Flash and POWER For VUG

Download as pdf or txt
Download as pdf or txt
You are on page 1of 82

FlashSystem family 2015

TM
IBM FlashSystem : Where to Use Flash
Jeffrey Stilwell,
Business Unit Executive, NA SDFS

© 2015 IBM Corporation Title of presentation goes here


Over 1000 FlashSystem Customers in 6 Months

2 © 2015 IBM Corporation


A new generation of IBM POWER systems addresses CPU
performance

Designed for
Big Data
POWERS822L POWER S824 or POWER S814

Superior
Cloud POWER S822
POWER S812L
Economics
 1 or 2 sockets  1 or 2 sockets
Open  10 or 12 cores/socket
 Up to 1 TB of Memory
 6, 8,10 or 12 cores/socket
 Up to 1 TB of Memory
Innovation
Platform

© 2016
3 IBM Corporation
© 2015 IBM Corporation IBM FlashSystem | 3
E880
HANA on POWER Portfolio E880C
(for Prod and non-Prod usage)
E870
S824 E870C
S814*
S822 E850
E850C
Scale-Out

Enterprise
S824L
HANA on Power IBM fast-start option
Industry best-practice
S822L Solution Editions Quick configurations
Aggressively priced
S812L* Rapid deployment of
Linux only Systems SAP HANA on IBM Power Systems

4
*not SAP certified for productive usage,
4 © 2015 IBM Corporation only at least 2 socket systems are certified for HANA productive usage
IBM Systems Flash Storage Offerings Portfolio
IBM
Power
Systems
Storwize Storwize FlashSystem FlashSystem FlashSystem IBM Elastic DS888xF
V5030F V7000F V9000 A9000 A9000R Storage Server

Entry Storage Midrange Storage Highest


Virtualization Virtualization Performance
Cloud service
Large deployments Big Data Business Critical
providers
z/OS | AIX
• Simplified management • Consolidates file &
• Flexible consumption model
• Full time data • Full time data Power HA
reduction reduction object workloads
• Virtualized, enterprise-class, flash-optimized, modular storage • Faster data Power i HA
• Workloads: Cloud, • Workloads: Mixed
• Enterprise class heterogeneous data services and selectable data reduction analysis
VDI, VMware and cloud • Business critical,
• Global sharing
deepest
integration with z
Systems
• Superior
performance &
FlashSystem 900 • Extreme performance reliability
SVC Application acceleration • Targeting database acceleration • Three-site / Four-
& Spectrum Storage booster site replication
• DS8884F,
Enhanced data storage functions, economics DS8886F,
and flexibility with sophisticated virtualization IBM FlashCore™ Technology Optimized DS8888F
© Copyright IBM Corporation 2017 IBM & IBM Business Partner Use Only

5 5
© 2015 IBM Corporation
Infrastructure challenges impact analytics

CPU performance up 10x this last decade


Storage has grown capacity but unable to
keep up in performance
Systems are now Latency & IO bound Performance
resulting in significant performance gap Gap

© 2016
6 IBM Corporation
© 2015 IBM Corporation IBM FlashSystem | 6
Client responses to infrastructure challenges

Wasteful,
HDD Expensive &
Most Costly Add More Ineffective
Performance
& Volatile Memory with Storage
Enhancement Latency
Issues
Typical
Performance
Mitigation
Tactics
Only Time
effective if Consuming,
Tune & Modify
storage is Add CPUs Very
not a
Application Expensive
bottleneck & Risky

© 2016
7 IBM Corporation
© 2015 IBM Corporation IBM FlashSystem | 7
These adjustments are responsible for most IT woes
Countless SW Dissatisfied Users
Licenses and Customers

Complicated SW Slow or Unstable


Applications

Expensive and
Bloated Infrastructure Loss of Business
Opportunity

© 2016
8 IBM Corporation
© 2015 IBM Corporation IBM FlashSystem | 8
FlashSystem

Enabling the use of


Flash in the Enterprise

9 ©© 2015
2015 IBM
IBM Corporation
Corporation
We’ve arrived here by way of 3 technology
transformations and one epic problem
• Processor performance increases have followed
Moore’s law
• Unprecedented growth in concurrent users and
transactions
• Rapid adoption of flash in consumer electronics
drove up flash density and drove down flash price

• Disk drives increased capacity at the expense of


performance/capacity
This demands an revolutionary solution,
the next generation IBM FlashSystem
10 © 2015 IBM Corporation
Verizon Networkfleet
Fleet telematics provider tracks extreme performance

6x faster
performance for fleet queries

15x reduction
in IOPS latency, from 15ms or
greater to less than 1ms

80 percent Business challenge: Online GPS tracking solution by Verizon required innovative
answers for problems of latency, queries, and reporting, but needed to find clarity
reduction in power, 50% in
among flash alternatives such as hybrid arrays and inline accelerators.
space to facilitate TCO savings
The solution: Winning against other flash alternatives, two IBM FlashSystem arrays
Solution Components were deployed to support fleet telematics such as location, where they are going, what
• IBM FlashSystem they are doing, and monitoring operational health of vehicles.
• Oracle
Business impact: More efficient and cost effective monitoring of fleet vehicles, e.g.
GPS location, vehicle activity, engine diagnostics, miles per gallon and emissions,
improper driving, vehicles outside of performance spec, all generating millions of daily
messages and alerts and monthly reports.
11 © 2015 IBM Corporation
FlashSystem to the Rescue
Actual client experience moving an application to IBM FlashSystem 840.

Can You Guess Which Day the Switch was Made?

12 © 2015 IBM Corporation


IBM FlashSystem Crushes the Competition
Random Reads/Writes; IOPS vs. Response Time
Random 4K Read/Write
5.00
Response Time (ms)

4.00 100% Read

90% Read
Brand V
3.00
70% Read
2.00
50% Read Throughput vs. Response Time:
1.00
0% Read Comparing side-by-side on the same
0.00 scale highlights FlashSystem’s vastly
0 200,000 400,000 600,000 800,000 1,000,000 1,200,000
superior performance for all mixed
5.00 workloads.
Response Time (ms)

4.00

3.00

2.00
FlashSystem
1.00

0.00
0 200,000 400,000 600,000 800,000 1,000,000 1,200,000
IOPS

13 © 2015 IBM Corporation


Why is Latency So Important?

Consider Little’s Law of mathematical queue theory as it applies to Application Performance

Q = the number of parallel threads running in the


application
t = the Time it takes for an IO request to be serviced
R = Result, typically measured in IOPS or Bandwidth
Let’s Assign Values to this equation:
Now let’s see how FlashSystem alters this equation

That’s a 50X improvement in response time and amount of work completed!

