Power a Smarter Planet with POWER7 - e-TechServices · IBM Power Systems Power a Smarter Planet...
Transcript of Power a Smarter Planet with POWER7 - e-TechServices · IBM Power Systems Power a Smarter Planet...
© 2010 IBM Corporation
IBM Power Systems
Power a Smarter Planet with POWER7Smarter Systems for a Smarter Planet
John SheehySystems [email protected]
© 2010 IBM Corporation
IBM Power Systems
2
IBMVision
Designed, integrated systems are part of the transformational story
of the next decade.
© 2010 IBM Corporation
IBM Power Systems
3
IBM Systems & Technology lay the foundation
Workload Optimization
Deep Understanding of Client Needs
Approach challenge from the client’s perspective
A Comprehensive PortfolioA family of “fit-for-purpose” servers, storage & system software
Technology LeadershipInvesting in future technology for differentiation and sustained
leadership
© 2010 IBM Corporation
IBM Power Systems
4
Smarter Money
Power Systems performance, security and availability are capabilities that provide the world’s largest banks with the ability to move today’s money - intangible, invisible information - from a paycheck to a bank to a retailer and back into another business account.
Smarter Cities
Cities large and small depend on the ability of Power Systems to sift through the data needed to not only solve crimes and respond to emergencies, but to help prevent them. Power Systems help manage traffic, share information across city agencies, keep citizens informed and give them access to services.
Smarter TelecomTelcos are using Power Systems to deliver new services dynamically to an exploding number of devices - and Power’s scalability means that new services can be added quickly, new clients can be billed accurately, and costs can be reduced with consolidation.
© 2010 IBM Corporation
IBM Power Systems
5
Organizations are moving from “what” to “how”
How do I infuse intelligence into a system for which no one enterprise or agency is responsible?
How do I bring all the necessary constituents together?
How do I make the case for budget?
How do I get a complex solution through procurement?
How do I coalesce support with citizens?
Where should I start?
How fast should I move?
“
”
© 2010 IBM Corporation
IBM Power Systems
6
Transformations to “smarter” solutions require smarter systems that:
Scale quickly and efficiently
Optimize workload performance
Flexibly flow resources
Avoid downtime
Save energy
Automate management tasks
© 2010 IBM Corporation
IBM Power Systems
8
Source: IDC Quarterly Server Tracker Q309 release, November 2009
UNIX Server Rolling Four Quarter Average Revenue Share
POWER4Dynamic LPARsDynamic LPARs
POWER6Live Partition Live Partition
MobilityMobilityPOWER5
Micro-PartitioningMicro-Partitioning
Customers are moving to higher value…as shown by the largest shift of customer spending in UNIX History
© 2010 IBM Corporation
IBM Power Systems
9
successful Power Migration Factory migrations to date. There were over 500 Power migrations during 2009, with more than 90% from Sun and HP customers (including x86 consolidation). In 4Q09 alone, Power achieved nearly 200 competitive migrations.
2,100
© 2010 IBM Corporation
IBM Power Systems
•POWER7 Processor
•POWER7 Servers–Power 750
–Power 755
–Power 770
–Power 780
•Active Memory Expansion
•Upgrades
•I/O Update
Agenda….
