© 2009 ibm corporation the value of consolidation with enterprise power systems name of the...

© 2009 IBM Corporation

The Value of Consolidation with Enterprise Power Systems

Name of the presenterTitle of the presenter

© 2009 IBM Corporation3

Globally, systems and infrastructure are reaching a breaking point.

Proliferation of servers and networking devices Excessive energy usage and heating problems Inadequate power and cooling infrastructure Data silos and data synchronization Expectations that “everything” is connected Linear staffing costs Skyrocketing software costs Unexplained outages

Meanwhile, customer expectations, competitive pressures, regulatory requirements and fiscal pressures are increasing.


23%

25%

26%

28%

33%

36%

37%

42%

44%

53%

0% 10% 20% 30% 40% 50% 60%

Storage virtualization

Disaster recovery

Storage hardware

Security

Business intelligence

Data center consolidation

Application integration

Cost cutting

Server consolidation

Server virtualization

% of respondents rating issue as high priority

CIO key spending initiatives

IT Leaders Need to do More With Less

Operational issues can inhibit business growth and innovation

Overburdened IT Overburdened IT staffstaff

Complex service Complex service deliverydelivery

Strained IT Strained IT budgetsbudgets

Source: Goldman Sachs Group IT Spending Survey, July 2008


Consolidating with Enterprise Power Systems enables clients to…

Server consolidation improves service to clients by delivering flexible performance, dynamic provisioning and enabling clients to avoid disruption

Server consolidation with shared resources enables high system utilization, which lowers the cost of ownership by reducing networking, energy, floor space, and software costs.

REDUCE COST

IMPROVE SERVICE

MANAGE RISK

Server consolidation manages IT risk by improving security, increasing business resiliency and simplifying operations.

Power™ Systems and PowerVM™ are designed to deliver effective consolidation in the most demanding data centers


REDUCE COST by consolidating with Power Systems

Resource sharing Sharing system resources through virtualized

consolidation reduces unused system overhead

Virtualized consolidated systems are evidenced by high utilization rates

High utilization means less hardware

Environmentally friendly Less power and cooling is required

Less floor space is required

Fiscally responsible Fewer processor cores drives less software costs

Newer systems are more reliable and less costly to maintain than older systems

Fewer systems translates to reduced people costs


0%

20%

40%

60%

80%

100%

Configuration planned for growth (20% unused?)

Configuration planned for peaks (50% unused?)

System waits for I/O and memory access even when it is working (20% unused?)

Typical UNIX or x86 serving running a single operating environment is 10 - 20% utilized

Result is that 80% of the hardware, software, maintenance, floor space, and energy that you pay for, is wasted

What you pay for

What you get

Typical small server utilization


Example consolidation environment

Single system / single workload– Hypothetical workload– No partitions, no virtualization– System is dedicated to application– System could be UNIX or x86 based– Assumed average utilization of 20% (industry averages are 8 - 10% for x86

and 15 - 20% for UNIX)– Assumed peak of ~4X the average

utilization– Peaks are assumed to be random

Single Application Server

0%

10%

20%

30%

40%

50%

60%

70%

80%


Typical Scale-out Approach

Single workload on a single system– Assumed average utilization of 20%– Assumed peak of 4X– Peaks are assumed to be random

Eight separate workloads on eight identical systems– Same assumptions


0%

10%

20%

30%

40%

50%

60%

70%

80%


0%

10%

20%

30%

40%

50%

60%

70%

80%


0%

10%

20%

30%

40%

50%

60%

70%

80%


0%

10%

20%

30%

40%

50%

60%

70%

80%


0%

10%

20%

30%

40%

50%

60%

70%

80%


0%

10%

20%

30%

40%

50%

60%

70%

80%

Single Application Serve

0%

10%

20%

30%

40%

50%

60%

70%

80%


0%

10%

20%

30%

40%

50%

60%

70%

80%


0%

10%

20%

30%

40%

50%

60%

70%

80%

Result is 8X the hardware, software, maintenance, and floor space that you pay for, is wasted


Same Scenario with Physical Server Consolidation

Single workload on a single system– Average Utilization: 20%

8 separate workloads on one partitioned system– Average utilization is still 20%

8 times the hardware in one physical system


0%

10%

20%

30%

40%

50%

60%

70%

80%

Utilization remains the same due to no resource sharing

Result is 8X the hardware and software that you pay for, is wasted

Partitioned Application Server with eight fixed partitions

0%

10%

20%

30%

40%

50%

60%

70%

80%

0%

10%

20%

30%

40%

50%

60%

70%


Same Scenario with Virtualized Server Consolidation(Shared resources)

8 to 1 Systems Consolidation (16 cores)

0%

10%

20%

30%

40%

50%

60%

70%

80%

* Assumes independent workloads

Single Application Server (4 cores)

0%

10%

20%

30%

40%

50%

60%

70%

80%


0%

10%

20%

30%

40%

50%

60%

70%

80%


0%

10%

20%

30%

40%

50%

60%

70%

80%


0%

10%

20%

30%

40%

50%

60%

70%

80%


0%

10%

20%

30%

40%

50%

60%

70%

80%


0%

10%

20%

30%

40%

50%

60%

70%

80%


0%

10%

20%

30%

40%

50%

60%

70%

80%


0%

10%

20%

30%

40%

50%

60%

70%

80%

Eight separate workloads on eight identical systems– Average utilization is 20%

Eight separate workloads on one system*– Average utilization is 39%

32 cores reduced to 16 cores (2 to 1)

Utilization increases


0%

10%

20%

30%

40%

50%

60%

70%

80%


The effect of sharing system resources Average utilization per system almost doubles

– 20% to 39%

Peak usage actually drops– 79% to 76%

Any single application now has access to more resource– Previous peaks, capped at 4 cores, can now go to 16 cores– Critical workloads can now be prioritized and enabled to run faster– Batch jobs, for instance, now run faster, due to the ability to access more capacity

16 cores of unused capacity eliminated and applications run faster

0%

10%

20%

30%

40%

50%

60%

70%

80%

Single Application Server (4 cores) 8 to 1 Consolidation (16 cores)


For example…by consolidating smaller IBM systems and sharing resources

Eight 4-core 520/550 systems could be consolidated into one 16 core 570 system

Achieved by consolidating onto one system, sharing resources across eight workloads, doubling the system utilization rate, and improving single application performance


Example assumes independent workloads and equal per core performance


0%

10%

20%

30%

40%

50%

60%

70%

80%


Larger Scenario with Virtualized Server Consolidation(Shared resources)

Sixteen separate workloads on sixteen identical systems– Average utilization is 20%

– Peak is 79%

Sixteen separate workloads on one system*– Average utilization is 48%

– Peak is 78%

64 cores reduced to 24 cores (2.65 to 1)


0%

10%

20%

30%

40%

50%

60%

70%

80%



0%

10%

20%

30%

40%

50%

60%

70%

80%


0%

10%

20%

30%

40%

50%

60%

70%

80%


0%

10%

20%

30%

40%

50%

60%

70%

80%


0%

10%

20%

30%

40%

50%

60%

70%

80%


0%

10%

20%

30%

40%

50%

60%

70%

80%


0%

10%

20%

30%

40%

50%

60%

70%

80%


0%

10%

20%

30%

40%

50%

60%

70%

80%


0%

10%

20%

30%

40%

50%

60%

70%

80%


0%

10%

20%

30%

40%

50%

60%

70%

80%


0%

10%

20%

30%

40%

50%

60%

70%

80%


0%

10%

20%

30%

40%

50%

60%

70%

80%


0%

10%

20%

30%

40%

50%

60%

70%

80%


0%

10%

20%

30%

40%

50%

60%

70%

80%


0%

10%

20%

30%

40%

50%

60%

70%

80%


0%

10%

20%

30%

40%

50%

60%

70%

80%


0%

10%

20%

30%

40%

50%

60%

70%

80%



0%

10%

20%

30%

40%

50%

60%

70%

80%

For example…by consolidating smaller IBM systems and sharing resources

Sixteen 4-core 520/550 systems could be consolidated into one 24 core 570 system

Achieved by consolidating onto one system, sharing resources across sixteen workloads, increasing the system utilization rate, and improving single application performance


Example assumes independent workloads and equal per core performance


Very Large Scenario with Virtualized Server Consolidation(Shared resources)

