlock into TSM provides scalable incremental forever backups and intelligent policy based restore, allowing a multi-PB system back online with critical data in minutes or hours, not weeks or months.

Low Level integration into HSM to provide Quick Disaster Recovery of Metadata with Stub File Creation (Restores a Filesystem without any Tape Movement)

Multi Site Replication (synchronous as well as asynchronous) integrated in Filesystem Layer

Full management suite included

End-to-End Monitoring of all included Components (Network, SAN, Storage, Servers, OS, NAS Components)

Tools for user, Group and Project level quotation

Self Service Space Creation Utilities with pre-defined user templates

System utilization and Statistic Reporting

end to end performance reporting Capabilities

Commandline, Scripting and Graphical Management Capabilities

Distributed Datacenter Design

Data exchange Protocol for Intersite Communication (to prevent Throughput limitations at high Latency's)

Global Namespace (integration of existing NAS devices at Namespace Level) with User Locality Feature

Remote Caching Capabilities



SoFS

parallel file system

clustered NAS services

CIFS NFS HTTPS FTP

IP Layer

Classic Filers

DFS AD

local file system

NAS services

CIFS NFS

local disk

What is a Scale-out NAS Solution ?

shared SAN disks

storage pools

From a client perspective it is yet another filer

In the datacenter it dramatically changes the handling of your Data



SoFS vs. NAS virtualizers

SoFS

- True clustered NAS

- All nodes serve all files- Maximum single file performance is equal to

the aggregated performance of the cluster

- No disk, backup, management, etc. islands

- No bottlenecks on single directory branches

Virtualizer (VFM, StorageX, ONTAP GX)

- No clustering, just some pretty paint on an island topology

- Each individual file is pinned to a single NAS filer

- Maximum single file performance is equal to the performance of the hosting filer

- Possible bottlenecks on individual directory branches

- Islands regarding disks, backup, etc.

/

/sales

/finance

/web

/

/sales

/finance

/web



SoFS

IP Layer

Classic Filers

DFS AD

The SoFS Story: Performance & Scalability

/

/sales

/finance

/web

/

/sales

/finance

/web

parallel data flowall nodes serve all files

each file is pinnedto a single filer



Does Performance matter ?

729 MB/sec from a single Windows Client to a single SoFS Node !!



The SoFS Story: Utilization & ILM

Capacity managed individually

Avarage Utilization usually <50%

All files online, but 80% haven't been accessed during the last 6 months

Capacity managed centrally.

Average utilization >80%

Policy driven file movement between tiers, the most inactive files can be migrated to tape

SoFSClassic Filers

With SoFS you just buy the capacity you really need

$$ $$ $$ $$



The SoFS Story: Integrate new Storage

Replace a storage system in 3 steps without any service interruption

SoFS

SoFS makes adding and replacing storage one of the easiest things in theworld.

No more planned downtimes and weekend shifts to install storage.

1. Attach new storage system

2. Tell SoFS to redistribute data excluding thedecomissioned system

3. Remove storage system



The SoFS Story: Administration

“I loved my first filer. It was so easy to administrate. When we installed the 20th I started loathing them.”

Each filer has to be individually administrated

Operation costs grow linear with the amount of filer

The SoFS cluster is managed centrally

Thanks to ILM you manage storage tiers instead of individual disks

Thanks to easy growth and data migration you do not have to plan changes weeks upfront

SoFSClassic Filers

SoFS makes operation easy



The SoFS Story: Backup

TSM is tightly integrated into SoFS

The scan for changes in the file system is not done by the TSM backup client, but by the ILM engine of the file system.

We scan >15 million files per minute

This means the backup window is more or less reduced to the time needed to copy the changes onto tape

With LAN-free backup this can be done with the bandwidth of your SAN environment taking load off your TSM server

SoFS

LAN-free

TSMServer

Metadata

Data



agenda

Conclusion5

The NAS Layer4



NAS today1



Overview IBM’s General Parallel Filesystem (GPFS) available 1996

Used in other IBM Products (BIA for SAP, VTS)

Used on many of the largest supercomputers in the world.