© 2013 IBM Corporation Title of presentation goes here


14 © 2015 IBM Corporation
Solution
The TMS/IBM team started by analyzing ABB’s Oracle AWR report. The results were incredible:

The TMS/IBM team was able to conclude that ABB could decrease wait time from 70.84% to 2.78%.
That’s a 213.10% INCREASE in performance with the addition of the RamSan product.

250.00%

200.00%

150.00%

100.00%

50.00%

0.00%
1
Current wai t ti me total 70.84%
New proj ect wai t ti me 2.78%
Total Improvement 213.10%

© 2013 IBM Corporation Title of presentation goes here


15 © 2015 IBM Corporation
RamSans Added
250+ meters Between
Data Hall 301 Data Halls 301 & 001 Data Hall 001
Pod1 Pod2 Region1
(backs up to ATL_A) (backs up to ATL_B)
SAN40B Switch
Red Fabric
AIX-Level LUN Mirroring Supported

DS8700_01 with RamSan 810 SSD RamSan 810 SSD DS8700_02 with
Expansion Frame Expansion Frame
2 sets of 4 redundantly linked

Between ATL Regions


Dual-fabric SVC Nodes
ATL_A TS3500 with

40+ meters
Mirrored 3 Expansion Frames / 18 Drives
ATLs Pod1 active set; Pod2 copy set
Region2
SAN768B Director
Red Fabric 2 pairs of redundantly linked SAN768B Director
6509E core Ethernet Switches Blue Fabric

ATL_B TS3500 with


3 Expansion Frames / 18 Drives
Pod2 active set; Pod1 copy set
HACMP Clustering
p520 R6 TSM Server p520 R7 TSM Server SAN40B Switch
(R6TSM010, 030, 050 ) (R7TSM020, 040, 060) Blue Fabric
7 meters Note: Not all SAN connections shown - devices labeled in purple are attached to both SAN
Between Pods Fabrics. ATL drives are directly attached, each to a single fabric; Odd # drives to Red;
© 2013 IBM Corporation Even # drives to blue
ABB/IBM Confidential Title of presentation goes here
16 © 2015 IBM Corporation
IBM FlashSystem – Flash Realized

Drive a complete paradigm shift in Enterprise Storage with the all new
IBM FlashSystem Family
• Experience real-time analytical insights with up to
50x better performance than enterprise disk
systems with IBM FlashCore™ technology
• Future-proof infrastructure while scaling to over
2 petabytes of effective all-flash capacity under a
single integrated interface
• Deliver agile integration and superior data
economics with 4x greater capacity in less rack
space than competitive all-flash products

17 © 2015 IBM Corporation


Introducing
IBM FlashCore™

Enabling the use of


Flash in the Enterprise

18 ©© 2015
2015 IBM
IBM Corporation
Corporation
Introducing IBM FlashCore™
IBM uses tools and technologies to drive a robust
solution
Industry standard interface

Hardware accelerated I/O

IBM MicroLatency™ module

Advanced flash management


Industry standard chip technology

19 © 2015 IBM Corporation


FlashCore: IBM MicroLatency™ Module
• IBM Engineered

• Massively Parallel Design

• FPGAs in the Data Path

• Distributed RAM

• High Speed Interface

• Hardware-based Data
at Rest Encryption

20 © 2015 IBM Corporation


IBM MicroLatency™ Module vs SSD
IBM MicroLatency™ Module SSD

CUSTOM FAST RELIABLE COMMODITY SLOW FALLIBLE


Requires hardware Designed for Innovative flash No hardware Designed for More single
engineering low latency management skills needed low cost points of failure

Better Performance Better Reliability Better Density Lower $/IOPS

21 © 2015 IBM Corporation


IBM MicroLatency™ Module Key Differentiators
• Innovative ECC algorithm corrects very high bit error rates
(does not require additional reads)
• Health binning and heat segregation
– FlashCore technology continually assesses
health by monitoring several key parameters
– Hotter data placed in healthier blocks –
Up to 57% improvement in endurance!
– Write Amplification reduced by up to
Results:
45% by grouping like heat levels
• Large endurance gains w/o latency penalty
• The IBM and Micron collaboration provides special access
• Asymmetrical wear leveling results in higher
to the Flash
endurance in real life workloads!
– Extensive characterization determined optimal
read offsets to set as block ages • As the part ages and we determine the health
of each block, the ideal voltage levels are
• Powerful checks on entire data path, including inside set in order to minimize errors!
the flash modules
• Outstanding data reliability
• Variable Stripe RAID reduces what needs to be retired and improved endurance

22 © 2015 IBM Corporation


FlashCore: Advanced Flash Management
IBM Advanced Flash Management improves MLC flash endurance 9x
over standard implementations without sacrificing latency.

Backed by IBM’s Flash Wear Guarantee!


“Through close collaboration between IBM research, development and flash engineering, IBM created an
advanced flash characterization platform to test thousands of flash devices over a lifetime of wear. As a
result, IBM has made major strides in advanced flash management.” – Andy Walls, IBM Fellow

Key accomplishments:
Standard MLC Flash • Innovative ECC algorithm corrects very high bit error rates
IBM Enhanced MLC
• Variable voltage and read level shifting maximizes flash endurance
• Advanced Flash Management continually monitors the health of blocks
9.4X Endurance Improvement and performs asymmetrical wear leveling and sub-chip tiering:
- Health binning: up to 57% improvement in endurance!
- Heat segregation: up to 45% reduction in write amplification!
• Micron Technology provides special access to the chip

23 © 2015 IBM Corporation


FlashSystem 900

24 ©© 2015
2015 IBM
IBM Corporation
Corporation
FlashSystem data protection schema
Improving Availability and Serviceability
• Two-dimensional Flash RAID • No single point of failure
– IBM Variable Stripe RAIDTM
• Higher resilience • Fully redundant and hot-swappable
• Higher storage efficiency components
• Higher rebuild performance – Flash modules
• Lower set uptime
– I/O interfaces
– System-level RAID 5
– RAID controllers
• Non-disruptive (concurrent) code load – Management controllers
– Power supplies
• Data path is entirely independent of the – Batteries
control path – Fan modules

• Management features
– Call home
– Detailed logging
– Remote power on

25 © 2015 IBM Corporation


IBM Variable Stripe RAID™ (VSR)
Form Factor SSD
• Flash failure = Disk failure
• Requires top-level RAID
• Relatively frequent hot-swaps
No Parity

Enterprise Flash Drive or Memory Module


• Flash failure = Degraded state within
module
Value of Variable • Performance impact on RAID set
Parity
Stripe RAID • Hot-swap to resolve

Less maintenance touches FlashSystem with Variable Stripe RAID


while still preserving the life, • Preserves Flash life
protection, and performance • Preserves performance
of the Day-1 experience Parity • Re-parity data in microseconds

