zEnterprise Reduces Cost Per Workload

zEnterprise Economics

Dr. John Shedletsky
VP, Competitive Technology

New metric
for the age
of Smarter
Computing COST PER
WORKLOAD

Accurately allocating cost in a
virtualized environment

zEnterprise Economics - IBM SWG CPO - Mar 2011 2

Smarter Computing With zEnterprise
Platforms Optimized For Consistent Structured
Different Workloads Management

z/OS z/VM AIX Linux

Best fit for workload Consistent structured practices

Lowest Cost Of Lowest Cost Of
Acquisition Per Operation Per
Workload Workload

Lowest Cost Per Workload

Agenda

z/OS z/VM AIX Linux

Right fit for workload Consistent structured practices
Lowest cost per workload Lowest labor costs

Cost Per Workload Why zBX is better than
Examples do-it-yourself
Consolidate standalone zManager labor savings
workloads Benefits of workload
Claims processing management
Consolidate hybrid front ends

Consolidating Workloads With Heavy I/O
Requirements
Benchmark to 10 workloads Virtualized on Intel
determine which per Intel blade 8 core Blade
platform provides $21,413 per workload
the lowest TCA
over 3 years

15 workloads PowerVM on PS701
per POWER7 blade 8 core Blade
Workloads
$14,325 per workload
IBM WebSphere ND
Monitoring software
On 4 core “Older” Intel

Online banking workloads,
240 workloads z/VM on zEnterprise CPF
each driving 22
transactions per second, per 32-way z/VM 32 IFLs
with 1 MB I/O per $14,052 per workload
transaction I/O bandwidth
Consolidation ratios derived from IBM internal studies. z196 32-way performance large scale pool
projected from z196 8-way and z10 32-way measurements. zBX with x blades is
a statement of direction only. Results may vary based on customer workload
profiles/characteristics. Prices will vary by country.

Consolidating Heavy Workloads

Benchmark to Virtualized on Intel
determine which 1 workload 8 core Blade
platform provides per Intel blade $214,133 per
the lowest TCA workload
over 3 years

Workloads
more parallel $107,437 per
IBM WebSphere ND
threads workload
Monitoring software
On 8 core Nehalem
servers
Online banking workloads, 23 workloads
each driving 460 z/VM on zEnterprise CEC
transactions per second with per 32-way z/VM 32 IFLs
light I/O $146,631 per
Consolidation ratios derived from IBM internal studies. z196 32-way performance workload


Consolidating Light Workloads

Benchmark to 36 workloads Virtualized on Intel
determine which per Intel blade 8 core Blade
platform provides Fast low cost
$5,948 per workload
the lowest TCA threads
over 3 years

Workloads
$6,320 per workload
IBM WebSphere ND
Monitoring software
On 4 core “Older” Intel

Online banking workloads,
each driving 22 transactions 270 workloads z/VM on zEnterprise CEC
per second with light I/O per 32-way z/VM 32 IFLs
$12,491 per workload
Consolidation ratios derived from IBM internal studies. z196 32-way performance


Cost Per Workload Reveals True Value

25 heavy
True Value: Cost Per Workload
240 heavy workloads
I/O
Workloads
235 light
Cost Of Hardware Is Misleading
workloads

Run 500 workloads Cost of Cost of Cost per
Hardware Software Workload
56 Intel Blades
(8 cores per blade)
448 cores total $1.7M $9.8M $23.0K
1 zEnterprise
Best fit $2.7M $4.7M $15.0K
192 cores total

Note: 3yr TCA. CPO benchmarks. Equal mix of WAS ND and DB2 workloads. List prices.

Agenda

z/OS z/VM AIX Linux



Customer Case Study –
Insurance Claim Processing Company

Primary Stable Business
Government Insurance Claims Processing
Medicare/Medicaid/Defense/State

Growth Business
Commercial Claims Processing
Two Existing Commercial Claims Processing Systems
− Homegrown CICS/DB2 Application on Mainframe
− ISV 3rd-party Package running on HPUX

Which platform is the low-cost option for future growth in
commercial claims?