- Cluster: 1000+ nodes, fast reliable communication, common admin domain.

- Shared disk: all data and metadata on disk accessible from any node through disk I/O interface.

- Parallel: data and metadata flows from all of the nodes to all of the disks in parallel.

High performance

- Multi-Terabyte files, Multi-Petabyte file systems.

- Wide striping, large blocks, many GB/s to single file

Highly Reliable

- Can survive Disk and Node failures

- Allows Split site Configurations

Storage

StorageNetwork



GPFS: ASC Purple/C Supercomputer

1536-node, 100 Teraflop IBM System P cluster at Lawrence Livermore National Laboratory

2 PB GPFS file system (one mount point)

500 RAID controller pairs, 11000 disk drives

126 GB/s parallel I/O measured to a single file (134GB/s to multiple files)



GPFS: Information Lifecycle Management

ILM Support basic Technology

- Storage pool – group of LUNs

- Fileset - define subtrees of a file system

- Policies – for rule based management of files inside the storage pools

What does it offer

- One global file system name space across a pool of independent Storage

- Files in the same directory can be in different pools

- Files placed in storage pools at create time using policies

- Files can be moved between pools for policy reasons

- Can be used for hierarchical arranged storage based on files

- Allows classification of data according to SLAs

Bronze

StorageArea

Network

SilverGold

Tape

System



Storage Policies

Rules to control the placement, migration, and retention of files

Declarative SQL-like language

Rule types:- Placement policies, evaluated at file creation, example

rule 'home' set pool 'gold' for fileset 'home_sven'

rule 'otherfiles' set pool 'silver'

- Migration policies, evaluated periodically

rule 'cleangold' migrate from pool 'gold' threshold (90,70) to pool 'silver'

rule 'hsm' migrate from pool 'sata' threshold(90,85) weight(current_timestamp – access_time) to pool 'hsm' where file_size > 1024kb

rule 'cleansilver' when day_of_week()=Monday migrate from pool 'silver' to pool 'bronze' where access_age > 30 days

- Deletion policies, evaluated periodically

rule 'purgebronze' when day_of_month()=1 delete from pool 'bronze' where access_age>365 days

- There are Policies for Backup/Archive, Restore/Retrieve as well



File Placement is also useful for Databases

Databases by default places index and data table spaces in the same directory

File placement policies can direct the data to a pool with appropriate performance characteristics using file extension.

- Data SMS table spaces have a .DAT extension

- Index SMS table spaces have an .INX extension

RAID 1Index_pool

RAID 5data_pool

/DATA/MYDB/ SQL00002.DATSQL00002.INXSQL00003.DATSQL00003.INXSQL00004.DATSQL00004.INX

Rule ‘Data rule’ set pool ‘index_pool’Where UPPER(NAME) like ‘%.INX’

© Copyright IBM Corporation 2008


Scan 1 Billion files in 95 minutes on 1 node!Less than 30 on 4 and less that 15 on 8!

Using 8x IBM eServer xSeries 336 (3.2GHz Intel Xeon)attached at 4Gb/s to

IBM DS-4800 (2.1 TB RAID1) & IBM DS-4100 (10.2TB RAID5)

Scalable ILM Policy Scan Time

0102030405060708090

100

1 2 3 4 5 6 7 8

Number of nodes

Tim

e i

n m

inu

tes

Directory Walk Inode Scan Total Scan Time



GPFS: Replication & Cross-cluster mounts

Replication

- Synchronous intra-cluster replication on file level handled by GPFS nodes

- Replication can be set up for subsets of the file system

Cross-cluster mounts

- In a setup with two distinct GPFS clusters, one can mount the file system owned by the other cluster

- All locking and metadata operations are routed through the owning cluster, block I/O can be done directly by the cross-mounting cluster

- The cross-mounting cluster can mount the file system into one of its own file systems (single mount point).