64 separate workloads on 64 identical systems– Average utilization is 20%

– Peak is 79%

64 separate workloads on one system*– Average utilization is 61%

– Peak is 78%



0%

10%

20%

30%

40%

50%

60%

70%

80%



0%

10%

20%

30%

40%

50%

60%

70%

80%


0%

10%

20%

30%

40%

50%

60%

70%

80%


0%

10%

20%

30%

40%

50%

60%

70%

80%


0%

10%

20%

30%

40%

50%

60%

70%

80%


0%

10%

20%

30%

40%

50%

60%

70%

80%


0%

10%

20%

30%

40%

50%

60%

70%

80%


0%

10%

20%

30%

40%

50%

60%

70%

80%


0%

10%

20%

30%

40%

50%

60%

70%

80%


0%

10%

20%

30%

40%

50%

60%

70%

80%


0%

10%

20%

30%

40%

50%

60%

70%

80%


0%

10%

20%

30%

40%

50%

60%

70%

80%


0%

10%

20%

30%

40%

50%

60%

70%

80%


0%

10%

20%

30%

40%

50%

60%

70%

80%


0%

10%

20%

30%

40%

50%

60%

70%

80%


0%

10%

20%

30%

40%

50%

60%

70%

80%


0%

10%

20%

30%

40%

50%

60%

70%

80%


0%

10%

20%

30%

40%

50%

60%

70%

80%


0%

10%

20%

30%

40%

50%

60%

70%

80%


0%

10%

20%

30%

40%

50%

60%

70%

80%


0%

10%

20%

30%

40%

50%

60%

70%

80%


0%

10%

20%

30%

40%

50%

60%

70%

80%


0%

10%

20%

30%

40%

50%

60%

70%

80%


0%

10%

20%

30%

40%

50%

60%

70%

80%


0%

10%

20%

30%

40%

50%

60%

70%

80%


0%

10%

20%

30%

40%

50%

60%

70%

80%


0%

10%

20%

30%

40%

50%

60%

70%

80%


0%

10%

20%

30%

40%

50%

60%

70%

80%


0%

10%

20%

30%

40%

50%

60%

70%

80%


0%

10%

20%

30%

40%

50%

60%

70%

80%


0%

10%

20%

30%

40%

50%

60%

70%

80%


0%

10%

20%

30%

40%

50%

60%

70%

80%


0%

10%

20%

30%

40%

50%

60%

70%

80%


0%

10%

20%

30%

40%

50%

60%

70%

80%


0%

10%

20%

30%

40%

50%

60%

70%

80%


0%

10%

20%

30%

40%

50%

60%

70%

80%


0%

10%

20%

30%

40%

50%

60%

70%

80%


0%

10%

20%

30%

40%

50%

60%

70%

80%


0%

10%

20%

30%

40%

50%

60%

70%

80%


0%

10%

20%

30%

40%

50%

60%

70%

80%


0%

10%

20%

30%

40%

50%

60%

70%

80%


0%

10%

20%

30%

40%

50%

60%

70%

80%


0%

10%

20%

30%

40%

50%

60%

70%

80%


0%

10%

20%

30%

40%

50%

60%

70%

80%


0%

10%

20%

30%

40%

50%

60%

70%

80%


0%

10%

20%

30%

40%

50%

60%

70%

80%


0%

10%

20%

30%

40%

50%

60%

70%

80%


0%

10%

20%

30%

40%

50%

60%

70%

80%


0%

10%

20%

30%

40%

50%

60%

70%

80%


0%

10%

20%

30%

40%

50%

60%

70%

80%


0%

10%

20%

30%

40%

50%

60%

70%

80%


0%

10%

20%

30%

40%

50%

60%

70%

80%


0%

10%

20%

30%

40%

50%

60%

70%

80%


0%

10%

20%

30%

40%

50%

60%

70%

80%


0%

10%

20%

30%

40%

50%

60%

70%

80%


0%

10%

20%

30%

40%

50%

60%

70%

80%


0%

10%

20%

30%

40%

50%

60%

70%

80%


0%

10%

20%

30%

40%

50%

60%

70%

80%


0%

10%

20%

30%

40%

50%

60%

70%

80%


0%

10%

20%

30%

40%

50%

60%

70%

80%


0%

10%

20%

30%

40%

50%

60%

70%

80%


0%

10%

20%

30%

40%

50%

60%

70%

80%


0%

10%

20%

30%

40%

50%

60%

70%

80%


0%

10%

20%

30%

40%

50%

60%

70%

80%


0%

10%

20%

30%

40%

50%

60%

70%

80%


Virtualization enables higher system usage through consolidation and workload smoothing

0

10

20

30

40

50

60

70

1 8 16 64

# of workloads

% o

f u

tiliz

ati

on

System utilization increases as the number of mixed workloads are added

The amount of “leverage” increases as the number of workloads are added

The potential for savings increases with the amount of consolidation

Delta in system capacity as a result of shared resources

0

50

100

150

200

250

300

1 8 16 32 48 64

# of workloads

# o

f C

PU

s

2 to 1

2.6 to 1

3.5 to 1


Server Consolidation Concepts

Sharing system resources dramatically improves system utilization and reduces required capacity

Degree of compaction can be affected by:

Mixture of workloads (Different application characteristics)

Amount of consolidation (8 workloads or 80 workloads)

System design (Ability to feed processor)

Sophistication of scheduling mechanism (PowerVM)

Degree of priority (Ability to assign various priorities)

Granularity of work allocation (1/100th of a processor)

Server consolidation experience with SAP workloads on IBM Power Systems and System z™ support these conclusions


Power Systems Impact on Server Consolidation Concepts

3X reduction in number of cores is possible with POWER6™ and PowerVM in a virtualized consolidation environment

assuming all of the cores have equal performance

What if the new cores have better performance?

IBM POWER6 technology with PowerVM has the potential to reduce the number of cores by six fold or more

0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1

SPECint_rate2006per core

SPECfp_rate2006per core

SPECjbb2005 percore

2-tier SAP SD percore

Performance per core relative to the IBM Power 595

IBM Power 595 HP rx2660 Sun T5240

See “Power 595 performance results” in backup for full details


Very Large Scenario with Virtualized Server Consolidation

64 separate workloads on 64 identical systems– Average utilization is 20%– Peak is 79%

64 separate workloads on one system*– Average utilization is 61%– Peak is 78%



0%

10%

20%

30%

40%

50%

60%

70%

80%


with POWER6 technology

Single Application Server (4 CPUs)