26 © 2015 IBM Corporation


Two-dimensional RAID
• IBM Variable Stripe RAID is 1st level Canister-1 MC-1 MC-2 Canister-2
of protection (module-level)
FC-1 FC-2 FC-3 FC-4
• System-level RAID protects against
flash controller failure (abrupt
module failure) XBAR 0 XBAR 1
• RAID 5 across modules, hot
swappable from front of chassis
• Two independent crossbar (XBAR)
controllers
• Interface controllers stripe across
XBARs
• Granular capacities configurable
with 2, 4, 6, 8, 10 or 12 flash
modules
27 © 2015 IBM Corporation
IBM FlashSystem 900 components
IBM MicroLatency Modules (12)
Battery Packs (2)

RAID Controllers (2)

Interface Modules (4)

Fan Packs (4) Canisters (2)

Management Modules (2)

Power Supplies (2)

28 © 2015 IBM Corporation


Introducing IBM’s premier all-flash offering
IBM FlashSystem 900
the next generation in our lowest latency offering
• IBM MicroLatency™ with up to 1.1 million IOPS
• 40% greater capacity compared to previous generation
• IBM FlashCore™ technology, our secret sauce
Technical collaboration with Micron Technology – flash chips
• IBM enhanced flash technology
Performance at-a-glance
• MLC NAND flash offering and Flash Wear Guarantee
Minimum latency
Improved integration with VMware VASA and VAAI UNMAP, Write 90µs
OpenStack Cinder Read 155µs
Maximum IOPS 4 KB
Read (100%, random) 1,100,000
Read/write (70%/30%, random) 800,000
Module type 1.2TB 2.9TB 5.7TB Write (100%, random) 600,000
Maximum bandwidth 256 KB
Modules 4 6 8 10 12 6 8 10 12 6 8 10 12
Read (100%, sequential) 10GB/s
RAID 5 capacity 2.2 4.5 6.8 9.1 11.4 11.4 17.1 22.8 28.5 22.8 34.2 45.6 57.0 Write (100%, sequential) 4.5GB/s

29 © 2015 IBM Corporation


Where faster I/O will make a difference
RTO after
failure or
HANA
patching
Response time
Load time
of on-demand
dependent on
columns
throughput of dependent on
IOPS latency,
throughput of
IOPS
Part Shipped Item Item Package
Client Order date Delivery date Quantity Total price Instructions Status
number date Weight cost weight

Conventional DB Delta Table in HANA


To log file and in-memory row
abc 123 8/1/2015 9/6/2015 9/15/2015 15 0.56 1.35 9.5 $42.00 Leave on porch In-transit

Single Async IOP Delta merge


abc 123 8/1/2015 9/6/2015 9/15/2015 15 0.56 1.35 9.5 $42.00 Leave on porch In-transit

Async IOPS = number of columns


(at minimum)
abc 9/15/2015 9.5
123 15 $42.00
8/1/2015 0.56 Leave on porch

9/6/2015 1.35 In-transit

30 © 2015 IBM Corporation


IBM’s premier, fully-integrated, tier 1, all-flash offering

31 © 2015 IBM Corporation


What’s new with FlashSystem V9000
Announce: August 18, 2015
IBM FlashSystem V9000 General Availability: August 31, 2015

IBM introduces a fully integrated, flexible, feature rich all-flash storage system

Scalable Performance Enduring Economics Agile Integration

• Scalable all-flash architecture accelerates applications and


entire infrastructures

• FlashSystem software V7.5, including


o HyperSwap®
o FlashCopy® bitmap space increases
o Direct 16Gbps Fibre Channel support (except for AIX®)
Powered by o Microsoft Offloaded Data Transfer (ODX) support
IBM FlashCore™ Technology
• Full 8x8 scaling support
32 © 2015 IBM Corporation
Scalable
Performance

FlashSystem V9000 options for deployment


Scalable Performance
Small Data Center Application Accelerator

• Great for database and VDI • Great for large data sets with
workloads big I/O requirements and
• Accelerate, migrate, tier, needing storage services
clone, snapshot, replicate, Public or Private Cloud • Scale up for more all flash
compress existing storage capacity
• Great for tier 1 disk
• Up to 630K IOPS, 200µs • Up to 630K IOPS, 200µs
replacement
• Up to 57TB usable, 285TB • Up to 285TB usable, 1.4PB
• Scale up and out for
effective Effective
more processing, more
capacity and more I/O
Mixed Workload Accelerator • Up to 2.5M IOPS, Virtualized Data Center
200µs • Great for data centers with
• Great for multiple mixed • Up to 456TB usable, heterogeneous storage
workloads that drive huge I/O 2.2PB Effective • Extends core feature set to
• Scale out for more all flash
other storage arrays
capacity, IOPS and bandwidth
• Up to 2.5M IOPS, 200µs
• Up to 2.5M IOPS, 200µs
• Up to 3 Exabytes virtualized
• Up to 228TB usable, 1.1PB
Effective

33 © 2015 IBM Corporation


Mixed workload consolidation Agile
Integration

Agile Integration

Transaction Virtualization
Qualities needed for
Big Data
mixed workloads:
ERP / SCM
Cloud • Capacity scaling
• Performance scaling
• Automatic tiering
• Quality of service
• Strong mixed R/W
performance

34 © 2015 IBM Corporation


IBM XIV
XIV System
FC
Host XIV Module
SSD
Host Data
Modules
iSCSI XIV Module
SSD

Interconnect
Host Interface
Modules
XIV Module
Full Rack – up to 485TB SSD
• 15 modules Interface
• 90 cores (180 logical
cores w/ hyper-threads) XIV Module Modules
• 180 2/3/4/6 TB SAS disks SSD
• 720 GB memory
Data
Modules
• 24 8GB FC ports
• 12x10GbE / 22x1GbE
iSCSI ports UPS UPS UPS
• 12TB SSD cache (opt)

36 © 2015 IBM Corporation


Data distribution

• COMPLETELY Automated • Zero Tuning


Process • LINEAR Scalability
• NO Pre-Planning • NO Hot-Spots
• AUTOMATIC Load Balancing • MAXIMUM capacity utilization
• EVEN Disk utilization • WORLD-RECORD disk rebuilds

37 © 2015 IBM Corporation


Leveraging the power of the XIV grid
• XIV Gen3 can have up to 15 RACE engines, depending on the configuration. These
RACE engines compress and decompress data in parallel

In XIV, Real-time Compression is distributed across a grid.


