#ibmedge© 2016 IBM Corporation

A Diversified Analytic Solution Based on Various Storage ServicesSep 21, 2016

#ibmedge2

Agenda

• Spectrum Scale Deliveries for Data Analytics

• Spectrum Scale HDFS Transparency Hadoop Connector

• Advanced Data Management and Protection for Analytics

• Spectrum Scale Data Analytics Deployment Models

• Solutions dedicated to Client’s Success

#ibmedge3

Hadoop Data Analytics with IBM Spectrum Scale

• IBM Spectrum Scale is a proven enterprise ready alternative of HDFS for Hadoop data analytics

• Completely compatible with HDFS client API and Hadoop applications

• Unified File, Object and HDFS access interface

• Mature enterprise features make it easy to manage and protect data

#ibmedge4

IBM Spectrum Storage and Computing for Data Analytics

• IBM Spectrum Storage family and Spectrum Computing provide integrated solution for Data Analytics workload.

• Mature enterprise class and proven features in Spectrum Storage provides rich of data management, monitor, security enhancement.

• Spectrum Computing provide high performance and cost effective services for Data Analytics workload.

IBM Spectrum Control

IBM Spectrum Protect

IBM Spectrum Archive

IBM Spectrum Virtualize

IBM Spectrum Accelerate

IBM Spectrum Scale

IBM Spectrum Conductor

IBM Spectrum ConductorWith Spark

IBM Spectrum LSF

IBM Spectrum Storage IBM Spectrum Computing

Data Analytics Services

#ibmedge 5

Spectrum Scale - Unified Software Defined StorageUnified Scale-out Data Lake• File In/Out, Object In/Out; Analytics on demand.• High-performance native protocols• Single Management Plane• Cluster replication & global namespace• Enterprise storage features across file, object & HDFS

Spectrum Scale

Expand to other Storage Systems

(DeepFlash 150 or others)Expand to server local disk drives

……

FDR/QDR IB or 10/40 GBE

POSIX File-System APISMB

Windows File Sharing

NFS Object iSCSI

Linux / UNIX File Sharing

Web & Swift / S3 Compatible

Block Storage (Experimental)

Protocol Nodes managed by CES Framework

General Applications

Elastic Storage Server

HDFS Transparency

External Hadoop / Spark Cluster

And CloudSpectrum Scale

TCT

#ibmedge 6

IBM Elastic Storage Server and IBM Data Engine

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24

FC 5887

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24

FC 5887

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24

FC 5887

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24

FC 5887

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24

FC 5887

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24

FC 5887

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24

FC 5887

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24

FC 5887

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24

FC 5887

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24

FC 5887

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24

FC 5887

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24

FC 5887

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24

FC 5887

IBM Elastic Storage Server (ESS) is Powered by IBM Spectrum Scale RAID

delivers:

Analytics

• ESS• Spectrum Computing• IOP with Hadoop & Spark• Watson Explorer

Hadoop & Spark

• Spectrum Scale• Spectrum Symphony• Platform Cluster Manager• IOP with Hadoop and Spark

NoSQL• CAPI enabled Power System

SAP NANA

• ESS Certified• IBM Power Certified

IBM Data Engine for:

Optimal Solution for All Data Needs Single repository of data with unified file, Hadoop and object support

Highest Reliability Mean Time to Failure many orders of magnitude longer than the life of

the hardware

Highest Performance Faster than rated disk speeds

Cost Effectiveness Less then 1 cent per GB per month

7© 2016 IBM Corporation #ibmedge

Spectrum Scale HDFS Transparency Hadoop Connector

#ibmedge8

We are transforming

• Storage & Compute integrated in the same node• Compute node has to access Spectrum Scale

file system directly• Build system from scratch

Hadoop Client

Hadoop FileSystem API

Connector on Spectrum Scale

API & POSIX

Spectrum Scale

Node

Hadoop Client

Hadoop FileSystem API

Connector on Spectrum Scale

API & POSIX

Node

Hadoop Client

Hadoop FileSystem API

Connector on Spectrum Scale API & POSIX

Node

Hadoop Client

Hadoop FileSystem API

HDFS Client

Node

Hadoop Client

Hadoop FileSystem API

HDFS Client

Node

Hadoop Client

Hadoop FileSystem API

HDFS Client

Node

HDFS RPC over network

Spectrum Scale Connector Service

Node

Connector on Spectrum Scale API &

POSIX

Spectrum Scale Connector Service

Node

Connector on Spectrum Scale API

& POSIX

Spectrum Scale

Dedicate Hadoop compute cluster Compatible with more Hadoop components

and features, such as Impala, distcp, webhdfs and so on.