0%

10%

20%

30%

40%

50%

60%

70%

80%


0%

10%

20%

30%

40%

50%

60%

70%

80%


0%

10%

20%

30%

40%

50%

60%

70%

80%


0%

10%

20%

30%

40%

50%

60%

70%

80%


0%

10%

20%

30%

40%

50%

60%

70%

80%


0%

10%

20%

30%

40%

50%

60%

70%

80%


0%

10%

20%

30%

40%

50%

60%

70%

80%


0%

10%

20%

30%

40%

50%

60%

70%

80%


0%

10%

20%

30%

40%

50%

60%

70%

80%


0%

10%

20%

30%

40%

50%

60%

70%

80%


0%

10%

20%

30%

40%

50%

60%

70%

80%


0%

10%

20%

30%

40%

50%

60%

70%

80%


0%

10%

20%

30%

40%

50%

60%

70%

80%


0%

10%

20%

30%

40%

50%

60%

70%

80%


0%

10%

20%

30%

40%

50%

60%

70%

80%


0%

10%

20%

30%

40%

50%

60%

70%

80%


0%

10%

20%

30%

40%

50%

60%

70%

80%


0%

10%

20%

30%

40%

50%

60%

70%

80%


0%

10%

20%

30%

40%

50%

60%

70%

80%


0%

10%

20%

30%

40%

50%

60%

70%

80%


0%

10%

20%

30%

40%

50%

60%

70%

80%


0%

10%

20%

30%

40%

50%

60%

70%

80%


0%

10%

20%

30%

40%

50%

60%

70%

80%


0%

10%

20%

30%

40%

50%

60%

70%

80%


0%

10%

20%

30%

40%

50%

60%

70%

80%


0%

10%

20%

30%

40%

50%

60%

70%

80%


0%

10%

20%

30%

40%

50%

60%

70%

80%


0%

10%

20%

30%

40%

50%

60%

70%

80%


0%

10%

20%

30%

40%

50%

60%

70%

80%


0%

10%

20%

30%

40%

50%

60%

70%

80%


0%

10%

20%

30%

40%

50%

60%

70%

80%


0%

10%

20%

30%

40%

50%

60%

70%

80%


0%

10%

20%

30%

40%

50%

60%

70%

80%


0%

10%

20%

30%

40%

50%

60%

70%

80%


0%

10%

20%

30%

40%

50%

60%

70%

80%


0%

10%

20%

30%

40%

50%

60%

70%

80%


0%

10%

20%

30%

40%

50%

60%

70%

80%


0%

10%

20%

30%

40%

50%

60%

70%

80%


0%

10%

20%

30%

40%

50%

60%

70%

80%


0%

10%

20%

30%

40%

50%

60%

70%

80%


0%

10%

20%

30%

40%

50%

60%

70%

80%


0%

10%

20%

30%

40%

50%

60%

70%

80%


0%

10%

20%

30%

40%

50%

60%

70%

80%


0%

10%

20%

30%

40%

50%

60%

70%

80%


0%

10%

20%

30%

40%

50%

60%

70%

80%


0%

10%

20%

30%

40%

50%

60%

70%

80%


0%

10%

20%

30%

40%

50%

60%

70%

80%


0%

10%

20%

30%

40%

50%

60%

70%

80%


0%

10%

20%

30%

40%

50%

60%

70%

80%


0%

10%

20%

30%

40%

50%

60%

70%

80%


0%

10%

20%

30%

40%

50%

60%

70%

80%


0%

10%

20%

30%

40%

50%

60%

70%

80%


0%

10%

20%

30%

40%

50%

60%

70%

80%


0%

10%

20%

30%

40%

50%

60%

70%

80%


0%

10%

20%

30%

40%

50%

60%

70%

80%


0%

10%

20%

30%

40%

50%

60%

70%

80%


0%

10%

20%

30%

40%

50%

60%

70%

80%


0%

10%

20%

30%

40%

50%

60%

70%

80%


0%

10%

20%

30%

40%

50%

60%

70%

80%


0%

10%

20%

30%

40%

50%

60%

70%

80%


0%

10%

20%

30%

40%

50%

60%

70%

80%


0%

10%

20%

30%

40%

50%

60%

70%

80%


0%

10%

20%

30%

40%

50%

60%

70%

80%


0%

10%

20%

30%

40%

50%

60%

70%

80%


Consolidation Template Example

Processor cores per systemQuantity of serversTotal processor coresConsolidation factorPerformance ratioRequired cores with consolidation

41664 2.65*

212

Mixture of workloadsAmount of consolidationAverage utilizationSystem designScheduling sophisticationDegree of priorityGranularity of work allocation

Range 2.0 to 6.0

X

..

..=

Performance per core of consolidated solutionPerformance per core of existing solution

Performance ratio =(Range 1.0 to 4.0)

5.3 consolidation ratio in processor cores (64/12)

(Range 2.0 to 24.0)

*Assumes no limitations or restrictions


POWER6 and PowerVMMade for each other

POWER6 systems are designed with scalable memory, memory bandwidth, and I/O bandwidth to handle high utilization rates

POWER6 and PowerVM enable sharing of:

POWER6 processors (Shared processors, Micro-Partitioning™)

POWER6 memory (Active Memory Sharing™)

POWER6 I/O (Virtual I/O Sharing)

PowerVM dynamically manages and adjusts system resources

PowerVM manages priorities with Partition Availability Priority

PowerVM allocates resources down to 1/100th of a processor

POWER6 and PowerVM support virtual servers up to 64 cores


Reduce cost with Active Memory Sharing

Dynamically adjusts memory available on a physical system for multiple virtual images based on their workload activity levels:

– Different workload peaks due to time zones– Mixed workloads with different time of day

peaks (e.g. CRM by day, batch at night)– Ideal for highly-consolidated workloads with

low or sporadic memory requirements

Available with PowerVM Enterprise Edition– Supports AIX®, IBM i and Linux workloads

Blends Power Systems hardware, firmware and software enhancements to optimize resources

– Supports over-commitment of logical memory

0

5

10

15

Asia

Americas

Europe

Mem

ory

Usa

ge

(GB

)

0

5

10

15

Night

Day

Time

Mem

ory

Usa

ge

(GB

)

Time

PowerVM dynamically optimizes shared memory


Scalability Factors VMware ESX 3.5(in VMware Infrastructure 3)

VMware ESX 4.0(in VMware vSphere 4)

PowerVM

Virtual CPUs per VM 4 8 64

Memory per VM 64 GB 256 GB 4096 GB

Virtual NICs per VM 4 10 256

CPUs per physical server 32 64 64

Memory per physical server 256 GB 512 GB 4096 GB

Virtual CPUs per physical server 192 512 640

Reduce cost: PowerVM delivers superior scalability

to maximize consolidation and drive down IT costs


IBM Power 595Designed with the capacity for consolidation

> 3.5X memory per core

> 4X memory bandwidth per core

> 4X I/O bandwidth per core

> 8X L2 + L3 cache per core

You can use the tremendous capacity of the IBM Power™ 595 to run challenging applications in every virtual server.

IBM Power 595 HP DL370 G6 Sun T5240

Cores 64 8 16 Memory (GB) 4,096 144 128 Memory Bandwidth (GB/s) 1,376 38.4 63.4 I/O Bandwidth (GB/s) 640 19.2 8 L2 + L3 cache 1,280 18 8

Memory (GB) per core 64 18 8 Memory Bandwidth (GB/s) per core 21.5 4.8 3.9625 I/O Bandwidth (GB/s) per core 10 2.4 0.5 L2 + L3 cache per core 20 2.25 0.5

0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1

Memory (GB) percore

Memory Bandwidth(GB/s) per core

I/O Bandwidth(GB/s) per core

L2 + L3 cache percore

Capacity per core relative to the Power 595

IBM Power 595 HP DL370 G6 Sun T5240

Source: HP QuickSpecs available at www.hp.com; Press Kit - Intel® Xeon® Processor 5500 Series available at www.intel.com; http://www.sun.com/servers/coolthreads/t5240All data is current as of May 5, 2009.


External NetworkExternal Network

Virtual LAN

Reduce cost and complexity with Virtual LAN

Virtual networking for server-to-server communication can sustain performance while reducing network utilization and physical points-of-failure

Removes the need for network adapters and switches for communication between LPARs

Can drastically improve network recovery in the event of a disaster – no physical cables to re-connect or diagnose

Reduce network latency

Increase server network throughput up to 3X!

*See “Virtual networking claim in backup for full details


Reduce Cost with Virtual I/O Server

VLAN

LPAR1 LPAR2 LPAR3 LPAR4 LPAR5 LPAR6 LPAR7 LPAR8

VLAN

LPAR9 LPAR10 LPAR11 LPAR12 LPAR13 LPAR14 LPAR15 LPAR16

VIOS

External Network

VIOS can dramatically reduce physical resources and associated costs through more effective resource sharing

Reduce I/O adapters & cabling with VIOS !Redundant VIOS enhances reliability

Network


Reduce cost by eliminating hardware and points-of-failure


0%

10%

20%

30%

40%

50%

60%

70%

80%


0%

10%

20%

30%

40%

50%

60%

70%

80%


0%

10%

20%

30%

40%

50%

60%

70%

80%


0%

10%

20%

30%

40%

50%

60%

70%

80%


0%

10%

20%

30%

40%

50%

60%

70%

80%


0%

10%

20%

30%

40%

50%

60%

70%

80%


0%

10%

20%

30%

40%

50%

60%

70%

80%


0%

10%

20%

30%

40%

50%

60%

70%

80%


0%

10%

20%

30%

40%

50%

60%

70%

80%


0%

10%

20%

30%

40%

50%

60%

70%

80%


0%

10%

20%

30%

40%

50%

60%

70%

80%


0%

10%

20%

30%

40%

50%

60%

70%

80%


0%

10%

20%

30%

40%

50%

60%

70%

80%


0%

10%

20%

30%

40%

50%

60%

70%

80%


0%

10%

20%

30%

40%

50%

60%

70%

80%


0%

10%

20%

30%

40%

50%

60%

70%

80%

0%

10%

20%

30%

40%

50%

60%

70%

80%

Reduce 32 disk drives to 8! Reduce 32 fibre adapters to 4! Reduce 32 LAN adapters to 4! Reduce 128 cables to 16!