In a 15-module XIV system, 180 CPU cores are processing every
volume’s I/O in parallel (including Real-time Compression)

38 © 2015 IBM Corporation


IBM FlashSystem A9000
FlashSystem A9000 FlashSystem A9000R
(Rack)
• 8RU, rack-less offering  Integrated rack offering
• 60TB – 145TB effective capacity  Scale-out grid architecture
• 450K IOPS (70/30, 4K random)  290TB – 1.7PB effective capacity
 1.2M – 3.6M IOPS (70/30, 4K random)

40 © 2015 IBM Corporation


Software stack capability highlights
• Data services • New
– Snapshots – Data Reduction: DeDup and Compression
– Sync/Async replication – Cross platform management GUI
– Thin-Provisioning
– Hyper-Scale features
• Integrations
– VMWare
– Encryption
– Microsoft (Hyper-V)
– Multi-tenancy
– OpenStack
– Data Migration
– QoS per host, domain, pool
and volume

41 © 2015 IBM Corporation


Architectural building blocks
• Scalable grid architecture
Interface Module
• Effective capacity scalable to 1,710TB  System x3650 M5
with 5:1 data reduction  2U 2S x86 Server
 4x16Gb FC + 2x10Gb iSCSI or 4x10Gb iSCSI
• Fully redundant system  384GB DRAM
• Redesigned power distribution and  XIV Flash-Optimized SW stack
protection

IBM FlashSystem 900 Infiniband interconnect


 2D RAID5  FDR 56Gbps
 12 x 1.2TB or 2.9TB or 5.7TB

42 © 2015 IBM Corporation


FlashSystem A9000
•8U rack-less offering • 12TB, 29TB or 57TB Storage Enclosure

• 2U 2S x86 Server (4x16Gb FC + 2x10Gb

•Storage enclosure iSCSI or 4x10Gb iSCSI)


• 8 cores, 196GB DRAM

12TB Storage 29TB Storage 57TB Storage


Enclosure Enclosure Enclosure
(Useable / Effective 5:1) (Useable / Effective 5:1) (Useable / Effective 5:1)

12 / 60TB 29 / 145TB 57/ 285TB

• 3 x M5 controllers
• 450K IOPS with data reduction (2)
• Manage through IBM Hyper-Scale Manager
(1) 5:1 Data Reduction Ratio
• Scalability with Hyper-Scale (2) 70% R / 30%W 4K random, compression & deduplication

43 © 2015 IBM Corporation


FlashSystem A9000R scale-out
• Mellanox Neptune based system
 The entry configuration: • 36 port Infiniband switch – FDR
56Gbps
- Switch/PDU/Mgmt. Module,
• Internal redundant BBU solution
- 4 x Interface Modules and 2 x FS900 Flash
enclosures • Management Module

 Rack can be upgraded from 2 to 6 scale-out


• Custom design 20KW PDU
building blocks (MES groups)
• Cascaded control connectors
Number of 29TB Flash Media 57TB Flash Media
building blocks modules modules
(Useable / Effective 5:1) (Useable / Effective 5:1)
2 58 / 290 TB 114 / 570 TB • 29TB or 57TB FlashSystem 900
Enclosure
3 87 / 435 TB 171 / 855 TB

4 116 / 580 TB 228 / 1,140 TB • 2U 2S x86 Server


• 4x16Gb FC + 2x10Gb iSCSI or
5 145 / 725 TB 285 / 1,425 TB 4x10Gb iSCSI
• 384GB DRAM
6 174 / 870 TB 342 / 1,710 TB

44 © 2015 IBM Corporation


System architecture FS900
FC Interface
Host FS900 Grid Controller
Module

Host
Interface FS900
iSCSI Module

Host Interface FS900

IB Interconnect
Module
IB Interconnect

Interface FS900 Power Distribution


Module
Full system
• 1.7 PB effective capacity with
5:1 data reduction Interface
(342 TB useable) FS900
Module
• 4.6 TB memory
• 48 x 16GB FC ports +
24 10GbE iSCSI Interface FS900
• 48 x 10GbE iSCSI ports Module
• Dual 56Gbps FDR IB network

45 © 2015 IBM Corporation


Data distribution in a grid-architecture
• Data is spread across
all the Interface Modules
and Flash Enclosures
Oracle
• Partitions are protected
on the Flash Enclosure
• No manual intervention
VMware
• No Tuning

• Completely automated
• No Pre-Planning ERP
• Automatic load balancing
• Zero tuning
• Linear scalability

46 © 2015 IBM Corporation


Bank in India
Optimizing resources while improving performance

Reduction
of report run times

Highly efficient
copy services for core banking
solution

300TB Business challenge: A bank in India desired to evaluate and implement new
of effective capacity utilizing technologies offering significant reduction in data center requirements without
data reduction techniques compromising performance and function.

The solution: IBM FlashSystem A9000 was chosen over competition to support core
Solution Components
• IBM FlashSystem A9000
banking applications.
• Competition: EMC, NetApp
Business impact: Substantial expected benefits in performance, report run time and
copy services will enable the client to optimize data center resources while
improving service levels and business operations.
© Copyright IBM Corporation 2017 IBM & IBM Business Partner Use Only 47
47 © 2015 IBM Corporation
Where to use Flash
Typical use cases for IBM FlashSystem: SAP Business Suite

4. Flexible deployment architectures for any SAP Business Suite workload


• Use SVC stretched cluster for HA purposes
• Gain advantage of FlashSystem write performance
• Simple integration and installation
• Capacity enhanced by using traditional storage
IBM Power • Enhance read and write performance of traditional storage by
using EasyTier (FlashSystem can be utilized as high
performance tier)
• Cost savings by using heterogenous storage devices
• Full flexibility to shift workload between different storage classes
Preferred read
IBM SAN Volume on local site
Controller IBM SAN Volume
EasyTier Controller
switched on EasyTier
switched on
SCV stretched cluster
logs only Short disctance (10 km) to avoid
SVC cache disabled! latency

IBM FlashSystem 900


IBM Storwize or XIV IBM FlashSystem 900
IBM Storwize or XIV

49 © 2015 IBM Corporation


Where faster I/O will make a difference
RTO after
failure or
HANA
patching
Response time
Load time
of on-demand
dependent on
columns
throughput of dependent on
IOPS latency,
throughput of
IOPS
Part Shipped Item Item Package
Client Order date Delivery date Quantity Total price Instructions Status
number date Weight cost weight

Conventional DB Delta Table in HANA


To log file and in-memory row
abc 123 8/1/2015 9/6/2015 9/15/2015 15 0.56 1.35 9.5 $42.00 Leave on porch In-transit

Single Async IOP Delta merge


abc 123 8/1/2015 9/6/2015 9/15/2015 15 0.56 1.35 9.5 $42.00 Leave on porch In-transit