Enhanced security. Full Kerberos support. Improved performance. Leverage HDFS client

cache.

#ibmedge9

Based on Hadoop RPCIntegrated system for data analytics

• Seamless supports Hadoop application without any changes

• Not necessary to mount file system in compute node

• Leverage HDFS client cache to improve performance and security

HDFS RPC Similar FileSystem method Implemented By ------------------------------------------------------------------ ------------------------------------------ ---------------------- rpc getBlockLocations(GetBlockLocationsRequestProto); getFileBlockLocations Spectrum Scale API rpc getServerDefaults(GetServerDefaultsRequestProto); get default value from configuration file Spectrum Scale API rpc create(CreateRequestProto)returns(CreateResponseProto); create POSIX I/O Interface rpc append(AppendRequestProto) returns(AppendResponseProto); append POXIX I/O Interface rpc setReplication(SetReplicationRequestProto) setReplication

Spectrum Scale API rpc setPermission(SetPermissionRequestProto) setPermission Spectrum Scale API rpc setOwner(SetOwnerRequestProto) returns(SetOwnerResponseProto); setOwner Spectrum Scale API rpc setSafeMode(SetSafeModeRequestProto) HDFS specific, can skip…………

#ibmedge10

Data Integration with HDFS

https://hadoop.apache.org/docs/r1.2.1/distcp.html https://hadoop.apache.org/docs/r2.7.2/hadoop-project-dist/hadoop-hdfs/Federation.html

distcp

Use MapReduce for large inter/intra-cluster data copying between native HDFS and Spectrum Scale

Use the same semantic as copying data between HDFS

Locality awareness and data is also balanced

Federation

Spectrum Scale storage can be co-exist with native HDFS storage to service the same Hadoop workload

Spectrum Scale storage can be added to exsiting HDFS storage without any change.

viewFs is invoked to provide single namespace view

11© 2016 IBM Corporation #ibmedge

Advanced Data Management and Protection for Analytics

#ibmedge12

Locality Awareness Data Recovery• Map/Reduce jobs

• Map/Reduce job comprises of tasks and each task will process one HDFS-called data block

• Disk failure or node failures won’t impact much on data locality in native HDFS

• Hbase Hfile data locality• If one node fails, the Hbase region server over that node is down• The region server file will be assigned to other region server for

service• The data locality of region files can only be restored when major

compaction happens if taking native HDFS•Hive table files

• In native HDFS, no easy way to restore/control the hive table file’s data locality

• Easy data locality restore in Spectrum Scale• WADFG/mmchattr/mmresripefile

A B C D

Node A has all data for Hbase region / Hive hfile for better performance

XA B C D

When node A fails, data have to migrate to other node to keep data safe

√

A B C D A B C D

√

Spectrum Scale locality awareness

data recovery

HDFS natively won’t recovery data to original node

#ibmedge13

Rack-awareness Data Locality for Shared Storage, IBM DeepFlash and ESS Benefit for:

• More than 1 rack in the file system• One shared storage(or IBM DeepFlash and

ESS) per rack• Top of Rack switch network

Advantages:• Map/Reduce could be scheduled to the rack

where the input data are located• Reduce the network traffic for inter-rack

switch• Performance gain from fast reading from local

rack

<property> <name>gpfs.storage.type</name> <value>rackaware</value> </property>

#ibmedge14

Transparent Data Compression

• Transparent compression for HDFS transparency, Object, NFS, SMB and POSIX interface.