10

Two Commercial Claims Processing Systems

HP Servers + ISV IBM System z CICS/DB2

Which system
costs less for
Production Servers
HP 9000 Superdome rp4440 future
HP Integrity rx6600
growth?
Total MIPS 11,302

MIPS Used for commercial
Dev/Test Servers claims processing
HP 9000 Superdome rp5470 Calculate production/dev/test 2418
HP Integrity rx6600
cost per
Claims per year 327,652 Claims per year 4,056,000
workload
Buy Build

Allocated Annual Costs For Two Systems
Mainframe Distributed
Provided by customer
Hardware 1,302,205 87,806 finance department
Hardware Maint 315,548
Software IBM MLC 4,842,384
Software Non IBM OTC 647,843 196,468
Software Non IBM MLC 5,027,936
Storage 877,158
Network 418,755
Support Staff 2,324,623 257,289
Platform + Staff Total 15,756,452 541,563

Platform + Staff Claims Allocation 3,371,880 (21.4%) 541,563 (100%)
Billing Center 1,611,650
Call Center 2,920,090
Development 1,907,382
Total 9,811,002 541,563
Claims Processed 4,056,000 327,652
$ Per Claim 2.42 0.87

Mainframe costs easily
Software Non IBM OTC 647,843 196,468 identified, distributed
Software Non IBM MLC 5,027,936 costs difficult to identify
Storage 877,158
Network 418,755
Support Staff 2,324,623 257,289

Total 9,811,002 541,563
$ Per Claim 2.42 0.87

Storage 877,158
Network 418,755 Billing and Call center
costs allocated to
Support Staff 2,324,623 257,289
mainframe, but would be
Platform + Staff Total 15,756,452 541,563 the same for either
option
Total 9,811,002 541,563
$ Per Claim 2.42 0.87

Storage 877,158
Network 418,755 Billing and Call center
costs allocated to
Support Staff 2,324,623 257,289
mainframe, but would be
Platform + Staff Total 15,756,452 541,563 the same for either
option
Billing Center 1,611,650 Development still
required to customize
Call Center 2,920,090 packaged software for
Development 1,907,382 each new contract
Total 9,811,002 541,563
$ Per Claim 2.42 0.87

True Workload Costs For Either Option
Hardware 1,302,205 87,806
Software Non IBM OTC 647,843 196,468
Software Non IBM MLC 5,027,936
Storage 877,158 ?
Network 418,755 ?
Support Staff 2,324,623 257,289

Billing Center same same
Call Center same same
Development 1,907,382 123,031
Total 5,279,262 664,594
Claims Processed 4,056,000 327,652 Mainframe has lower
$ Per Claim 1.30 2.03 cost per workload

A Note On Support Staff Annual Costs

HP Servers + ISV IBM System z CICS/DB2
$0.79 per claim

Production Servers
HP 9000 Superdome rp4440
$0.12 per claim
HP Integrity rx6600

Total MIPS 11,302

MIPS Used for commercial
Dev/Test Servers claims processing
HP 9000 Superdome rp5470 production/dev/test 2418
Mainframe
HP Integrity rx6600
support staff has
Claims per year 327,652 6.6x better Claims per year 4,056,000
productivity


A Note On Processing Equivalence
Current Computing Capacity Allocated For Distributed is
38918 RPE’s

Using the Scaling Factor (Claims/Sec Distributed /
Claims/Sec Mainframe) we calculate that processing the
equivalent Claims volume on the mainframe would require
only 195.46 MIPS

This gives us an RPE/MIPS factor of 199.10 (Distributed
capacity to Mainframe capacity ratio)

We have studied Offloads that range from 122 RPE/MIPS
to 670 RPE/MIPS
18

Agenda

z/OS z/VM AIX Linux



Collapse Web Front End Workloads Into
zEnterprise Platform
Web facing Message CICS/DB2
front-end hub core system
Run as ensemble of virtual
servers

Unified management of
virtual machines

Manage ensemble as a
AIX on Power DataPower z/OS single workload with service
Blade
XI50z goals

WAS ESB CICS/DB2 Assign best fit to Power
blade and XI50z for lowest
cost per workload
AIX z/OS
Embedded pre-configured
zEnterprise BladeCenter zEnterprise data network
Extension (xBX) z196

DataPower XI50z – Built For Purpose Appliance
Enterprise Microsoft BizTalk Server 492 messages per sec
Service Bus Windows on Intel Server
benchmark BizTalk
$764 per mps
4 sockets, 32 cores
comparison Server
128 GB
Windows

Oracle Service Bus 5,839 messages per sec
OSB Oracle Linux on HP DL380
$120 per mps
2 sockets, 12 cores
Linux 128 GB

messages messages

DataPower
XI50z
5,117 messages per sec
$33 per mps
Tests consists of measuring maximum throughput DataPower
of ESB while performing a variety of message
mediation workloads: pass-through, routing,
XI50z in zBX
transformation, and schema validation HS 22, 8 cores

zEnterprise Economics - IBM SWG CPO -- Feb. 2011
zEnterprise Economics SWG CPO Mar 2011 21