Async IOPS = number of columns


(at minimum)
abc 9/15/2015 9.5
123 15 $42.00
8/1/2015 0.56 Leave on porch

9/6/2015 1.35 In-transit

50 © 2015 IBM Corporation


SVC/FlashSystem Integration – Architecture
DB Servers
Boost Performance
Boost Redundancy
Write IO to both mirrors - Without Disruption
- Without Risk
SAN Volume
Read IO from FlashSystem for Controller - Without Feature Loss
µsec response times SAN
SAN

READS
READS

Vdisk Mirror 2 WRITES


Vdisk Mirror 1
IBM ACTIVE DATA
WRITES
FlashSystem
ACTIVE DATA

20 TB
20 TB

TRANSITIONAL DATA TRANSITIONAL DATA


Mirror
5 TB 5 TB

51 © 2015 IBM Corporation


Operation and financial impact: Tipping Point POC
Results Hardware Configuration
Normalized $ / IOPS
Performance Results
• 1.3 Million IOPS 11x Less
• 43K+ Transactions per second
• 13K Updates per second
IBM 2,500Spindles 5,000
FlashSystem + 128 SSDs Spindles

Energy Space

37x Less 80x Less

IBM 2,500Spindles 5,000 IBM 2,500Spindles 5,000


FlashSystem + 128 SSDs Spindles FlashSystem + 128 SSDs Spindles

52 © 2015 IBM Corporation


All-Flash is About Economics
“spectacular”; processing T5TB in 3.5 inches of rack space vs.1,300 T75% less rack space,
huge number of transactions in disk for 400K IOPs, 90% less power, 83% faster
one day, lower response timesT less than 1/10th the costT data compressionT
Core Financial Transactions Cloud Storage Source: IBM Client Experiences

$7.1M All FlashP


 38% Lower Software License Costs
Over 30% Less – Fewer cores, lower maintenance
Expensive Systems – Database, infrastructure SWT
 Higher Storage Utilization
$4.9M – As much as 50%
– Lower maintenance
– Simplified management
$4,997,000  17% Fewer Servers
– Fewer cores
$3,197,000 – Fewer network connections
– Lower maintenance
$420,000  74% Lower Environmental Costs
$296,000 $638,000 – Power / cooling, floor space
$377,000
$314,000  35% Lower Operational Support Costs
$1,113,000 $723,000 – Server / storage administration

Disk Flash Source: Wikibon, March 2013


© 2013 IBM Corporation Title of presentation goes here
53 © 2015 IBM Corporation
Cegeka Case Study Video

Enabling lightning-fast performance to lead enterprises to the cloud

66 percent
reduction in SAP ERP
application response times

97 percent
cut in footprint requirement

95 percent Business challenge: To broaden its cloud-services client base, Cegeka targeted
decrease in power enterprises with demanding SLA requirements. Could it meet data performance
consumption and scalability needs without costs spiraling out of control?

The solution: Cegeka gained the capacity to support enterprise clients in the
cloud, helping to meet corporate growth objectives, while continuing to attract and
Solution Components retain business from existing customers.
• SAP ERP
• IBM FlashSystem
“IBM FlashSystem Enterprise Solution accelerates some SAP ERP
• IBM SAN Volume Controller
• Oracle database ”
processing jobs by up to 90 percent.”
• VMware
—Luc Greefs, Director of Cloud Services, Cegeka
54 © 2015 IBM Corporation
Coca-Cola Bottling Co. Consolidated CRDB Case Study Video

Delivering deeper insight into customer demand four times faster


97 percent
reduction in processing time
without replacing a single server

20x more
data crunched within the existing
window and SLAs
Business challenge: CCBCC needed to process more data to accurately
150 million anticipate customer demand without increasing time-to-insight or exceeding SLAs.
cases of product per week
The solution: CCBCC now processes 20 times more forecasting data, down to
the store level, which enables them to match manufacturing output with demand,
Solution Components enables earlier logistics planning and increases profitability.
• SAP ERP
• JDA Manugistics “By using IBM FlashSystem to accelerate our insights into customer
• IBM FlashSystem demand, we’re better placed than ever before to offer unbeatable levels
• IBM SAN Volume Controller
of service to our customers across the United States.”
- Tom DeJuneas, Infrastructure Manager, CCBCC
55 © 2015 IBM Corporation
Major home improvement retailer
Building competitive advantage with flash analytics
12x faster
business analytics performance

90x faster
than tier 1 EMC VMAX storage

85 percent
reduction in batch process time Business challenge: A major home improvement retailer needed to quickly
download information from multiple production databases and perform analytics.
The existing storage took 6 to 8 hours to complete, impacting end user access.
Solution Components The solution: The company moved mission critical applications off EMC VMAX on
• SAS to IBM FlashSystem and reduced key batch processes from hours to minutes.
• SPLUNK
• IBM FlashSystem 840, 820
• IBM SAN Volume Controller “IBM FlashSystem performance and management is incredible.
• IBM Storwize V7000 We will be implementing a lot more IBM Flash this year.”
• VMware VDI - Senior Director, Enterprise Storage

56 © 2015 IBM Corporation


Brokerage Workload
IOPS and Response time in the cases of 6,000 users and
50,000 customers Oracle Approved EXTERNAL White Paper

Increasing the workload from IOPS Latency us


300000 10000
6,000 users to 50,000 users Oracle 11g R2, AIX 7.1 Power
9000
yielded a 229% increase in 250000
S824, FlashSystem 840
225,726 8000
IOPS, while average IO
7000
latency increased by only 200000
6000
10us
150000 5000

4000
98,551
100000
3000

2000
50000
380µs 390µs 1000

0 0
6,000 Customers 50,000 Customers
6000Number of Customers 50000
This is an IBM internal study designed to replicate a typical IBM customer workload usage in the marketplace. The results were obtained under laboratory conditions, and not in an actual customer environment. IBM's internal workload studies are not benchmark applications, nor are they based on
any benchmark standard. As such, customer applications, differences in the stack deployed, and other systems variations or testing conditions may produce different results and may vary based on actual configuration, applications, specific queries and other variables in a production environment.