• Improved storage efficiency• Typically 2x improvement in storage efficiency

• Improved i/o bandwidth• Read/write compressed data reduces load on

storage

• Improved client side caching• Caching compressed data increases apparent

cache size

Traditional Application (POSIX)

Spectrum Scale Node

Native Compression

Block Device / NSD

Disk

Object/NFS/SMB

HDFS Transparency

Vision

• Per file compression

• Use policies

• Compress low use data

#ibmedge15

Transparent Data Encryption• Native Encryption of data at rest

• Files are encrypted before they are stored on disk

• Keys are never written to disk• No data leakage in case disks are stolen

or improperly decommissioned

• Secure deletion• Ability to destroy arbitrarily large subsets

of a file system• No ‘digital shredding’, no overwriting:

Security deletion is a cryptographic operation

• Use Spectrum Scale Policy to encrypted (or exclude) files in fileset or file system

• Generally < 5% performance impact Sequential Read Sequential Write

Traditional Application (POSIX)

Spectrum Scale Node

(Encryption/Decryption)

IBM Security Key Lifecycle Manager

Vormetric Key ManagerBlock Device / NSD

Disk

Object/NFS/SMB

HDFS Transparency

#ibmedge16

Spectrum Scale for Hadoop Security and Audit• Spectrum Scale HDFS Transparency + IBM

Security Guardium Data Activity Monitor• Spectrum Scale enables IBM Security Guardium

Data Active Monitor for Hadoop data access Audit

• Spectrum Scale HDFS Transparency support Ranger

• Ranger is open source and used for security/audit• IBM IOP 4.2 includes this component.

• Spectrum Scale HDFS Transparency fully supports Kerberos

• ACL

• Supports NameNode Block Access Token

Spectrum Scale

Ranger

HDFS Transparency

Hadoop Platform

YARN MapReduceSpark

HBase

Zookeeper

Flume

Hive Pig

Sqoop

HCatalog

Solr/Lucene

kafka

#ibmedge

Transparent Cloud TieringExpanding storage to the cloud

|

17

• Archive to object or tape• Tier to Spectrum Archive incl. dual copies for long term

retention• Tier to Cleversafe as an option for disk at lower TCO

• Cognitive engine to move data based upon ‘heat’ or policy, even across the Internet

High capacity tiered storage, mixed data access requirements for high-performance & retention

• Media storage for raw, processing, storage

• Digital surveillance for ingest, plus retention

• Technical Computing for clusters, data analytics and aged results

System pool(Flash)

Gold pool(Disk)

Tape LibraryCloud Object Storage

OR

One cloud or tape per filesystem

#ibmedge18

Data DR Derived from Spectrum Scale• Hbase Cluster Replication

• WAL-based asynchronous way• All nodes in both cluster should be accessible for each other• Both clusters could provide Hbase service on the same time• Only available for Hbase• Supported over HDFS Transparency + Spectrum Scale

• Spectrum Scale AFM IW-based replication• AFM IW-based replication from production cluster(cache site)

to standby cluster(home site)• Both sites should be Spectrum Scale cluster for Hadoop

application failover• Only one cluster can provide Hbase service(conflict in

assigning region servers if Hbase is up on both cluster)

• Spectrum Scale Active-Active DR• 2 replica in production cluster; another replica in standby

cluster• Dedicated network for two clusters(10Gb bandwidth)• Distance between two clusters is less than 100Km• Can achieve RPO=0 in DR

Spectrum Scale

HDFS TransparencySpectrum Scale

HDFS TransparencyAFM

WAN

Spectrum ScaleHDFS Transparency HDFS Transparency

Dedicated network

19© 2016 IBM Corporation #ibmedge

Spectrum Scale Data Analytics Solution Deployment Models

#ibmedge 20

In-place Data Analytics

A A A

Existing System Analytics SystemData ingest

Export result

A A A

Existing System

Spectrum Scale File SystemFile Object

Analytics System

HDFS Transparency

Traditional Analytics Solution

In-place Analytics Solution

• Build analytics system from scratch, not only for compute but also for storage

• Add storage and compute resource at the same time no matter it’s required

• Native HDFS doesn’t support native POSIX• Lacks of enterprise data management and

protection capability

Can leverage existing Spectrum Scale storage Unified interface for File and Object analytics POSIX compatibility Mature enterprise data management and

protect solutions derived from Spectrum Storage family and 3rd part components

Transparency Cloud Tiering

#ibmedge 21

Storage-rich Server Based Deployment Model

S FServer License FPO License

Storage-Rich Servers• IBM Power• Commercial X86

S FS S F

Spectrum Scale BD&A Cluster

10 GigE / 40 GigE / InfiniBand (Spectrum Scale I/0)

10 GigE / 1 GigE (Application, Management)

Spectrum Scale FPO

Spectrum Scale HDFS Transparency

Hadoop

Application

HDFS RPC

Spectrum Scale FPO, Hadoop and Application are installed in all nodes in cluster.