Web Front Ends Cost 58% Less On zEnterprise
28 front end Competitive App Server
applications on older 57 SPARC T3-1B blades
in SUN racks
SPARC T2+ servers 2 HP DL380 servers
(for ESB)
936 cores total
Web
Facing
28 workloads
each driving Upgrade to $11.7M
1975 tps
new SPARC T3 3yr TCA
hardware HW+SW

SPARC T5440 WebSphere App Server
32 core servers 28 POWER7 blades
2 DataPower XI50z
HP DL380 servers in zBX
(for ESB) 224 cores total

Power Blades $4.9M
in zBX 3yr TCA
HW+SW

Web Front Ends Cost 58% Less On zEnterprise

Why?
Competitive App Server
57 SPARC T3-1B blades
in SUN racks
2 HP DL380 servers WAS on PS701
(for ESB)
936 cores total delivers 1.84x
processing capacity
Competitive
Upgrade to $11.7M
new SPARC T3 Application Server
3yr TCA
hardware HW+SW cannot effectively
utilize the threads
available in T3
blade
WebSphere App Server DataPower better
28 POWER7 blades
2 DataPower XI50z price/performance
in zBX
224 cores total Need to over
provision SPARC T3
Power Blades $4.9M since no zManager
in zBX 3yr TCA
HW+SW

SAP Applications Cost 20% Less On zEnterprise

20 front end SAP
applications on older 38 SPARC T3-1B
blades in SUN rack
SPARC T2+ servers 608 cores total

T3-1B z196
SAP
Upgrade to $1.2M
20 workloads
new SPARC T3
3yr TCA
hardware HW+SW

20
SPARC T5440
32 core servers 23 POWER7 blades
538,120 total SAPs in zBX
640 cores total 184 cores total
zBX z196

Power Blades $0.97M
in zBX 3yr TCA
zEnterprise Economics - IBM SWG CPO - Mar 2011 HW+SW 24

Agenda

z/OS z/VM AIX Linux


Claims processing zManager labor savings
Consolidate standalone Benefits of workload
workloads management

Compare The Costs For 92 Hybrid Workloads
I see how fit for
purpose can cut costs,
POWER7 blades in
but why do I need zBX BladeCenter chassis
racks?

Can’t I just use
BladeCenters to do it Web Facing

myself? Do-It-Yourself
72 workloads

SAP

20 workloads

POWER7 blades
in zBX racks with
zManager zEnterprise
(Manage+Automate)


Labor Cost Trends Favor A Centralized
Structured Approach To Management

Intel - Virtual Servers/FTE z/OS - MIPS/FTE
500
450
Large scale consolidation and

t
400

en
consistent structured

em
350
management practices drive
ov
300
pr increases in labor productivity
Im
250
x
18

200
150
Small scale consolidation with
100
ment
50 3.9x Improve ad hoc management achieves
0 lesser gains
2000/01 2007/08
Year

The more workloads you consolidate and manage with
consistent structured practices…
the lower the management labor cost
Source: IBM Scorpion Studies and IDC Three Data
Centers – One Vision.PDF (IDC, 2010) zEnterprise Economics - IBM SWG CPO - Mar 2011 27

zManager Provides Structured Management For
zEnterprise Virtual Environments

Process Typical Distributed zManager
Management Practices
Asset Discover assets with ad hoc Automated discovery and
Management methods management of entitlement assets
Manual entitlement management
Deployment Manually configure hypervisor and Automated deployment of
Management build networks hypervisor and attachment to
integrated networks

Security Different ways to manage Centralized, fine-grained
Management administrator access administrator access management

Change No visibility into impact of changes Track dependencies for change
Management impact

Capacity and No end-to-end transaction End-to-end transaction monitoring to
Performance monitoring isolate issues
Management Manually adjust CPU resources to Automatic CPU resource
meet changing workload demands adjustments to meet changing
workload demands


Hypervisor Setup And Configuration Lab Test –
Do-It-Yourself vs. zManager
DIY Tasks (per Blade) Elapsed Time Labor Time
Initial communication setup & education 6 min 26 sec 6 min 26 sec
Boot VIOS disc & install (creates LPAR for VIOS automatically) 37 min 59 sec 36 min
Configure VIOS networking 2 min 49 sec 2 min 49 sec
Create new storage pool for LPARs 35 sec 35 sec
Install VIOS service fix packs 61 min 5 sec 20 sec

TOTAL TIME 1 hr 48 min 52 sec 46 min 10 sec

zManager Tasks (per Blade) Elapsed Time Labor Time
Add entitlement for a blade 90 min 92 sec