57 © 2015 IBM Corporation


AIX/Power & FlashSystem Deliver Amazing Analytics Result
Rapid Discovery of Opportunities & Risk
IBM InfoSphere Identity Insight
• Analytics to help organizations understand and identify data about
individuals and groups from multiple sources and formats
• Provides actionable information to quickly discover opportunities or risk in
areas such as customer service, fraud detection and threat avoidance
• The need to process massive amounts of small data fragments results in
very high IOPS
• Spinning disk struggles to keep pace with the performance requirements
of analytics
Higher Performance = Faster Insight
• Faster query processing yields quicker results
• FlashSystem technology provides greater I/O density and much lower
latency than disk
• For best performance IBM InfoSphere Identity Insight recommends
AIX/Power, FlashSystem technology, and DB2 Power 780+ 64c, Flash Adapter 90, FlashSystem 840 40TB

58 © 2015 IBM Corporation


Understanding Storage in a Healthcare Environment: Why
it matters
Why FlashSystem storage matters
Doctors get the data they need quickly
without waiting for storage to address their
request. IT personnel get peace of mind
1,536 HDD 12 Flash
knowing the patient data is stored in a safe,
69,000 IOPS Modules
high performing solution that delivers the 85,000 IOPS
4.3
data when needed at a lower cost. milliseconds 0.39
milliseconds

Epic Healthcare Epic is a set of software that has changed the way
40% lower healthcare interacts with patients.
90% lower random read latency than SSDs Epic covers all aspects of the healthcare organization which directly affect patients,
doctors, and operations. This presents challenges to the infrastructure.
random read latency than high-end disk The ability of EPIC to do what it was designed to do is what directly affects doctors
and patients. Due to this, solutions have to meet minimum requirements when
30X below deployed in an EPIC environment.
recommended performance requirements
Storage ensures that Epic can function efficiently and
70% compression effectively.
rate achieved The processing ability of IBM POWER8 coupled with the enterprise capabilities and
performance of IBM® FlashSystem™ V840 surpassed EPIC needs at a lowest cost*
Learn more at http://www-03.ibm.com/systems/storage/flash/ecosystem/isv.html

© 2013 IBM Corporation Title of presentation goes here


59 © 2015 IBM Corporation
© 2014 IBM Corporation
IBM FlashSystem® provides performance boost
Helps improve patient healthcare experience

Healthcare consumers often accept care without


knowing the financial implications of their decision.
TriZetto strives to change that through their proven
claims administration systems and require high-
performing infrastructure to capture and reliably deliver
healthcare information seamlessly, with no unplanned
downtime.
12% improvement 90% OPEX reduction
in application performance compared to disk
Replacing the disk storage solution with IBM FlashSystem provided
90% reduction TriZetto tangible performance improvement in its payer software. The
in storage latency up to 5X improvement reduction in batch cycles and improvement in online response in
application and DB maintenance operations TriZetto payer applications seen with FlashSystem are vital.
Hardware
• IBM FlashSystem storage
• Flex System™ server With IBM FlashSystem storage, TriZetto can process claims faster,
with IBM POWER7+™,
2u of which can improve the patient experience, help customers make more
IBM AIX® 20u of
Software Flash, informed healthcare decisions and reduce healthcare costs for
• TriZetto QNXT application Disk, vs
w/Microsoft SQL Server 24 TB insurers and members.
• TriZetto Facets 70TB
application w/Sybase You can read more this solution in our whitepaper.
ASE 15.7

60 © 2015 IBM Corporation


IBM storage achieves record-breaking performance on FIS benchmark

IST/Switch is FIS’ open platform software for high-volume, high-availability,


and highly-scalable EFT processing environments. It is used to run billions of
EFT transactions each year for leading financial institutions, networks,
retailers and third-party processors.
How did FlashSystem perform?
FlashSystem produced the performance needed to service current and future
growth. It allowed IST/Switch to scale far past prior limits, smashed the
previous benchmark, and did so failure free.
“Using IBM Storage and Power 10,000 transactions per
Systems servers, we are able to second
scale our FIS IST/Switch application
far beyond previous limits and meet for over 30 minutes with 0
errors
the demanding requirements of our 3.6X Improvement
financial services clients. These Over previous tests of the
results take payment processing to
A Smarter, Faster Solution a new level of performance for our IST/Switch benchmark
IBM FlashSystem, FIS IST/Switch application, clients,” says Andy Roe, Director,
IBM Power Systems, IBM AIX, IBM DS8800 Strategic Development, FIS.

You can read more about this solution in our case study.

61 © 2015 IBM Corporation


IBM FlashSystem improves DB2
workloads IBM® Power Systems™ provide the processing capabilities
How can the growing data volumes between
needed to scale out or scale up while powering the most
different processing clusters, departments and
complex analytics.
sites be shared most effectively? How can the
explosion of unstructured data be managed? IBM FlashSystem® storage offers performance, efficiency and
reliability, as well as the virtualization tools that enable
enterprises to turn the explosion of data into insight for
business advantage.

These IT infrastructure components provide the foundation


upon which to build a next generation data platform—using
IBM DB2®—that keeps transactional workflows and analytics
operating at maximum efficiency.
Brokerage Workload Transactional Workload Big data insights analytics workload

40X better TPS 190X better TPM 2.5X performance improvement


performance performance for complex queries
40X better IOPS 86X more IOPS
7X performance improvement
95% price-per-performance
90% price-per-performance for intermediate queries
improvement
improvement

62 © 2015 IBM Corporation


IBM Data Engine for
NoSQL - Power Systems Edition

IBM INNOVATION

63 © 2015 IBM Corporation


IBM Data Engine for NoSQL - Power Systems Edition
The Solution:
• Integrated platform for large and fast growing NoSQL
data stores which utilizes high speed access to both
RAM and flash storage for lower cost and higher
workload density than a standard RAM based system
– with equal or better performance than scale out
• Can be offered as a cloud-based service or as an on
premise solution for enterprises 24:1 3x cost
consolidation3 savings

The Value:
• Lower Total Cost of Ownership for NoSQL
infrastructure with high levels of consolidation and no
performance impact or degradation
Up to 40TB of extended memory with one POWER8 server + CAPI attach FLASH

© 2016
64 IBM Corporation
© 2015 IBM Corporation IBM FlashSystem | 64
Introducing game-changing innovation that
accelerates big data & analytics
Smart Acceleration enabled by CAPI (Coherence Accelerator
Processor Interface) Technology

• Smart, simplified attach for accelerators:


GPUs, flash memory, networking & FPGAs

• Connects directly to processor, sharing the


same address space

• Improves performance, reduces latency, and


provides more workload for your dollar
CAPI enables innovation from the OpenPOWER Foundation

© 2016
65 IBM Corporation
© 2015 IBM Corporation IBM FlashSystem | 65
CAPI attached IBM FlashSystem optimization
• Attach IBM FlashSystem to POWER8 via CAPI
• Issues Read/Write commands from applications to eliminate 97% of code path length
• Saves 20-30 cores per 1M IOPs

Application Application

Read/Write Syscall 20K Posix Async aio_read()


instructions I/O Style API aio_write()1
FileSystem reduced to User Library
<500
strategy ( ) iodone ( )
LVM
Shared
strategy ( ) iodone ( ) Memory Work
Queue
Disk and Adapter DD
Pin buffers, Interrupt, unmap,
Translate, Map unpin,Iodone
DMA, Start I/O scheduling