Write data to directly to local disk when RPC destination is local node.

Spectrum Scale FPO• Use the same binary but with different deployment model

and configuration.• Spectrum Scale (FPO) has capability to store data and

metadata in different Storage Pool (set of disks) to provide high performance and availability.

• The same locality as HDFS. Saves the 1st replica in local disks and stripe other replica(s) to other failure group.

• Optimized for READ.• WRITE data has to go through network.

FPO storage pool configurations:• AllowWriteAffinityDepth

• Enable FPO storage pool (horizontal block allocation type)• WriteAffinityDepth

• 0 indicates that each replica is to be striped across the disks in a cyclical fashion. (General GPFS)

• 1 indicates that the first copy is written to the writer node. The second copy is written to a different rack. The third copy is written to the same rack as the second copy, but on a different half. (FPO)

• 2 indicates that the first copy is written to the writer node. The second copy is written to the same rack as the first copy, but on a different half.

• BlockGroupFactor• Effective block size = file system block size * BlockGroupFactor

#ibmedge22

SAN Shared Storage Based Deployment Model• Compute Nodes only needs to install Hadoop components,

like Hbase, and HDFS client to service Analytics workload.

• Spectrum Scale cluster is built on SAN shared storage.

• Spectrum Scale HDFS Transparency Connector (includes NameNode and DataNode) are deployed in Spectrum Scale NSD Servers or dedicate client nodes.

Software defined storage supports various platform, such as IBM Power and commercial X86, and storage system, such as IBM DeepFlash 150 and other RAID system.

No data recovery when compute node or Spectrum Scale Transparency node (NameNode or DataNode) failed.

Easy to extend storage capacity, I/O performance, Analytics performance separately.

Leverage SAN shared storage to provide high performance and high availability derived from FC and InfiniBand (and RDMA) and Spectrum Scale replication.

S Server License

S S S

FC Switch / InfiniBand (RDMA)

10 GigE / 1 GigE

RAID Storage

Compute Nodes• IBM Power or X86• Only Hadoop services and

HDFS client

Spectrum Scale BD&A Cluster

•HDFS Transparency NameNode/DataNode

•IBM Power or X86

C Client License

C C C

#ibmedge23

IBM DeepFlash 150 Based Deployment Model• IBM DeepFlash 150

• 3U chassis starting at 128TB up to 512TB • 8 to 64 8TB flash cards• Up to 2M raw IOPS • sub 1ms latency• 12GB/s throughput

• Data platform for Analytics (Hadoop, Spark, SAP Hana)• Faster time to insights• Load and off load data to and from memory faster• Shared data platform for multiple instance and

forms of analytics

IBM DeepFlash 150

S Server License

S S S

SAS Connection

10 GigE / 1 GigE

RAID Storage

Compute Nodes• IBM Power or X86• Only Hadoop services and

HDFS client

Spectrum Scale BD&A Cluster

•HDFS Transparency NameNode/DataNode

•IBM Power or X86

#ibmedge 24

ESS Based Deployment Model

C C C

10 GigE / 1 GigE

Spectrum Scale BD&A Cluster

ESSE E

InfiniBand (RDMA) / 40 GigE / 10 GigE

C Client License

Compute Nodes• IBM Power or X86• Only Hadoop services and

HDFS client

• Compute Nodes only needs to install Hadoop components, like Hbase, and HDFS client to service Analytics workload.

• Spectrum Scale cluster is built on IBM Elastic Storage Server (ESS).

• Spectrum Scale HDFS Transparency Connector (includes NameNode and DataNode) are deployed in dedicate client nodes which joined into Spectrum Scale cluster in ESS.