Virtualize & Consolidate

Eliminate underutilized adapters Reduce switches and cost of network operations and maintenance


Reduce cost of software licensing

0%

10%

20%

30%

40%

50%

60%

70%

80%

Reduce software costs by up to 80%! Reduce software support costs Reduce software subscription costs Costs are ongoing operational costs


0%

10%

20%

30%

40%

50%

60%

70%

80%


0%

10%

20%

30%

40%

50%

60%

70%

80%


0%

10%

20%

30%

40%

50%

60%

70%

80%


0%

10%

20%

30%

40%

50%

60%

70%

80%


0%

10%

20%

30%

40%

50%

60%

70%

80%


0%

10%

20%

30%

40%

50%

60%

70%

80%


0%

10%

20%

30%

40%

50%

60%

70%

80%


0%

10%

20%

30%

40%

50%

60%

70%

80%


0%

10%

20%

30%

40%

50%

60%

70%

80%


0%

10%

20%

30%

40%

50%

60%

70%

80%


0%

10%

20%

30%

40%

50%

60%

70%

80%


0%

10%

20%

30%

40%

50%

60%

70%

80%


0%

10%

20%

30%

40%

50%

60%

70%

80%


0%

10%

20%

30%

40%

50%

60%

70%

80%


0%

10%

20%

30%

40%

50%

60%

70%

80%


0%

10%

20%

30%

40%

50%

60%

70%

80%

Based on previous example of consolidating sixteen 4 core servers to one 12 core serverAssumes 2.65 consolidation factor and 2.0 performance ratio

Virtualization can generate significant software savings

Database software $40k per core!


Reduce cost with environmental efficiency

By consolidating 16 x16-core Sun V890s into ONE Power 570 system -- Save up to 92% in annual energy costs! -- Reduce floor space required by over 85% -- Reduce processing cores by over 87%

One Power 570 (@ 60% utilization)

256 total cores @ 2.1 GHz over 51 sq. ft. floor space up to 678 MWh annual energy

16 Sun V890s(@ 20% utilization)

32 total cores @ 4.2 GHz Only 7.6 sq. ft. floor space Save up to 629 MWh annually – up to $94k in energy savings per year!

See “Power 570 consolidation claim” in backup for full details.


Reduce cost: Administration of virtual workloads

New white paper from ITG (April 2009) provides comparative TCO data for typical workload consolidations in three industry segments – financial services, technology and government:

– PowerVM delivers better value than VMware, due to lower total administration costs


IMPROVE SERVICE by consolidating with Power Systems

Delivering flexible performance to tailor capacity based on service level agreements and business priorities

Supporting dynamic provisioning of virtual servers and resources to rapidly deploy new applications and infrastructure

Enabling clients to avoid disruption to deliver growth or perform system maintenance


Power Systems capacity is fluid and can be tailored to changing requirements

A partition size is not fixed and may be adjusted dynamically to match demand; resources added or moved are instantly available

Capacity on Demand offers insurance to efficiently respond to steady increases as well as unexpected spikes

Power Systems virtual server size is not limited by small physical system boundaries

Partition sizes can scale from a fraction of a processor to the full size of the system, dynamically

Unused capacity in one partition can be used by other workloads when required, improving their performance

Power Systems scale seamlessly to deliver balanced growth

Massive memory and I/O bandwidth enable linear performance as workload and capacity increase

Modular, Hot-node Add options enable smooth system growth

Improve service with Flexible Performance


Improve service with Dynamic Provisioning

Virtual servers can be quickly and easily deployed to accommodate new applications and respond to rapidly changing business needs

Virtual provisioning takes minutes and can even be done remotely

Virtual provisioning leverages existing resources and does not require installation of additional hardware

Workload Partitions can simplify provisioning even further

New workloads can be tested simply on virtual servers with minimal resources

Testing is accomplished in secure, isolated partitions that are created when needed and consume resource only when used

Virtual servers may be de-commissioned or re-used with minimal effort, easily recycling their resources

Resources are infinitely reusable and continuously shareable with all workloads, enabling high utilization and efficiency


Flexibility Factors VMware ESX 3.5(in VMware Infrastructure 3)

VMware ESX 4.0(in VMware vSphere 4)

PowerVM

Dynamic virtual CPU changes in VM No Add (but not Remove) Yes

Dynamic memory changes in VM No Add (but not Remove) Yes

Dynamic I/O device changes in VM No No Yes

Direct access to I/O devices from within VM No Some (with Nehalem) Yes

Maximum simultaneous live migrations 4 4 8

Improve service: PowerVM delivers superior flexibility

… to optimize IT resource utilization and boost responsiveness


Improve service by Avoiding Disruption

Enterprise Power Systems are designed for growth without application outage

Available system resources can be automatically increased without disruption to the application through PowerVM

New resources can be physically added to the server without downtime through Capacity on Demand and Hot-node Add

Applications can be moved to other systems without disruption using Live Partition Mobility and Live Application Mobility

Enterprise Power Systems are designed for concurrent service to avoid costly application downtime

System resources can be repaired while operations continue using Hot-node Repair

Operating systems and firmware can often be updated without disruption to operations

Pervasive redundancy, dynamic failover and extensive self-healing functions convert unplanned component failures into autonomic events

Failed processors and memory DIMMS can be automatically replaced by the system, maintaining application availability


Power Systems avoid disruption for our clients- Best availability according to Yankee group

Live Partition Mobility and Live Application Mobility for seamless upgrades and maintenance

Power Systems have exclusive features to keep you going even when something goes wrong– Bit Steering helps avoid double bit memory errors

– Alternate Processor Retry keeps applications running even when a processor has a hard failure

– I/O EEH to recover from transient PCI bus errors

New features reduce your risk even more– RAID controllers

– Virtual tape support

– Solid State Drives with fewer moving parts than disk

0

1

2

3

4

5

6

7

8

9

AIX HP-UX Solaris Windows

Hours of downtime per year*

According to a recent Yankee Group study* of 400 Windows, Linux and UNIX users, AIX

was the most reliable server operating system:

“IBM’s AIX achieved the highest level of reliability, with corporate enterprises

reporting an average of only 36 minutes of downtime per server in a 12-month period”

* Source: “Unix, Linux Uptime and Reliability Increase; Patch Management Woes Plague Windows” © 2008 Yankee Group Research, Inc. All rights reserved


POWER6 RAS – clearly the one to keep you in the GREEN

Feature POWER6 SPARC Integrity Xeon

Application/Partition RAS

Live Partition Mobility Yes No No Yes

Live Application Mobility Yes No No No

Partition Availability Priority Yes No No No

HW Isolated Partitions No Yes1 Yes1 No

System RAS

OS independent First Failure Data Capture Yes No No No

Redundant System Interconnect No Yes No No

Processor RAS

Processor Instruction Retry Yes Yes No No

Alternate Processor Recovery Yes No No No

Dynamic Processor Deallocation Yes Yes Yes No

Dynamic Processor Sparing Yes Yes2 Yes2 No

Memory RAS

Memory Keys Yes No No No

Chipkill™ Yes Yes Yes Yes

Redundant Memory Yes3 Yes3 Yes3 Yes3

I/O RAS

Extended Error Handling Yes No No No

#1,2,3 - See “POWER6 RAS” in backup; See the following URLs for addition details:http://www-03.ibm.com/systems/migratetoibm/systems/power/availability.htmlhttp://www-03.ibm.com/systems/migratetoibm/systems/power/virtualization.html


POWER6 Instruction Retry & Alternate Processor Recovery Industry leading hard- and soft- failure resilience Caches and interfaces are ECC protected and correct hard and soft fails Remaining circuits are protected by processor swap or instruction retry

Core restarts from last check point – Processor Instruction Retry

Intermittent Error Case

Hypervisor moves workload to an alternate core – Alternative Processor Recovery

Solid Error Case

Hard Error

Core architected state is check-pointed at every instruction completion Circuitry checked every cycle

Normal Operation

Instruction StateCheckpointGP1=0x14343433CT=-0x12344324...