Snapshot Support

/fs1/file1/fs1/file2/fs1/subdir1/file3/fs1/subdir1/file4/fs1/subdir2/file5

/fs1/file1/fs1/file2/fs1/subdir1/file3/fs1/subdir1/file4/fs1/subdir2/file5/fs1/.snapshots/snap1/file1/fs1/.snapshots/snap1/file2/fs1/.snapshots/snap1/subdir1/file3/fs1/.snapshots/snap1/subdir1/file4/fs1/.snapshots/snap1/subdir2/file5

Read-only copy of directory structure and files

Only changes to the original file consume disk space

mmcrsnapshot fs1 snap1

Writing dirty data to disk Quiescing all file system operations Writing dirty data to disk again Creating snapshot. Resuming operations.

Creating a snapshop



Windows VSS Snapshots Integration

•Integrated into Windows Explorer using the Volume Shadow Copy Services (VSS)



agenda

Conclusion5

The NAS Layer4



NAS today1



IBM Service

What is inside SoFS ?

IBM Storage

IBM BladeCenter

Enterprise Linux

IBM GPFS

CTDB & NAS Exporter

SoFS Mgt Extensions Tivoli S

torage Manager



Another View at the SoFS Components

GPFS

shared SAN disks

storage pools

ctdb

Linux kernel

CIF

S

nfs

vsft

p

apac

he

ssh

d

CIF

S

nfs

vsft

p

apac

he

ssh

d

CIF

S

nfs

vsft

p

apac

he

ssh

d

Linux kernel Linux kernel

pool of public IPscontrolled by ctdb



SoFS CIFS Enhancements

Clustering

- Multiple exports of the same file system over multiple nodes including distributed lock, share and lease support

- Failover capabilities on the server

- Integration with NFS, FTP, HTTP daemons in regard of locking, failover and authorization

Performance optimization with GPFS

NTFS ACL Support in Samba using the native GPFS NFSv4 ACL Support

HSM support within Samba to allow destaging of files to tape and user transparent recall.

VSS integration of GPFS snapshots

Registry based Samba configuration

Alternate Data Streams

Simple install and configuration tools



Availability and Scaling Architecture

SoFS uses an active/active architecture

- All nodes are always up, running and serving files

- In case of a node failure the healthy nodes take over the load of the failed node

- Spare nodes improve overall performance during normal operation

Most NAS solutions use an active/passive architecture

- A spare node is sleeping until its partner fails. At that moment it starts up and takes over the load of the failed partner.

- Failures, misconfiguration, broken cables, etc. of the spare node might remain undiscovered until the partner node fails.

Node 1

Client

failover

Client Client Client

Node 2

...

Node 1

Client Client Client Client

Node 2

...

Active/Passive

Node 1

Client

...failover


Node 2 Node n

...

Node 1

Client

...


Node 2 Node n

...

Active/Active



Availability vs. Disaster Recovery

Availability features in SoFS- Clustering: Protects against individual node failures

- RAID 0,1,5 & 6: Protects against individual disk failures

- Symmetric replication: Protects against storage subsystem failures

Disaster Recovery features in SoFS- Snapshots: Protects against accidental deletion or modification of files

- TSM backup: Protects against complete loss of file system or loss of individual files

- Asynchronous replication: Protects against complete loss of file system or individual files