ESS provides high performance and reliability powered by Spectrum Scale RAID

Software De-Cluster RAID Erasure code ready End-to-end checksum

Supports FDR/QDR InfiniBand and RDMA ESS can also service other workload, such as OpenStack

at the same time

25© 2016 IBM Corporation #ibmedge

Solutions dedicated to Client’s Success

#ibmedge 26

Analytics on Docker with Spectrum Scale

Unified Data Plane for Docker’s Orchestration

Fileset X Fileset Y

Fileset Z

Image File A

Image File B

Image File C

……DockerContainer A

SharedPriv.

DockerContainer B

SharedPriv.

DockerContainer C

SharedPriv.

Fileset X Fileset YVirtual Data Plane acrossDocker instances

Multi-tenant Storage-Centric Services in Cloud

Managed by Docker’s Orchestration

IBM Spectrum Computing

GPFS File (w/ file clone) as the runtime Image

GPFS Fileset as shared volume between Docker containers

Shared Fileset build up the virtual data planes across containers.

Spectrum Scale control-plane can transparently control the virtual data plane

Achieve/Snapshot/DR can be done at Fileset level.

IBM DeepFlash 150 Disk

Tape

ESS

Single name space

Spectrum Scale

Share NothingCluster

Spectrum Scale AFM

Spectrum Scale TCT

#ibmedge 27

Solutions for Data Warehouse – SAP HAHA

node03

Spectrum Scale FPO + SAP NAHAnode01 node02

shared filesystem - GPFSHDD flash

data01 log01

DB partition 1

- index server- statistic server- SAP HANA® studio

- SAP HANA® DB worker node

HDD flash

DB partition 2

- index server- statistic server

- SAP HANA® DB worker node

SAP HANA® DB

data02 log02

HDD flash

DB partition 3

- index server- statistic server

- SAP HANA® DB worker node

data03 log03 Key advantages from Spectrum Scale: Disk failure tolerance Node failure tolerance Data locality/data locality restore Fast repair when bringing down disk back

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24

FC 5887

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24

FC 5887

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24

FC 5887

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24

FC 5887

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24

FC 5887

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24

FC 5887

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24

FC 5887

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24

FC 5887

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24

FC 5887

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24

FC 5887

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24

FC 5887

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24

FC 5887

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24

FC 5887

IBM Elastic Storage Server is SAP NAHA Certified Enterprise Storage

#ibmedge28

Solutions for Data Warehouse – IBM DB2• Each Data Module

• Has a set of local disks• Hosts a set of Partitions (1-N)

• 3 Module building block• GPFS File system over building block

• One or Two replicas• Failure of any one Failure Group does not affect GPFS

operation

• High Availability• Distributed or mutual failover• Data fully visible across DPF cluster• Data locally available within building block

• Optional - local RAID for disk• Avoid rebuild due to disk failure

Spectrum Scale + IBM DB2 DPF / DashDB

#ibmedge 29

Bill Analytics System for a Telecom Operators

Data Collection Sub-system ETL System

Network equipment's data collection

Internet activities capture andpre-processing (merge, etc)

Real time sales/marketing

sub-system

PowerLinux

PowerLinux

PowerLinux

Spectrum Computing

Spectrum Scale FPO

Spectrum Scale Multi-cluster

PowerLinux

PowerLinux

PowerLinux

Spectrum Computing

Spectrum Scale FPO

Built on 2013 runs Hadoop 1.x Built on 2015 runs Hadoop 2.x

NoSQL database

Import aggregated results into NoSQL database

Customer Behavior

Subsystem

Other business

sub-system

HBase MapReduce MapReduce Spark HBase Hive

PowerLinux

More than 100 Nodes Process 40 TB data each day Spectrum Scale FPO manages more than

2600 disks (2.6 PB raw capacity) in this system.

Spectrum Scale and Spectrum Computing is the most stable Big Data system we have. Spectrum Scale POSIX access interface and Multi-cluster feature resolves data balance issue and provides 5X performance improvement in data loading phase. -- Client IT Architecture

#ibmedge

ESS Services Hadoop and HPC Mixed Workload

IBM Power E880

IBM Elastic Storage Server (ESS)

Compute Nodes

HDFS Transparency Nodes

Hadoop Dev Cluser Hadoop Test Cluser Hadoop PROD Cluser HPC Cluser

Spectrum Scale Cluster

Key Advantages - IOP with Value-Add (Spectrum Scale + Spectrum Computing) - Share the same ESS storage for both Hadoop clusters and HPC cluster - Deploying several Hadoop clusters over one file system - Mixed workloads, scaling storage and computing capability separately

#ibmedge

Notices and Disclaimers

31

Copyright © 2016 by International Business Machines Corporation (IBM). No part of this document may be reproduced or transmitted in any form without written permission from IBM.