RUCore 0

No Error

Soft Error

On Fault

Core 1

Availability POWER6 Improvements


MANAGE RISK by consolidating with Power Systems

Delivering Proven Security that allows you to stay connected with confidence

Offering Business Resiliency solutions that allow you to weather the unforeseen

Enabling Effective Systems Management to allow you to simplify your operations


Manage risk with Proven Security

Power Systems security offers assurance to aggressively pursue opportunities in today’s interconnected world IBM is a trusted, stable IT vendor with decades of experience with virtualization &

consolidation Built as a completely integrated solution, not a collection of products from three or more vendors

World-class operating system security with AIX and i Fewer avenues of entry

Recognized security certification for the whole platform, including PowerVM virtualization EAL4+ certification in place for POWER6 hardware and software

It is easier to secure & administer one enterprise server with virtualization than a collection of smaller servers Fewer operating systems to install security patches on Fewer systems to monitor for intrusion and compliance Eliminates need to synchronize security policies across multiple systems Reduced number of administrators and administrative touch points Eliminates domino effect of a single system with weak security falling and being

used to leverage escalation of privileges and node hopping


Manage risk: Virtualization security

VMware vSphere 4– Dependence on x86 architecture makes the

hypervisor vulnerable to an increasing number of published security exploits

– Intel/AMD virtualization support (VT) was a relatively recent addition, with known issues

– VMware recently gained CC EAL 4 certification, but this only relates to security of the virtual network traffic between VMs

PowerVM– Power (and System/z) architecture built from the

ground up for secure partitioning & virtualization– PowerVM has CC EAL 4+ certification

* For further details, read the System Virtualization Security white paper from the IBM SWG Competitive Projects Office (CPO).


Manage risk with Business Resiliency

Enterprise Power Systems are designed to minimize the risk of an unplanned outage

Enterprise Power Systems leverage superior quality components whenever feasible to minimize component failure

Enterprise Power Systems endure extensive system testing that includes PowerVM, AIX, and IBM i to ensure solid performance under pressure

Enterprise Power Systems utilize extensive predictive failure technologies to avoid impending component failures

Enterprise Power Systems enable memory keys to avoid inadvertent memory over-writes that lead to unexplained system crashes

Enterprise Power Systems inspire confidence to know that regardless what happens, your business will continue to operate

Disaster recovery becomes part of the plan, not just an afterthought

System failover becomes manageable with fewer pieces to manage

Supports regular and sustained role swap operations with PowerHA


Manage risk with Effective Systems Management

Simplify your operations by consolidating to one system

One system, one set of firmware, one set of virtualization software

Fewer cables, switches, routers to order, install, track, and maintain

Share operating system images with AIX Workload Partitions

Support multiple levels of AIX, Linux and IBM i if required

Simplify your operations with one consistent management tool

Let System Director be your view into your system and across your partitions, including energy management

Systems Director features are integrated with POWER6 and PowerVM and work across operating systems

Systems Director can even manage Windows, Linux, zLinux servers, and storage systems in the same consistent interface

Automate your IT environment by enabling your system to diagnosis problems

When components are internal to a single system, troubleshooting and repair times are reduced

Fewer networking and storage device interconnections simplifies monitoring and control of service offerings


The Challenge of Do-It-Yourself Assembly


The Advantage of Full Service

POWER6, Power Systems, PowerVM, AIX, IBM i

Technology and solutions that are:

Designed for each other

Architected together

Built together

Tested together

Delivered together

Supported together


Manage risk with security, control and simplicity

16 production entry servers, limited RAS 16 additional servers for HA/DR Multiple failover solutions 2X networking & increased operations costs Complex recovery – days or weeks

OR

Two enterprise Power Systems servers, extensive RAS Single, high-speed network connection Simple failover with PowerHA Regular role-swap ensures proven and rapid recovery of

all applications


“Nobody else had as eloquent and simple a design as IBM did. They also offered very cost-effective solutions.”

— Al Reineking, Executive Director of IT

Operations and Technical Services,

Baylor College of Medicine

Business challenge:

As a business with a mission-critical, 24x7 environment, Baylor College of Medicine needed an energy-smart infrastructure solution in a small footprint to support an SAP upgrade and provide an IT foundation for the organization’s new hospital.

Solution:

With the support of IBM Premier Business Partner Mark III and the IBM Migration Factory, Baylor migrated 32 legacy Sun servers onto three IBM Power™ 570 servers running IBM PowerVM™ on the IBM AIX® platform.

Benefits:

Improved overall performance by 30 percent

Reduced SAP environment footprint by 60 percent

Reduced cooling and power costs by more than 40 percent

Reduced server management and maintenance

Baylor College of Medicine

Migration with PowerVM enables significant SAP performance improvement


University of Pittsburgh Medical Center (UPMC) Rewriting the rules on IT investment to facilitate tomorrow’s healthcare innovations

Business ChallengeUPMC, Pennsylvania’s largest integrated healthcare delivery network, sought to lower the cost and complexity of IT infrastructure to enable the continued investment in next-generation clinical systems and to lay the foundation for the best possible patient care.

SolutionNow in the middle of a landmark, 8-year strategic partnership with IBM, UPMC is transforming its systems through consolidation, standardization and virtualization. Relying on IBM products and services, the mid-stream effort has already resulted in the reduction of hundreds of servers across the UPMC network and achieved more than a quantum improvement in resource efficiency. It has fundamentally changed the link between processing and resource needs — enabling it to meet an ambitious clinical agenda with a far lower rate of IT investment growth.

Benefits $30 million in capital and operating cost reductions 150 percent increase in processing capacity with no increase in IT support costs 40 percent reduction in IT infrastructure floor space requirements, freeing up space for revenue generating services 67 percent reduction in number of physical servers

“ Considering that IBM and UPMC are only midway through this trans-formation project, the results have been impressive. We have already proven that standardization, along with aggressive implementation of virtualization, yields unprecedented productivity and efficiency.”

– Paul Sikora, VP of IT Transformation, UPMC

Solution Components IBM BladeCenter® IBM Component Infrastructure Roadmap IBM Global Technology Services IBM Healthcare and Life Sciences IBM Research IBM STG Services IBM SWG Services IBM System p, System x, System z IBM Tivoli® product suite IBM TotalStorage® Enterprise Storage IBM WebSphere Business Integration IBM WebSphere® Application Server

© 2008 IBM Corporation


IBM Power Systems Success Stories Power technologies provide real business value

Annual savings of over $500,000 + accelerated new services deployment from 1 month to 2 days

Reduced application downtime + increased flexibility to dynamically scale and

change workload capacity

Fewer processor cores than previous Sun servers reduced software and

management costs + PowerVM simplified the effort to deploy or scale new services

Consolidated and virtualized 76 servers to 6 – reducing complexity and total cost of ownership

of global IT infrastructure


ROI Study Conclusions

The key findings of this ROI study show that respondent sites using IBM Power Systems and IBM System z realized the following benefits, based on the ROI analysis and in-depth interviews with respondents:

Consolidation and virtualization reduced annual IT infrastructure costs by $17,700 per 100 users per year.

More efficient server platforms increased the companywide average utilization rate capacity of a server utilized from 36% to 79%, compared with the previous deployment.

Cross-platform management improved application availability from 99.2% to 99.9% per year, leveraging high-availability, reliability, and management features that are built into the scalable server systems.

The use of fewer server footprints, made possible by the use of scalable servers, resulted in overall power/cooling savings of 50%.

In summary, the total annual value of the benefits came close to $30,000 per 100 users per year, based on the ROI analysis, and generated a three-year ROI of nearly 500%. The payback period for the investment, including hardware, software, consulting services, training, and IT staff to manage the server platforms was 6.3 months, following initial purchase and consolidation of Linux and open source workloads on the scalable IBM Power and IBM System z servers.