SoFS Management GUI

Installation wizards

- GPFS

- CIFS & CTDB

Cluster monitoring

- Node availability

- System utilization

- File system utilization

Self-service

- Create and manage project spaces

GPFS management

- Cluster configuration

- File systems

- Disks & NSDs

- Nodes

- Quotas

- Snapshots

- ILM: Storage Pools, Policies, Filesets



Availability Management



Central Event Monitoring System



Central Event Notification System



Node Management



Filesystem Management



Usage/Quota Management



System Utilization Reports



Entry SoFS configuration

3 I/O Nodes, 1 Management Node

up to 48 TB usable capacity per DS3400

Streaming performance with large blocks up to 700 MB/s

Up to 7500 active concurrent users

DS 3400

EXP 3000

EXP 3000

EXP 3000

Midplane

SoF

S B

lade

So

FS

Blade

So

FS

Bla

de

HS

21 Blade

HS

21 Bla

de

HS

21 Bla

de

HS

21 Blade

HS

21 Blade

HS

21 Blade

HS

21 Blade

HS

21 Blade

HS

21 Blade

HS

21 Blade

Mgt. B

lade

AM

MA

MM

Ethernet Switch Module

Midplane


FC Switch Module FC Switch Module

Clientnetwork

BladeC

ente

r H C

hassis



Medium SoFS configuration


2*DS4700 with up to 132 TB usable capacity



MidplaneS

oFS

Blade

So

FS

Blade

So

FS

Bla

de

SoF

S B

lade

SoF

S B

lade

SoF

S B

lade

HS

21 Blade

HS

21 Blade

HS

21 Blade

HS

21 Blade

HS

21 Blade

HS

21 Blade

HS

21 Blade

Mgt. B

lade

AM

MA

MM


Midplane


Clientnetwork

BladeC

ente

r H C

hassis

DS 4700

EXP 810EXP 810

EXP 810

EXP 810EXP 810

EXP 810

DS 4700

EXP 810EXP 810

EXP 810

EXP 810EXP 810

EXP 810




Large SoFS configuration


4*DS4800 with up to 528 TB usable capacity



Midplane

SoF

S B

lade

So

FS

Blade

So

FS

Bla

de

SoF

S B

lade

SoF

S B

lade

SoF

S B

lade

SoF

S B

lade

SoF

S B

lade

SoF

S B

lade

SoF

S B

lade

SoF

S B

lade

SoF

S B

lade

SoF

S B

lade

Mgt. B

lade

AM

MA

MM


Midplane


Clientnetwork

BladeC

ente

r H C

hassis


EXP 810

DS 4800

EXP 810EXP 810EXP 810

EXP 810EXP 810

EXP 810EXP 810EXP 810EXP 810

EXP 810EXP 810

EXP 810EXP 810

External FC Switch

EXP 810

DS 4800

EXP 810EXP 810EXP 810

EXP 810EXP 810

EXP 810EXP 810EXP 810EXP 810

EXP 810EXP 810

EXP 810EXP 810

...



Minimalistic SoFS Configuration

3 Nodes

DS3200 direct SAS-attached

Up to 48 TB usable capacity per DS3200



Ethernet SwitchEthernet Switch

Clientnetwork

DS 3200

x3650 x3650 x3650

EXP 3000

EXP 3000

EXP 3000



SoFS BladeCenter Hardware

BladeCenter H chassis- Up to 6 Ethernet switch modules

• Either GbE or 10 GbE uplinks

- Two 4 Gb Fibre Channel switch modules

- Two management modules- Up to 14 HS-21 blades (one dedicated for management)

• Intel Xeon Quadcore, 8 GB RAM4x1Gb I/O+

1Gb GPFS +1Gb CTDB / Mgt

Midplane

HS

21 B

lade

HS

21 B

lad

e

HS

21 B

lade

HS

21 B

lade

HS

21 B

lade

HS

21 B

lad

eH

S21

Blad

e

HS

21 B

lade

HS

21 B

lade

HS

21 B

lade

HS

21 B

lade

HS

21 B

lad

eH

S21

Blad

e

HS

21 B

lade (M

gt)

AM

M2

AM

M2


Midplane

10GbE Uplink SwMd


2x4Gb = 8Gb

4x2x10Gb= 80Gb

Clientnetwork

2x6x4Gb = 48Gb

Storage

Blad

eCe

nter H

Ch

assis

Dedicated networks for GPFS and CTDB/mgt



SoFS Rack mounted Hardware

X3650- 2 * Intel Xeon Quadcore

- 2 GigE Adapter On-board

- Up to 6* Ethernet or FC Adapter optional

- 4 - 48 GB RAM



Why BladeCenter ? Management efficiency:

- All network interconnect (Ethernet & Fibre Channel) is hard-wired on the midplanes. No chance for wrong or missing cabling. (In a SoFS cluster with 14 blades the midplanes replace 112 network cables)

- Blades can be discovered, monitored, started, stopped etc. through BC management modules.

- Switches are an integrated part of the cluster and can be monitored through BC management modules.

Space efficiency: Fits 14 servers in an 9U package

Power efficiency: The BC power modules are up to 50% more efficient then smaller power supplies found in rack-mounted servers.