© 2016
66 IBM Corporation
© 2015 IBM Corporation IBM FlashSystem | 66
Innovative “In-Memory” NoSQL/KVS integrated solution
4Q14 - Differentiated NoSQL
Today’s NoSQL in memory (x86) (POWER8 CAPI + FlashSystem)

WWW
10Gb Uplink WWW
10Gb Uplink
Load Balancer
24U POWER8 Server
500GB Cache 4U
500GB Cache
Node
500GB Cache Flash Array w/ up
Node
500GB Cache
Node
500GB Cache to 40TB
Node
Node

Backup Nodes

Target LoB / solution architects and MSPs IBM Data Engine for
NoSQL - Power Systems Edition:
• Supporting or building mobile/web/social apps
• Leveraging Key Value Store (KVS) for fast lookups • 24:1 physical server consolidation
• Require high performance in-memory data access • 6x less rack space (2U server+2U
FlashSystem vs. 24 1U servers)
© 2016
67 IBM Corporation
© 2015 IBM Corporation IBM FlashSystem | 67
Demonstrating the value of CAPI attachment
Identical hardware with 2 different
120000 500
paths to data 108,438 466
450
100000
FlashSystem 840 400

350
80000
300

Conventional 60000 250

CAPI 199
PCIe I/O 200
40000
150

16,150 100
20000
50

0 0
PCIe I/O CAPI PCIe I/O CAPI

IOPS per HW Thread Latency (µs)


POWER S822
© 2016
68 IBM Corporation
© 2015 IBM Corporation IBM FlashSystem | 68
Summary and Next Steps
• FlashStorage has traditionally be an application acceleration tool allowing CPU resources to be better
utilized – faster applications!
• Today, FlashStorage can be used for data center modernization because “Flash is cheaper than disk”
• Tying FlashStorage with Power provides best in class on servers AND storage
• Which application causes you the most headache from a performance standpoint? Let’s look at it together
and see what Flash could do.
• Check out the FlashSystem website: http://www-03.ibm.com/systems/storage/flash/

69 © 2015 IBM Corporation


70 © 2015 IBM Corporation
Manufacturing firm
Multinational company validates innovative IBM FlashSystem A9000R

930 percent
better index range scan Reads

142 percent
better parallel load of Reads

50 percent
better rebuild Writes Business challenge: Large company in Europe with very large databases and high-
*all compared to disk-based system with SSD performance imperatives realized traditional single class shared storage would not
acceleration
accommodate future business needs.

Solution Components The solution: IBM FlashSystem A9000R was tested in a successful Proof of
• IBM FlashSystem A9000R Concept reflecting daily business service requirements for the SAP environment.
• SAP
• Oracle
Business impact: Significant overall performance improvements and excellent
expectations for data reduction (compression & deduplication) provided the results
needed to conclude the A9000R fits the storage service requests requirement.
© Copyright IBM Corporation 2017 IBM & IBM Business Partner Use Only 71
71 © 2015 IBM Corporation
Financial services provider
IBM FlashSystem and Spectrum software combine to add great value

Flash-only
solution with optimal
performance for all workloads

TCO costs
dramatically reduced by less
power, energy, rack space
Advanced features Business challenge: Banking services provider in Europe needed to implement a
highly available storage architecture and DR site to comply with regulatory rules and
include virtualization, accommodate 30% year-to-year growth.
compression, deduplication
The solution: IBM FlashSystem A9000 pods were geographically dispersed on 4
Solution Components sites with IBM SmartCloud VSC for management of entire solution.
• IBM FlashSystem A9000
• IBM SmartCloud Virtual Storage Center Business impact: Substantial improvements in high availability, disaster recovery,
• IBM Spectrum Virtualize & Spectrum Scale performance, power and space costs, and flexibility have provided the client a
• Competition: HDS, NetApp (SolidFire) consolidated storage infrastructure to address present and future requirements.
© Copyright IBM Corporation 2017 IBM & IBM Business Partner Use Only 72
72 © 2015 IBM Corporation
Investment bank
IBM rises above in financial company’s evaluation of flash solutions

Key benefits
include better performance and
reduced power/cooling/space

Data reduction
provided by deduplication and
compression

Grid architecture Business challenge: Investment bank in Europe looking to move to flash storage
evaluated multiple vendors for the best solution for their requirements.
and unique features to promote
resiliency and performance The solution: IBM FlashSystem A9000R and A9000 installed in two sites –
production and DR – in support of various OS and application environments.
Solution Components
• IBM FlashSystem A9000 & A9000R Business impact: Superior reduction factors, uniqueness and performance of grid-
• Oracle, Sybase based architecture, and enterprise-class characteristics and features compelled the
• Solaris, Linux, VMware client’s evaluation and technical team to choose IBM.
• Competition: Pure Storage, EMC, HDS
© Copyright IBM Corporation 2017 IBM & IBM Business Partner Use Only 73
73 © 2015 IBM Corporation
Hospital and medical system
Large multispecialty health system moves to flash with IBM

Excellent
environment for Cerner VDI,
Oracle, VMware & compression

Simplified
storage management GUI based
on IBM XIV

500 physicians Business challenge: A system of integrated health services located in northwest
USA was seeking a storage system upgrade to flash technology in the midst of a
and 5500 employees new data center implementation and increased requirements for performance.

The solution: IBM FlashSystem A9000R was implemented in support and fulfillment
Solution Components of Oracle and Cerner healthcare applications and a virtualized environment.
• IBM FlashSystem A9000R
• IBM XIV
• Cerner, Oracle, VMware Business impact: Expected improvements in performance, efficiency, flexibility and
• Business Partner: Mainline storage management will provide the client with the storage infrastructure to help
• Competition: Pure Storage continue its mission of offering superior treatment outcomes for patients.
© Copyright IBM Corporation 2017 IBM & IBM Business Partner Use Only 74
74 © 2015 IBM Corporation
Investment funds
Enhancing XIV experience with new IBM FlashSystem A9000R

Excellent
user experience based on IBM
XIV

Data reduction
to help optimize resources in
new datacenter

Faster Business challenge: Global fund manager based in Singapore decided to combine a
move to a new datacenter with an upgrade of storage infrastructure.
performance than previous
storage The solution: IBM FlashSystem A9000R was deemed better than the competition
and will be implemented in support of analytics applications.