TOTAL TIME 1 hr 30 min 1 min 32 sec

97% reduction
in labor time

Source: IBM CPO Internal Study zEnterprise Economics - IBM SWG CPO - Mar 2011 29

Network Setup And Configuration Lab Test –
Do-It-Yourself vs. zManager
Do-It-Yourself Tasks (for two BladeCenters) Elapsed/Labor Time
Planning (includes time to go over docs, etc) 5 hrs
Cabling 2 hrs
AMM Configuration 2 hrs
Logical Configuration (L2) 8 hrs
Blades network configuration 4 hrs
Testing 2 hrs
Documenting the configuration 3 hrs

TOTAL TIME 26 hrs

zManager Tasks (for two BladeCenters) Elapsed/Labor Time
Planning 3 hrs
Cabling (pre-cabled in zBX) 0 hrs
AMM Configuration (done in zBX) 0 hrs
Logical configuration (L2) 30 mins
Blades network configuration 1 hr 30 mins
Testing (pre-tested) 0 hrs
Documenting the configuration (all part of zManager) 0 hrs

TOTAL TIME 5 hrs 81% reduction
in labor time


Performance Manager Lab Test –
Automatic Allocation Of CPU Resource

Time
zManager monitors
virtual machine
Workload 2 (W2)
performance and
automatically adjusts
15 minutes CPU resources as
Goal
needed

Considers priority and
Workload 1 (W1)
performance relative
to service level
agreement goals
2. Performance Manager is 4. W1 reaches
turned on. zManager detects performance
W1 is underperforming. goal. Reduces the need to
1. No performance
over-provision CPU
management. W1
is underperforming,
3. Over time, zManager resources
adjusts CPU resources,
and W2 is over- taking from W2 and
performing. giving to W1.

zManager Performance Management Reduces
Need To Overprovision CPU Resource
Without zManager With zManager

Total CPU Resource Needed
Total CPU Resource Needed

110% No additional CPU
No additional CPU
resourcesneeded
resource needed
100% 100%
10% more CPU
resources needed Unexpected 20% spike
in average demand,
Performance Manager
Unexpected increases entitlement by
20% spike in 20%
average 13.2% 13.2%
demand
Workload 1 Workload 1
11% 11%

Average utilization
Average utilization

Performance manager
reduces entitlement by
20%
11% Workload 2 Workload 2
11%

8.8%

Must overprovision CPU resource for Performance manager enables trading off
both workloads by 10% to handle resource from lower priority workload,
unexpected spike in demand avoiding the need to overprovision

Case Study:
Compare The Costs For 92 Hybrid Workloads
62
POWER7 blades in 5 Blades Cost
BladeCenter chassis of Acquisition $7,325K
Labor $ 987K
Total (3yr) $8,312K

$30.1K per
Web Facing
Do-It-Yourself
72 workloads Data
Processing
workload per year
20 workloads

zBX Cost
of Acquisition $6,994K
Labor $ 714K
72 Web facing hybrid
applications, 2 per POWER7
blades
56 Total (3yr) $7,708K
POWER7 blades
20 SAP hybrid applications, 1
per POWER7 blade. DIY blades
over provisioned by 10%
in two zBX racks with
zManager zEnterprise $27.9K per
because no zManager
performance manager (Manage+Automate) workload per year
Cost of CICS/DB2 and SAP
components of workloads not
included
Labor costs are for blade
management only 28% less labor cost and 7% less cost per workload
Labor rate $159,600 per year
Results may vary based on customer workload
profiles/characteristics. Prices will vary by country. zEnterprise Economics - IBM SWG CPO - Mar 2011 33

zManager Labor Reduction Benefits

4289 total hours per year reduced
by 28% to 3104 hours per year

Change Management Deployment Management
Automatic setup and
Standardization of images
configuration of the
and firmware, visibility into Reduced
hypervisor and out-of-the-
relationships among by 33%
box networks
resources
Reduced Reduced
by 41% by 52% Capacity/Performance
Management
Automation to
Reduced isolate and fix issues
by 9%

Security Management Reduced
by 20% Asset Management
Centralized fine-grain Automated discovery,
administrator access entitlement management
control