Scalability: Add up to 14 blades, the infrastructure (power, cooling, network, management) is already there.

Price: Initial investment looks expensive, but equipped with approximately 5 blades or more the TCO is lower than that of rack mounted servers.



Backup, offline files & Tape

Full support of GPFS snapshots. Users can see existing snapshots in “Previous Version” dialogue of the Windows Explorer and restore them autonomously

SoFS supports LAN-free backup using the Tivoli Storage Manager 5.5

SoFS supports destaging of files to tape using the Tivoli Hierarchical Storage Manager 5.5

Customers have the choice of IBM’s premium LTO3 Tape Libraries

Full support of IBM’s tape encryption technology

TS3500 Tape Library



agenda

Conclusion5

The NAS Layer4



NAS today1



SoFS vs Classic NAS

SoFS combines a Clustered Filesystem with a Global Namespace. It‘s deployed centrally, managed centrally, backed up centrally and grown centrally.

SoFS scales horizontally not vertically. When you need more capacity, you just add more disks. When you need more performance you add nodes and/or disks. There are almost no architectural limits for future growth.

NAS boxes growth vertically. You can add controllers and storage up to the maximum configuration of the box. Beyond that you have to buy a new one.

SoFS provides integrated Information Lifecycle Management

- Different tiers of storage, e.g. FC, SATA and tape

- Policy driven placement and migration of files over their entire Lifetime

SoFS offers synchronous Block Level replication for data and Metadata

SoFS offers multi-site configurations via the cross-cluster mount functionality.

SoFS allows to build Highly Available Systems including Disaster resistant Systems across multiple sites



Questions ?



Special Notices

Copyright IBM Corporation, 2006

– This presentation was produced in the United States. IBM may not offer the products, programs, services or features discussed herein in other countries, and the information may be subject to change without notice. Consult your local IBM business contact for information on the products, programs, services, and features available in your area. Any reference to an IBM product, program, service or feature is not intended to state or imply that only IBM's product, program, service or feature may be used. Any functionally equivalent product, program, service or feature that does not infringe on any of IBM's intellectual property rights may be used instead of the IBM product, program, service or feature.

– The [e(logo)server] brand consists of the established IBM e-business logo followed by the descriptive term "server".

– Information in this presentation concerning non-IBM products was obtained from the suppliers of these products, published announcement material or other publicly available sources. Sources for non-IBM list prices and performance numbers are taken from publicly available information including D.H. Brown, vendor announcements, vendor WWW Home Pages, SPEC Home Page, GPC (Graphics Processing Council) Home Page and TPC (Transaction Processing Performance Council) Home Page. IBM has not tested these products and cannot confirm the accuracy of performance, compatibility or any other claims related to non-IBM products. Questions on the capabilities of non-IBM products should be addressed to the suppliers of those products.

– IBM may have patents or pending patent applications covering subject matter in this presentation. The furnishing of this presentation does not give you any license to these patents. Send license inquires, in writing, to IBM Director of Licensing, IBM Corporation, New Castle Drive, Armonk, NY 10504-1785 USA.

– All statements regarding IBM's future direction and intent are subject to change or withdrawal without notice, and represent goals and objectives only. Contact your local IBM office or IBM authorized reseller for the full text of a specific Statement of General Direction.

– The information contained in this presentation has not been submitted to any formal IBM test and is distributed "AS IS". While each item may have been reviewed by IBM for accuracy in a specific situation, there is no guarantee that the same or similar results will be obtained elsewhere. The use of this information or the implementation of any techniques described herein is a customer responsibility and depends on the customer's ability to evaluate and integrate them into the customer's operational environment. Customers attempting to adapt these techniques to their own environments do so at their own risk.

– IBM is not responsible for printing errors in this presentation that result in pricing or information inaccuracies.

– The information contained in this presentation represents the current views of IBM on the issues discussed as of the date of publication. IBM cannot guarantee the accuracy of any information presented after the date of publication.

– All prices shown are IBM's suggested list prices; dealer prices may vary.