IDC: Adding Business Value with Linux Running on IBM Servers

IDC October 2008: http://www.ibm.com/linux/migrate.html

The payback period for the investment, including hardware, software, consulting services, training, and IT staff to manage

the server platforms was 6.3 months


REDUCE COST

– Resource sharing drives higher utilization rates and lowers costs

– Resource sharing reduces networking costs

– Resource sharing dramatically reduces energy consumption, floor space and software costs

0%

10%

20%

30%

40%

50%

60%

70%

80%

IMPROVE SERVICE

– Dynamically adjust partition resources to provide the performance your workloads require, when they need it

– Respond quickly and easily to changing business requirements– Avoid costly application outages and keep your business moving

MANAGE RISK

– Stay connected with confidence in today’s virtual world– Withstand the unforeseen, even in a turbulent time– Simplify your operations and your life

Consolidating with Enterprise Power Systems

enables clients to…


Backup


More PowerVM competitive resources

PPT: The Value of Consolidation on Enterprise Power Systems White Paper (ITG): Value Proposition for IBM Power Systems Report: System Virtualization Security Booklet: IBM Client Success with PowerVM


Additional Resources

IBM Power Platform Reliability, Availability, and Serviceability (RAS) - Highly Available IBM Power Systems Servers for Business-Critical Applications http://www.ibm.com/common/ssi/cgi-bin/ssialias?infotype=SA&subtype=WH&htmlfid=POW03003USEN&attachment=POW03003USEN.PDF&appname=STGE_PO_PO_USEN_WH

IBM EnergyScale for POWER6 Processor-Based Systems http://www.ibm.com/common/ssi/cgi-bin/ssialias?infotype=SA&subtype=WH&htmlfid=POW03002USEN&attachment=POW03002USEN.PDF&appname=STGE_PO_PO_USEN_WH

ITG: Value Proposition for IBM Power Systems - Virtualization Impact for Enterprise UNIX and Linux Server Infrastructureshttp://www.ibm.com/common/ssi/cgi-bin/ssialias?infotype=SA&subtype=WH&htmlfid=POL03019USEN&attachment=POL03019USEN.PDF&appname=STGE_PO_PO_USEN_CR

ITG: Value Proposition for IBM Power Systems - Platform Choices for the Enterprise SAP Infrastructurehttp://www.ibm.com/common/ssi/cgi-bin/ssialias?infotype=SA&subtype=WH&htmlfid=POL03022USEN&attachment=POL03022USEN.PDF&appname=STGE_PO_PO_USEN_CR

SAP® Application Server Consolidation on IBM System z™

http://www-03.ibm.com/support/techdocs/atsmastr.nsf/5cb5ed706d254a8186256c71006d2e0a/fe7d0390c72f4d0d862573b800268459/$FILE/SAP_Systemz_Sercon.pdf

IBM PowerVM Active Memory Sharing : An Overviewhttp://www.ibm.com/common/ssi/cgi-bin/ssialias?infotype=SA&subtype=WH&htmlfid=POW03026USEN&attachment=POW03026USEN.PDF&appname=STGE_PO_PO_USEN_WH


RAS Item Power 550 Power 560 Power 570 Power 595

Redundant / Hot Swap Fans & Blowers

Hot Swap DASD / Media / PCI Adapters

Concurrent Firmware Update

Redundant / Hot Swap Power Supplies

Dual disk controllers (split backplane)

Processor Instruction Retry / Alternate Processor Recovery

Storage Keys

PowerVM™/Live Partition Mobility/Live Application Mobility

Redundant Service Processors * *

Redundant System Clocks * *

Redundant / Hot Swap Power Regulators

Dynamic Processor Sparing

Memory Sparing

Hot GX Adapter Add and Cold Repair

Hot-node Add / Cold-node Repair *

Hot-node Repair / Hot-memory Add * *

POWER6 Enhanced Memory

Dynamic System Clock Failover

Hot-node Repair / Hot-memory Add for all nodes *

Mid-plane connection for inter-nodal communication

Power = Uptime + Control Optional

Standard

Not available

* Requires two or more nodes


Unified IBM Power Platform

Common RAS Architecture Mainframe-inspired system reliability and

resilience First Failure Data Capture Dynamic Processor Deallocation and

sparing Instruction Retry Resilient memory and error handling I/O PowerVM Live Partition Mobility and AIX

Live Application Mobility

Power 520

Power 570

Power Blades

Power 560

Binary compatibility

Virtualization Architecture

Power 595

Power 550


Power 520

Power 570

Power Blades

Power 560

Power 550

CPU

L1 Cache

L2 Cache

Memory

CPU

L1 Cache

L2 Cache

Memory

CPU

L1 Cache

L2 Cache

Memory

CPU

L1 Cache

L2 Cache

Memory

CPU

L1 Cache

L2 Cache

Memory

CPU

L1 Cache

L2 Cache

Memory

CPU

L1 Cache

L2 Cache

CPU

L1 Cache

L2 Cache

Disk

Error Checker

Strategy for Detecting Failures

Extensive checkers throughout hardware

Error Detection and Self-Healing– Available during run-time– At IPL Time– And after a critical fault

Error Checker

Power 595 (64-core)

~ 200,000 checkers (captured in)

> 73,000 FIR bits

Analyze failure data before service call and reliably identify failing component reducing costly downtime

Power 595 r-unit(64-core uses)

~ 179,000,000 transistors > 5.8M

Register bits

Fault Isolation Register (FIR)

ServiceServiceProcessorProcessor

NonvolatileRAM

Log Error


NonvolatileRAM

Log Error


NonvolatileRAM

Log Error

Power 595











Power 520

Power 570

Power Blades

Power 560



Power 595

Power 550Common RAS Architecture

+ Alternate Processor Recovery

Power 520

Power 570

Power 560

Power 595

Power 550







Power 520

Power 570

Power Blades

Power 560



Power 595


+ Alternate Processor Recovery

Power 520

Power 570

Power 560

Power 595


+ Alternate Processor Recovery+ Increased redundancy (availability)

• Clocks, Service Processors, …+ “Node” structure (enclosure/book)

• Hot node add (upgrade)• Hot node repair or upgrade

Power 570Power 595


Power 595 performance results

Source: http://www.spec.org All results as of 4/28/09. Not all results listed.

Source www.sap/com/benchmark/ All results as of 4/28/09. Not all results listed.

System Name Cores ChipsCores /

ChipThreads /

Core Peak Base PublishedIBM Power 595 64 32 2 2 2,155 1,866 April 2009HP rx2660 4 2 2 2 62.8 58.1 November 2007Sun T5240 16 2 8 8 157 142 April 2008


ChipThreads /

Core Peak Base PublishedIBM Power 595 64 32 2 2 2,184 1,681 April 2009HP rx2660 4 2 2 2 55.8 54.5 November 2007Sun T5240 16 2 8 8 119 111 April 2008


ChipThreads /

Core bopsbops / JVM Published

IBM Power 595 64 32 2 2 3,435,485 107,359 April 2008HP rx2660 4 2 2 2 80,884 80,884 February 2007Sun T5240 16 2 8 8 384,934 24,058 November 2008

SPECint_rate2006 Results

SPECfp_rate2006 Results

SPECjbb2005 Results

Technology Partner System

Number of Benchmark

UsersOperating System

Average Dialog

Response Time (sec)

Dialog Steps Per

Hour SAPS

Fully Processed Line Items Per Hour Cores Chips Threads

Certification Number

IBM Power 595 35,400 AIX 6.1 1.94 10,677,000 177,950 3,559,000 64 32 128 2008019HP rx2660 1,090 HP-UX 11i 1.93 329,000 5,480 109,670 4 2 8 2007016Sun Microsystems T5240 4,170 Solaris 10 1.97 1,254,000 20,900 418,000 16 2 128 2008021

SAP SD 2-tier Benchmark Results


Power 570 consolidation claim

Comparisons between the IBM Power 570 and the Sun Fire V890.– The virtualized system count and energy savings were derived from several factors: A performance factor of

5.67X was applied to the virtualization scenario based on SPEC® results source: www.spec.org . Power 570 (32-core, 16 chips, 2 cores per chip, 4.2 GHz) SPECjbb2005 1,390,087 bops, 86,880 bops/JVM as of 4/28/2009; Sun Fire V890 (16-core, 8 chips, 2 cores per chip) 2.1 GHz, SPECjbb2005 244,846 bops, 30,606 bops/JVM as of 4/28/2009. A virtualization factor of 3X was applied to the virtualization scenario using utilizations derived from studies conducted by Alinean available at http://www-935.ibm.com/services/us/cio/optimize/opt_wp_ibm_systemp.pdf. Power consumption figures of 5600 W for the IBM Power 570 and 4843 W for the Sun Fire V890 were based on the maximum rates published by IBM and Sun Microsystems, respectively. Air conditioning power requirement estimated at 50% of system power requirement. Energy cost of $.1003 per kWh is based on 2009 YTD US Average Retail price to commercial customers per US DOE at http://www.eia.doe.gov/cneaf/electricity/epm/table5_6_b.html as of 4/28/2009. The reduction in floor space, power, cooling and software costs depends on the specific customer, environment, application requirements, and the consolidation potential. Actual numbers of virtualized systems supported will depend on workload levels for each replaced system.

– Systems were compared based on maximum processor configurations unless recommended wattage was available for other configurations. Maximum configurations were used because that is the data point for which power requirements are defined. Other configurations of these systems could have different performance per watt metrics.

– Performance/watt is calculated by dividing the performance metric by the recommended maximum power usage for site planning. Actual power used by the systems will be less than this value for both of the systems.