U.S. Government Users Restricted Rights - Use, duplication or disclosure restricted by GSA ADP Schedule Contract with IBM.

Information in these presentations (including information relating to products that have not yet been announced by IBM) has been reviewed for accuracy as of the date of initial publication and could include unintentional technical or typographical errors. IBM shall have no responsibility to update this information. THIS DOCUMENT IS DISTRIBUTED "AS IS" WITHOUT ANY WARRANTY, EITHER EXPRESS OR IMPLIED. IN NO EVENT SHALL IBM BE LIABLE FOR ANY DAMAGE ARISING FROM THE USE OF THIS INFORMATION, INCLUDING BUT NOT LIMITED TO, LOSS OF DATA, BUSINESS INTERRUPTION, LOSS OF PROFIT OR LOSS OF OPPORTUNITY. IBM products and services are warranted according to the terms and conditions of the agreements under which they are provided.

IBM products are manufactured from new parts or new and used parts. In some cases, a product may not be new and may have been previously installed. Regardless, our warranty terms apply.”

Any statements regarding IBM's future direction, intent or product plans are subject to change or withdrawal without notice.

Performance data contained herein was generally obtained in a controlled, isolated environments. Customer examples are presented as illustrations of how those customers have used IBM products and the results they may have achieved. Actual performance, cost, savings or other results in other operating environments may vary.

References in this document to IBM products, programs, or services does not imply that IBM intends to make such products, programs or services available in all countries in which IBM operates or does business.

Workshops, sessions and associated materials may have been prepared by independent session speakers, and do not necessarily reflect the views of IBM. All materials and discussions are provided for informational purposes only, and are neither intended to, nor shall constitute legal or other guidance or advice to any individual participant or their specific situation.

It is the customer’s responsibility to insure its own compliance with legal requirements and to obtain advice of competent legal counsel as to the identification and interpretation of any relevant laws and regulatory requirements that may affect the customer’s business and any actions the customer may need to take to comply with such laws. IBM does not provide legal advice or represent or warrant that its services or products will ensure that the customer is in compliance with any law

#ibmedge

Notices and Disclaimers Con’t.

32

Information concerning non-IBM products was obtained from the suppliers of those products, their published announcements or other publicly available sources. IBM has not tested those products in connection with this publication 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 does not warrant the quality of any third-party products, or the ability of any such third-party products to interoperate with IBM’s products. IBM EXPRESSLY DISCLAIMS ALL WARRANTIES, EXPRESSED OR IMPLIED, INCLUDING BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE.

The provision of the information contained h erein is not intended to, and does not, grant any right or license under any IBM patents, copyrights, trademarks or other intellectual property right.

IBM, the IBM logo, ibm.com, Aspera®, Bluemix, Blueworks Live, CICS, Clearcase, Cognos®, DOORS®, Emptoris®, Enterprise Document Management System™, FASP®, FileNet®, Global Business Services ®, Global Technology Services ®, IBM ExperienceOne™, IBM SmartCloud®, IBM Social Business®, Information on Demand, ILOG, Maximo®, MQIntegrator®, MQSeries®, Netcool®, OMEGAMON, OpenPower, PureAnalytics™, PureApplication®, pureCluster™, PureCoverage®, PureData®, PureExperience®, PureFlex®, pureQuery®, pureScale®, PureSystems®, QRadar®, Rational®, Rhapsody®, Smarter Commerce®, SoDA, SPSS, Sterling Commerce®, StoredIQ, Tealeaf®, Tivoli®, Trusteer®, Unica®, urban{code}®, Watson, WebSphere®, Worklight®, X-Force® and System z® Z/OS, are trademarks of International Business Machines Corporation, registered in many jurisdictions worldwide. Other product and service names might be trademarks of IBM or other companies. A current list of IBM trademarks is available on the Web at "Copyright and trademark information" at: www.ibm.com/legal/copytrade.shtml.

© 2016 IBM Corporation #ibmedge

Thank You