– IBM products are manufactured from new parts, or new and serviceable used parts. Regardless, our warranty terms apply.

– Information about non-IBM products was obtained from suppliers of those products. IBM makes no representations or warranties regarding these products. Non-IBM products are offered and warranted by third-parties, not IBM.



Special Notices (contd.)

– Information provided in this presentation and information contained on IBM's past and present Year 2000 Internet Web site pages regarding products and services offered by IBM and its subsidiaries are "Year 2000 Readiness Disclosures" under the Year 2000 Information and Readiness Disclosure Act of 1998, a U.S statute enacted on October 19, 1998. IBM's Year 2000 Internet Web site pages have been and will continue to be our primary mechanism for communicating year 2000 information. Please see the "legal" icon on IBM's Year 2000 Web site (www.ibm.com/year2000) for further information regarding this statute and its applicability to IBM.

– Any performance data contained in this presentation was determined in a controlled environment. Therefore, the results obtained in other operating environments may vary significantly. Some measurements quoted in this presentation may have been made on development-level systems. There is no guarantee these measurements will be the same on generally-available systems. Some measurements quoted in this presentation may have been estimated through extrapolation. Actual results may vary. Users of this presentation should verify the applicable data for their specific environment.

– The following terms are registered trademarks of International Business Machines Corporation in the United States and/or other countries: AIX, AIXwindows, AS/400, C Set++, CICS, CICS/6000, DataHub, DataJoiner, DB2, DEEP BLUE, DYNIX, DYNIX/ptx, e(logo), ESCON, IBM, IBM(logo), Information Warehouse, Intellistation, IQ-Link, LANStreamer, LoadLeveler, Magstar, MediaStreamer, Micro Channel, MQSeries, Net.Data, Netfinity, NUMA-Q, OS/2, OS/390, OS/400, Parallel Sysplex, PartnerLink, PartnerWorld, POWERparallel, PowerPC, PowerPC(logo), ptx/ADMIN, RISC System/6000, RS/6000, S/390, Scalable POWERparallel Systems, SecureWay, Sequent, SP2, System/390, The Engines of e-business, ThinkPad, Tivoli(logo), TURBOWAYS, VisualAge, WebSphere. The following terms are trademarks of International Business Machines Corporation in the United States and/or other countries: AIX/L, AIX/L(logo), AIX 5L, AIX PVMe, Application Region Manager, AS/400e, Blue Gene, Chipkill, ClusterProven, DB2 OLAP Server, DB2 Universal Database, e(logo)business, GigaProcessor, HACMP/6000, Intelligent Miner, iSeries, Network Station, NUMACenter, PowerPC Architecture, PowerPC 604, POWER2 Architecture, pSeries, Sequent (logo), SequentLINK, Service Director, Shark, SmoothStart, SP, Tivoli Enterprise, TME 10, Videocharger, Visualization Data Explorer, xSeries, zSeries. A full list of U.S. trademarks owned by IBM may be found at http://iplswww.nas.ibm.com/wpts/trademarks/trademar.htm.

– Lotus and Lotus Notes are registered trademarks and Domino and Notes are trademarks of Lotus Development Corporation in the United States and/or other countries.

– NetView, Tivoli and TME are registered trademarks and TME Enterprise is a trademark of Tivoli Systems, Inc. in the United States and/or other countries.

– Microsoft, Windows, Windows NT and the Windows logo are registered trademarks of Microsoft Corporation in the United States and/or other countries.

– UNIX is a registered trademark in the United States and other countries licensed exclusively through The Open Group.

– LINUX is a registered trademark of Linus Torvalds.– Intel and Pentium are registered trademarks and MMX, Itanium, Pentium II Xeon and Pentium III Xeon are trademarks of Intel Corporation in

the United States and/or other countries. – Java and all Java-based trademarks and logos are trademarks of Sun Microsystems, Inc. in the United States and/or other countries. – Other company, product and service names may be trademarks or service marks of others.

Documents

Sven Oehme – STG Industry Technology Lead [email protected] · GPFS: Information Lifecycle Management ILM Support basic Technology - Storage pool – group of LUNs - Fileset - define