– This information for the Power 570 is in "Model 9117-MMA server specifications" available at http://publib.boulder.ibm.com/infocenter/systems/scope/hw/topic/iphdx/sa76-0091.pdf. The power requirement for the Power 570 is 5600 watts.

– The information for the Sun Fire V890 Server is in the “Sun Fire™ V490/V890 Servers with UltraSPARC IV+ 2100MHz CPU/Memory Modules Supplement ” available at http://dlc.sun.com/pdf//820-0714-10/820-0714-10.pdf. The power requirement for the V890 is 4,843 watts.

– All data is current as of 4/28/2009


Virtual networking claim

Latency = TCP_RR, (TCP session performing 1 byte request/response test) performed on a Power 570 4.7GHz running AIX 5.3.10.0 comparing two 1-core LPARS communicating via virtual ethernet vs. two #5769 10GigE adapters communicating via a Force 10 S25N Data Center switch.

Virtual Ethernet #5769 via switch

Sessions Transactions/s

(half-round trip latency in microseconds)

1 24,378.8 (20.5 us) 16,300.5 (30.7 us)

Throughput = TCP_STREAM performed on a Power 570 4.7 GHz running AIX 5.3.10.0 comparing two 4-core LPARS communicating via virtual ethernet vs. two IVE/HEA adapters communicating via a Force 10 S25N Data Center switch

Measures application payload data rate (performance a user socket application would see)

Virtual Ethernet IVE/HEA via switch

Sessions Throughput

(Mbps)

1 15,193.76 8,7660.91

2 24,395.51 7,858.62

4 27,808.34 9,342.14

8 30,294.68 9,410.65


POWER6 RAS1 Limited to HW partitions at the system board or cell board level. Sun Dynamic System Domains with use the Quad XSB are not hardware isolated.

2 Analysis based on features described in: "IBM Power Platform Reliability, Availability, and Serviceability (RAS), October 2008, available at http://www-03.ibm.com/systems/power/hardware/whitepapers/ras.html; "High Availability in the Datacenter, SUN SPARC Enterprise Servers, April, 2007, available at www.sun.com, and “Meet the HP Integrity Superdome Server with the HP Super-Scalable Processor Chipset sx2000” and “Intel Itanium-based midrange servers from HP - The HP Integrity rx7640 and rx8640 Servers” available through www.hp.com. The Xeon features are described in specifications for Dell PowerEdge, HP ProLiant, and Sun x64 systems available at the respective Web sites: www.dell.com, www.hp.com, and www.sun.com. The Dell whitepaper “Dell™ PowerEdge™ Servers, 2009 – Memory” describing memory capability available at www.dell.com , and the Sun Whitepaper, “SUN FIRE™ X4170, X4270, AND X4275 SERVER ARCHITECTURE” available at www.sun.com were also used as sources. The actual names of features vary between vendors. This comparison is based on the description of the features. Only features where there is a significant difference between vendors are included in the table. All vendors provide many more features such as redundant power, cooling, etcetera. All data is current as of April 21, 2009.

3 Power uses Redundant Bit Steering which uses spare memory chips to replace failing bits. These spare chips are part of the standard memory system design. Integrity replaces failing pages with spare memory. Clients must purchase adequate spare memory for this feature to work. SPARC and Xeon systems uses optional mirroring. Customer must purchase two times the memory they need for production to use this feature.

4 SPARC and Integrity automatic recovery is limited. Alternative processor must be in the hardware partition or in HP's implementation must be in the partition or in a cell board that is not allocated to a partition. The Power system implementation allows any processor that is not currently allocated to be used.

5. See the following URLs for additional detailshttp://www-03.ibm.com/systems/migratetoibm/systems/power/availability.html http://www-03.ibm.com/systems/migratetoibm/systems/power/virtualization.html


Revised January 9, 2003

REQUIRED CHARTSThe presenter must display the Special Notices chart, the Notes on Benchmarks and Values charts (if the referenced

values are given), and the Notes on Performance Estimates chart (if the referenced performance estimates are given) during the course of the presentation. Any printed copies of this presentation that are distributed must include legible copies of these charts. If printed copies are not distributed, the attendees must be offered the option to receive legible printed copies of these charts.

TRADEMARKSPlease review the Special Notices page prior to updating this presentation to ensure all trademarks used are given

proper attribution.

SPEAKER NOTESThis presentation may contain speaker notes available imbedded or as a separate file. Please ensure these are utilized

if available.

Notes to presenter


This document was developed for IBM offerings in the United States as of the date of publication. IBM may not make these offerings available in other countries, and the information is subject to change without notice. Consult your local IBM business contact for information on the IBM offerings available in your area.

Information in this document concerning non-IBM products was obtained from the suppliers of these products or other public sources. Questions on the capabilities of non-IBM products should be addressed to the suppliers of those products.

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

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

The information contained in this document has not been submitted to any formal IBM test and is provided "AS IS" with no warranties or guarantees either expressed or implied.

All examples cited or described in this document are presented as illustrations of the manner in which some IBM products can be used and the results that may be achieved. Actual environmental costs and performance characteristics will vary depending on individual client configurations and conditions.

IBM Global Financing offerings are provided through IBM Credit Corporation in the United States and other IBM subsidiaries and divisions worldwide to qualified commercial and government clients. Rates are based on a client's credit rating, financing terms, offering type, equipment type and options, and may vary by country. Other restrictions may apply. Rates and offerings are subject to change, extension or withdrawal without notice.

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

All prices shown are IBM's United States suggested list prices and are subject to change without notice; reseller prices may vary.

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

Any performance data contained in this document was determined in a controlled environment. Actual results may vary significantly and are dependent on many factors including system hardware configuration and software design and configuration. Some measurements quoted in this document may have been made on development-level systems. There is no guarantee these measurements will be the same on generally-available systems. Some measurements quoted in this document may have been estimated through extrapolation. Users of this document should verify the applicable data for their specific environment.

Revised September 26, 2006

Special notices


IBM, the IBM logo, ibm.com AIX, AIX (logo), AIX 6 (logo), AS/400, BladeCenter, Blue Gene, ClusterProven, DB2, ESCON, i5/OS, i5/OS (logo), IBM Business Partner (logo), IntelliStation, LoadLeveler, Lotus, Lotus Notes, Notes, Operating System/400, OS/400, PartnerLink, PartnerWorld, PowerPC, pSeries, Rational, RISC System/6000, RS/6000, THINK, Tivoli, Tivoli (logo), Tivoli Management Environment, WebSphere, xSeries, z/OS, zSeries, AIX 5L, Chiphopper, Chipkill, Cloudscape, DB2 Universal Database, DS4000, DS6000, DS8000, EnergyScale, Enterprise Workload Manager, General Purpose File System, , GPFS, HACMP, HACMP/6000, HASM, IBM Systems Director Active Energy Manager, iSeries, Micro-Partitioning, POWER, PowerExecutive, PowerVM, PowerVM (logo), PowerHA, Power Architecture, Power Everywhere, Power Family, POWER Hypervisor, Power Systems, Power Systems (logo), Power Systems Software, Power Systems Software (logo), POWER2, POWER3, POWER4, POWER4+, POWER5, POWER5+, POWER6, System i, System p, System p5, System Storage, System z, Tivoli Enterprise, TME 10, Workload Partitions Manager and X-Architecture are trademarks or registered trademarks of International Business Machines Corporation in the United States, other countries, or both. If these and other IBM trademarked terms are marked on their first occurrence in this information with a trademark symbol (® or ™), these symbols indicate U.S. registered or common law trademarks owned by IBM at the time this information was published. Such trademarks may also be registered or common law trademarks in other countries. A current list of IBM trademarks is available on the Web at "Copyright and trademark information" at www.ibm.com/legal/copytrade.shtml

The Power Architecture and Power.org wordmarks and the Power and Power.org logos and related marks are trademarks and service marks licensed by Power.org.UNIX is a registered trademark of The Open Group in the United States, other countries or both. Linux is a registered trademark of Linus Torvalds in the United States, other countries or both.Microsoft, Windows and the Windows logo are registered trademarks of Microsoft Corporation in the United States, other countries or both.Intel, Itanium, Pentium are registered trademarks and Xeon is a trademark of Intel Corporation or its subsidiaries in the United States, other countries or both.AMD Opteron is a trademark of Advanced Micro Devices, Inc.Java and all Java-based trademarks and logos are trademarks of Sun Microsystems, Inc. in the United States, other countries or both. TPC-C and TPC-H are trademarks of the Transaction Performance Processing Council (TPPC).SPECint, SPECfp, SPECjbb, SPECweb, SPECjAppServer, SPEC OMP, SPECviewperf, SPECapc, SPEChpc, SPECjvm, SPECmail, SPECimap and SPECsfs are trademarks of the Standard Performance Evaluation Corp (SPEC).NetBench is a registered trademark of Ziff Davis Media in the United States, other countries or both.AltiVec is a trademark of Freescale Semiconductor, Inc.Cell Broadband Engine is a trademark of Sony Computer Entertainment Inc.InfiniBand, InfiniBand Trade Association and the InfiniBand design marks are trademarks and/or service marks of the InfiniBand Trade Association. Other company, product and service names may be trademarks or service marks of others.