Summary

Cost per workload is the
key metric for the new IT
economics

Fit for purpose reduces
cost of acquisition per
workload

zEnterprise integrated
management reduces cost
of labor per workload


Smarter Computing: System z Analyst Summit

Thank you!
ibm.com/smartercomputing

36 © 2011 International Business Machines Corporation
36

Smarter Computing: System z Analyst Summit

Trademarks and disclaimers
Intel, Intel logo, Intel Inside, Intel Inside logo, Intel Centrino, Intel Centrino logo, Celeron, Intel Xeon, Intel SpeedStep, Itanium, and Pentium are trademarks or registered trademarks of Intel Corporation or its subsidiaries in the United States
and other countries./ Linux is a registered trademark of Linus Torvalds in the United States, other countries, or both. Microsoft, Windows, Windows NT, and the Windows logo are trademarks of Microsoft Corporation in the United States, other
countries, or both. IT Infrastructure Library is a registered trademark of the Central Computer and Telecommunications Agency which is now part of the Office of Government Commerce. ITIL is a registered trademark, and a registered
community trademark of the Office of Government Commerce, and is registered in the U.S. Patent and Trademark Office. UNIX is a registered trademark of The Open Group in the United States and other countries. Java and all Java-based
trademarks and logos are trademarks or registered trademarks of Oracle and/or its affiliates. Other company, product, or service names may be trademarks or service marks of others. Information is provided "AS IS" without warranty of any
kind.

The customer examples described are presented as illustrations of how those customers have used IBM products and the results they may have achieved. Actual environmental costs and performance characteristics may vary by customer.

Information concerning non-IBM products was obtained from a supplier of these products, published announcement material, or other publicly available sources and does not constitute an endorsement of such products by IBM. Sources for
non-IBM list prices and performance numbers are taken from publicly available information, including vendor announcements and vendor worldwide homepages. IBM has not tested these products and cannot confirm the accuracy of
performance, capability, or any other claims related to non-IBM products. Questions on the capability of non-IBM products should be addressed to the supplier of those products.

All statements regarding IBM future direction and intent are subject to change or withdrawal without notice, and represent goals and objectives only.

Some information addresses anticipated future capabilities. Such information is not intended as a definitive statement of a commitment to specific levels of performance, function or delivery schedules with respect to any future products. Such
commitments are only made in IBM product announcements. The information is presented here to communicate IBM's current investment and development activities as a good faith effort to help with our customers' future planning.

Performance is based on measurements and projections using standard IBM benchmarks in a controlled environment. The actual throughput or performance that any user will experience will vary depending upon considerations such as the
amount of multiprogramming in the user's job stream, the I/O configuration, the storage configuration, and the workload processed. Therefore, no assurance can be given that an individual user will achieve throughput or performance
improvements equivalent to the ratios stated here.

Prices are suggested U.S. list prices and are subject to change without notice. Starting price may not include a hard drive, operating system or other features. Contact your IBM representative or Business Partner for the most current pricing in
your geography.

Photographs shown may be engineering prototypes. Changes may be incorporated in production models.

© IBM Corporation 2011. All rights reserved.
References in this document to IBM products or services do not imply that IBM intends to make them available in every country.

Trademarks of International Business Machines Corporation in the United States, other countries, or both can be found on the World Wide Web at http://www.ibm.com/legal/copytrade.shtml.

37 © 2011 International Business Machines Corporation
37

Backup


Web Workloads - Performance Data
Server Description Cores RPE RPE2 TPS TPS
(Measured) (Calculated)
T5440 Sun SPARC Enterprise 32 102,617 21,580 5,300
T5440 (4U) UltraSPARC Max throughput
SPARCT2+ T2+ 1.6GHz 4MB with multi JVM run
(4ch/32co) using Solaris
32 core containers

T3-1B Oracle SPARC T3-1B 16 61,249 11,520 2829
SPARC T3 1.65GHz Calculated as
16 core (1ch/16co) 5300*(11520/21580)

Power 710 Power 710 Express (2U) 8 60,208 14,920 5,670
Power7 3.55GHz
8 core (1ch/8co)

PS701 IBM BladeCenter PS701 8 53,189 13,730 5218
Express Power7 3.0GHz
Calculated as
8 core (1ch/8co) 5670*(13730/14920)

WAS on PS701 versus WLS on T3-1B is 1.84x
WLS on T5440 versus WLS on T3-1B is 1.87x
WLS on T5440 versus WAS on PS701 is nearly 1x

Case Study:
zEnterprise Reduces Infrastructure Labor Hours

5000
4500
4000
28%
fewer labor
Yearly Labor Hours

3500 hours Change Mgmt
3000 Security Mgmt
2500 Asset Mgmt
2000 Capacity/Perf Mgmt
1500 Deployment Mgmt
1000
500
0
Do-It-Yourself zEnterprise


zEnterprise Reduces Cost Per Workload

Recommandé

Recommandé

Contenu connexe

Tendances

Tendances (18)

Similaire à zEnterprise Reduces Cost Per Workload

Similaire à zEnterprise Reduces Cost Per Workload (20)

Plus de dkang

Plus de dkang (8)

Dernier

Dernier (20)

zEnterprise Reduces Cost Per Workload