Solution Components
• IBM FlashSystem A9000R Business impact: The client will find the ease of use and simplicity of their current
• Competition: EMC, Pure Storage IBM XIV will be matched by the FlashSystem A9000R and enhanced with new
function and capabilities.
© Copyright IBM Corporation 2017 IBM & IBM Business Partner Use Only 75
75 © 2015 IBM Corporation
Online conglomerate
Stellar e-commerce company choses IBM FlashSystem for speed

Sub-millisecond
response times for critical
systems

Peak season
targets successfully met

Overall value Business challenge: Parent company to many e-commerce retail brands based in
and performance were decisive Midwest USA was troubled by lagging response times which threatened business
for the A9000R vs. competition operations during an upcoming holiday shopping season.

The solution: IBM FlashSystem 900 was initially deployed for high-value workloads,
Solution Components
with a subsequent procurement of the IBM FlashSystem A9000R.
• IBM FlashSystem models A9000R, 900
• IBM Spectrum Accelerate software
• IBM XIV Business impact: Excellent benefits included better performance, accelerated
• Competition: Pure Storage, NetApp response times and ease of integration into the existing environment, all of which
gave the client a solid footing for growth and customer satisfaction.
© Copyright IBM Corporation 2017 IBM & IBM Business Partner Use Only 76
76 © 2015 IBM Corporation
Motor transport
Trucking company improves operations with IBM FlashSystem A9000

Excellent
data reduction rates

Single point
of control with management
interface and GUI

Primary storage Business challenge: Prominent long-standing truckload carrier in southeastern USA
A9000 expected to replace was seeking to improve performance and support for its VDI and VMware
spinning disk enterprise-wide virtualized environment.

The solution: (2) IBM FlashSystem A9000 pod systems were deployed with 150TB
Solution Components
of effective capacity each.
• IBM FlashSystem A9000
• VMware
Business impact: Impressive results include improved performance, data reduction
rates and single point of control via the management interface, which have
compelled the client to employ the A9000 as primary storage enterprise-wide.
© Copyright IBM Corporation 2017 IBM & IBM Business Partner Use Only 77
77 © 2015 IBM Corporation
Home furnishings
Retailer moves to IBM FlashSystem A9000 for Oracle workloads

50 percent
reduction in physical footprint

Performance
and response time improved for
Oracle database

Saved costs Business challenge: Furniture and home décor trend setter in Latin America had
with reduction of supply chain several initiatives for IT including better application response time, more efficient
delivery costs supply chain operations, and reduction in hardware footprint.

The solution: After an analysis of competition and traditional storage, the IBM
Solution Components FlashSystem A9000 was chosen to fulfill Oracle applications and other workloads.
• IBM FlashSystem A9000
• Oracle Business impact: Significant benefits included improved performance and response
• Competition: Oracle FS1 Flash Storage
System
time, reduced physical footprint, and more efficient business operations enabling the
• Business Partner: ITSERVICE stores to receive products in less time and with more accuracy.
© Copyright IBM Corporation 2017 IBM & IBM Business Partner Use Only 78
78 © 2015 IBM Corporation
Construction firm
Construction company looks to IBM FlashSystem for key benefits

80-90 percent
data reduction anticipated from
estimation tool
High performance
for Oracle applications

Similar interface Business challenge: A leading construction and infrastructure firm in Asia with
for management as IBM XIV installed IBM System z and XIV systems wished to improve Oracle performance in
a manner compatible with their current solutions.
Solution Components The solution: After a successful Proof of Concept an IBM FlashSystem A9000 was
• IBM FlashSystem A9000
• IBM zEnterprise 196
chosen over competition to support Oracle applications and other workloads.
• IBM LinuxONE
• IBM XIV Business impact: Expectations are that significant performance improvement and
• IBM SAN96B switch data reduction rates with easy data transfer and simplified management will provide
• Competition: NetApp
the client with storage infrastructure to maintain and grow market position.
© Copyright IBM Corporation 2017 IBM & IBM Business Partner Use Only 79
79 © 2015 IBM Corporation
e-commerce bookseller
Bringing educational services to a higher standard with IBM

5-to-1
data reduction results achieved
in data reduction study

Sub-millisecond
response times

Virtualization Business challenge: Midwest-based provider of products and services to the


integration with VMware and educational market wanted to use an upcoming refresh to replace disk drives with
flash technology in both primary and secondary data centers.
SVC was critical to the win
The solution: Pursuant to a data reduction study by IBM and Sirius Computer
Solutions the IBM FlashSystem A9000R was chosen for the client’s highly
Solution Components virtualized environment.
• IBM FlashSystem A9000R
• IBM Spectrum Virtualize (SVC)
• VMware, Oracle, MS SQL Business impact: Expectations include excellent improvements to performance,
• Business Partner: Sirius Computer Solutions response times, data reduction and reduced IT footprint, consolidation of storage
• Competition: HDS resources and significant cost savings over a 3 year period.
© Copyright IBM Corporation 2017 IBM & IBM Business Partner Use Only 80
80 © 2015 IBM Corporation
Department store
IBM FlashSystem A9000R excels with Oracle over competition

Grid storage
simplifies management and
deployment

Excellent
inline data reduction with no
need for post processing

Proof of Concept Business challenge: Major USA-based department store chain needed a top
performing storage solution to meet the performance demands of mission critical
successfully demonstrated application platforms.
superiority over competition
The solution: Multiple IBM FlashSystem A9000R storage systems will be deployed
in support of Oracle retail applications and other workloads.
Solution Components
• IBM FlashSystem A9000R
• Oracle
Business impact: Outstanding performance, simplified management, optimized
• Competition: EMC, Pure Storage capacity, and a highly available configuration have provided the client with storage
infrastructure to meet current and future needs.
© Copyright IBM Corporation 2017 IBM & IBM Business Partner Use Only 81
81 © 2015 IBM Corporation
Get the Details
• Healthcare - Epic: https://ibm.biz/FlashAcceleratesEpic
• Healthcare - TriZetto: https://ibm.biz/FlashAcceleratesTriZettoFacetsQNXT
• Financial - FIS Profile: https://ibm.biz/FlashAcceleratesFISProfile
• Financial - FIS IST/Switch: http://bit.ly/FAST-FIS-IST-Switch
• Financial - Finacle: https://ibm.biz/FlashAcceleratesFinacleCoreBanking
• Financial - STAC M3 World Record: https://ibm.biz/FlashAcceleratesMcObject
• Middleware - DB2: https://ibm.biz/FlashAcceleratesIBMDB2Database
• Middleware - Oracle https://ibm.biz/FlashAcceleratesOracleDatabase
• Analytics - IBM InfoSphere Identity Insight: https://ibm.biz/FlashAcceleratesIBMIdentityInsight
• General - IBM Systems Magazine Article https://ibm.biz/BdFFEC
• General - FlashSystem: The Ideal Storage for Power Systems
http://www.ibm.com/support/techdocs/atsmastr.nsf/WebIndex/WP102506

82 © 2015 IBM Corporation

You might also like