Revised April 24, 2008

Special notices (cont.)


The IBM benchmarks results shown herein were derived using particular, well configured, development-level and generally-available computer systems. Buyers should consult other sources of information to evaluate the performance of systems they are considering buying and should consider conducting application oriented testing. For additional information about the benchmarks, values and systems tested, contact your local IBM office or IBM authorized reseller or access the Web site of the benchmark consortium or benchmark vendor.

IBM benchmark results can be found in the IBM Power Systems Performance Report at http://www.ibm.com/systems/p/hardware/system_perf.html .

All performance measurements were made with AIX or AIX 5L operating systems unless otherwise indicated to have used Linux. For new and upgraded systems, AIX Version 4.3, AIX 5L or AIX 6 were used. All other systems used previous versions of AIX. The SPEC CPU2006, SPEC2000, LINPACK, and Technical Computing benchmarks were compiled using IBM's high performance C, C++, and FORTRAN compilers for AIX 5L and Linux. For new and upgraded systems, the latest versions of these compilers were used: XL C Enterprise Edition V7.0 for AIX, XL C/C++ Enterprise Edition V7.0 for AIX, XL FORTRAN Enterprise Edition V9.1 for AIX, XL C/C++ Advanced Edition V7.0 for Linux, and XL FORTRAN Advanced Edition V9.1 for Linux. The SPEC CPU95 (retired in 2000) tests used preprocessors, KAP 3.2 for FORTRAN and KAP/C 1.4.2 from Kuck & Associates and VAST-2 v4.01X8 from Pacific-Sierra Research. The preprocessors were purchased separately from these vendors. Other software packages like IBM ESSL for AIX, MASS for AIX and Kazushige Goto’s BLAS Library for Linux were also used in some benchmarks.

For a definition/explanation of each benchmark and the full list of detailed results, visit the Web site of the benchmark consortium or benchmark vendor.

TPC http://www.tpc.org SPEC http://www.spec.org LINPACK http://www.netlib.org/benchmark/performance.pdf Pro/E http://www.proe.com GPC http://www.spec.org/gpc VolanoMark http://www.volano.com STREAM http://www.cs.virginia.edu/stream/ SAP http://www.sap.com/benchmark/ Oracle Applications http://www.oracle.com/apps_benchmark/ PeopleSoft - To get information on PeopleSoft benchmarks, contact PeopleSoft directly Siebel http://www.siebel.com/crm/performance_benchmark/index.shtm Baan http://www.ssaglobal.com Fluent http://www.fluent.com/software/fluent/index.htm TOP500 Supercomputers http://www.top500.org/ Ideas International http://www.ideasinternational.com/benchmark/bench.html Storage Performance Council http://www.storageperformance.org/results

Revised March 12, 2009

Notes on benchmarks and values


Revised March 12, 2009

Notes on HPC benchmarks and valuesThe IBM benchmarks results shown herein were derived using particular, well configured, development-level and generally-available computer systems. Buyers should consult other sources of information to evaluate the performance of systems they are considering buying and should consider conducting application oriented testing. For additional information about the benchmarks, values and systems tested, contact your local IBM office or IBM authorized reseller or access the Web site of the benchmark consortium or benchmark vendor.

IBM benchmark results can be found in the IBM Power Systems Performance Report at http://www.ibm.com/systems/p/hardware/system_perf.html .

All performance measurements were made with AIX or AIX 5L operating systems unless otherwise indicated to have used Linux. For new and upgraded systems, AIX Version 4.3 or AIX 5L were used. All other systems used previous versions of AIX. The SPEC CPU2000, LINPACK, and Technical Computing benchmarks were compiled using IBM's high performance C, C++, and FORTRAN compilers for AIX 5L and Linux. For new and upgraded systems, the latest versions of these compilers were used: XL C Enterprise Edition V7.0 for AIX, XL C/C++ Enterprise Edition V7.0 for AIX, XL FORTRAN Enterprise Edition V9.1 for AIX, XL C/C++ Advanced Edition V7.0 for Linux, and XL FORTRAN Advanced Edition V9.1 for Linux. The SPEC CPU95 (retired in 2000) tests used preprocessors, KAP 3.2 for FORTRAN and KAP/C 1.4.2 from Kuck & Associates and VAST-2 v4.01X8 from Pacific-Sierra Research. The preprocessors were purchased separately from these vendors. Other software packages like IBM ESSL for AIX, MASS for AIX and Kazushige Goto’s BLAS Library for Linux were also used in some benchmarks.

For a definition/explanation of each benchmark and the full list of detailed results, visit the Web site of the benchmark consortium or benchmark vendor.SPEC http://www.spec.org LINPACK http://www.netlib.org/benchmark/performance.pdf Pro/E http://www.proe.com GPC http://www.spec.org/gpc STREAM http://www.cs.virginia.edu/stream/ Fluent http://www.fluent.com/software/fluent/index.htm TOP500 Supercomputers http://www.top500.org/ AMBER http://amber.scripps.edu/ FLUENT http://www.fluent.com/software/fluent/fl5bench/index.htm GAMESS http://www.msg.chem.iastate.edu/gamess GAUSSIAN http://www.gaussian.com ANSYS http://www.ansys.com/services/hardware-support-db.htm

Click on the "Benchmarks" icon on the left hand side frame to expand. Click on "Benchmark Results in a Table" icon for benchmark results.ABAQUS http://www.simulia.com/support/v68/v68_performance.php ECLIPSE http://www.sis.slb.com/content/software/simulation/index.asp?seg=geoquest& MM5 http://www.mmm.ucar.edu/mm5/ MSC.NASTRAN http://www.mscsoftware.com/support/prod%5Fsupport/nastran/performance/v04_sngl.cfm STAR-CD www.cd-adapco.com/products/STAR-CD/performance/320/index/html NAMD http://www.ks.uiuc.edu/Research/namd HMMER http://hmmer.janelia.org/

http://powerdev.osuosl.org/project/hmmerAltivecGen2mod


Revised April 2, 2007

Notes on performance estimatesrPerf for AIX

rPerf (Relative Performance) is an estimate of commercial processing performance relative to other IBM UNIX systems. It is derived from an IBM analytical model which uses characteristics from IBM internal workloads, TPC and SPEC benchmarks. The rPerf model is not intended to represent any specific public benchmark results and should not be reasonably used in that way. The model simulates some of the system operations such as CPU, cache and memory. However, the model does not simulate disk or network I/O operations.

rPerf estimates are calculated based on systems with the latest levels of AIX and other pertinent software at the time of system announcement. Actual performance will vary based on application and configuration specifics. The IBM eServer pSeries 640 is the baseline reference system and has a value of 1.0. Although rPerf may be used to approximate relative IBM UNIX commercial processing performance, actual system performance may vary and is dependent upon many factors including system hardware configuration and software design and configuration. Note that the rPerf methodology used for the POWER6 systems is identical to that used for the POWER5 systems. Variations in incremental system performance may be observed in commercial workloads due to changes in the underlying system architecture.

All performance estimates are provided "AS IS" and no warranties or guarantees are expressed or implied by IBM. Buyers should consult other sources of information, including system benchmarks, and application sizing guides to evaluate the performance of a system they are considering buying. For additional information about rPerf, contact your local IBM office or IBM authorized reseller.

========================================================================

CPW for IBM i

Commercial Processing Workload (CPW) is a relative measure of performance of processors running the IBM i operating system. Performance in customer environments may vary. The value is based on maximum configurations. More performance information is available in the Performance Capabilities Reference at: www.ibm.com/systems/i/solutions/perfmgmt/resource.html

© 2009 ibm corporation the value of consolidation with enterprise power systems name of the...

Documents