© 2010 IBM Corporation
IBM Power Systems
•Balance System Design
– Cache, Memory, and IO
•POWER7 Processor Technology
– 6th Implementation of multi-core design
– On chip L2 & L3 caches
•POWER7 System Architecture
– Blades to High End offerings
– Enhances memory implementation
– PCIe, SAS / SATA
•Built in Virtualization
– Memory Expansion
– VM Control
•Green Technologies
– Processor Nap & Sleep Mode
– Memory Power Down support
– Aggressive Power Save / Capping Modes
•Availability
– Processor Instruction Retry
– Alternate Process Recovery
– Concurrent Add & Services
POWER7 System Highlights
© 2010 IBM Corporation
IBM Power Systems
Operating Systems
Power Systems Portfolio (Feb 2010)Major Features: Modular systems with linear scalability PowerVM Virtualization Physical and Virtual Management Roadmap to Continuous Availability Binary Compatibility Energy / Thermal Management
Power 755
Power 750
520
BladeCenter JS12 / JS22JS23 / JS43
595
575
Power 770
Power 780
© 2010 IBM Corporation
IBM Power Systems
POWER7 Systems Technology Value…•Technology
–Roadmap
–Processor Instruction Retry
–Green Technology built in
–Common architecture from Blades to High-end
•Performance
–Power Systems scalability from blades to high end systems
–Performance leadership in a variety of workloads
–Best Performance per core
–Memory and IO bandwidth
•Virtualization
–Consolidate to higher levels
–Virtualize Processors, Memory, and I/O
–Dynamic movement of Partitions and Applications
–Reduce infrastructure costs
•RAS
–Power Systems mainframe inspired RAS features
–Hot Add support / Concurrent Maintenance
–Alternate Process Recovery
–Operating Systems Availability Leadership
Hypervisor
Virt I/O ServerShared I/O
Single SMP Hardware System
© 2010 IBM Corporation
IBM Power Systems
Processor Technology Roadmap
2001
Dual Core Chip Multi Processing Distributed Switch Shared L2 Dynamic LPARs (32)
2004
Dual CoreEnhanced ScalingSMTDistributed Switch +Core Parallelism +FP Performance +Memory bandwidth +Virtualization
2007
Dual Core High Frequencies Virtualization + Memory Subsystem + Altivec Instruction Retry Dyn Energy Mgmt SMT + Protection Keys
2010
Multi Core On-Chip eDRAM Power Optimized Cores Mem Subsystem ++ SMT++ Reliability + VSM & VSX (AltiVec) Protection Keys+
POWER8
Concept Phase
POWER4180 nm
POWER5130 nm
POWER665 nm
POWER745 nm
© 2010 IBM Corporation
IBM Power Systems
POWER5 POWER5+ POWER6 POWER7
Technology 130 nm 90 nm 60 nm 45 nm
Size 389 mm2 245 mm2 341 mm2 567 mm2
Transistors 276 M 276 M 790 M 1.2 B
Cores 2 2 2 4 / 6 / 8
Frequencies 1.65 GHz 1.9 GHz 3-5 GHz 3-4 GHz
L2 Cache 1.9 MB Shared 1.9 MB Shared 4 MB / Core 256 KB / Core
L3 Cache 36 MB 36 MB 32 MB 4 MB / Core
Memory Cntrl 1 1 2 / 1 2
LPAR 10 / Core 10 / Core 10 / Core 10 / Core
Processor Designs
© 2010 IBM Corporation
IBM Power Systems
POWER7 Processor Chip •Cores : 8 ( 4 / 6 core options )
•567mm2 Technology: – 45nm lithography, Cu, SOI, eDRAM
•Transistors: 1.2 B– Equivalent function of 2.7B
– eDRAM efficiency
•Eight processor cores– 12 execution units per core
– 4 Way SMT per core – up to 4 threads per core
– 32 Threads per chip
– L1: 32 KB I Cache / 32 KB D Cache
– L2: 256 KB per core
– L3: Shared 32MB on chip eDRAM
•Dual DDR3 Memory Controllers– 90 GB/s Memory bandwidth per chip
•Scalability up to 32 Sockets– 360 GB/s SMP bandwidth/chip
– 20,000 coherent operations in flight
Binary Compatibility with POWER6
POWER7CORE
L2 Cache
POWER7CORE
L2 Cache
POWER7CORE
L2 Cache
POWER7CORE
L2 Cache
POWER7CORE
L2 Cache
POWER7CORE
L2 Cache
POWER7CORE
L2 Cache
POWER7CORE
L2 Cache
L3 Cache and Chip Interconnect
MC1MC0
Local SMP Links
Remote SMP & I/O Links
FAST
L3 REGION
© 2010 IBM Corporation
IBM Power Systems
eDRAM technologyIBM’s eDRAM technology benefits: Greater density, Less power requirements, Fewer soft errors, and Better performance
Enables POWER7 to provide 32MB of internal L3 Cache
L3 Cache critical to balanced design / performance: 6:1 Latency improvement for L3 accesses vs external L3 2X Bandwidth improvement with on chip interconnect. 32B busses to and from each core
No off chip driver or receivers in L3 access path.
eDRAM is nearly as fast as conventional SRAM but requires far less space
1/3 the space of conventional 6T SRAM implementation 1/5 the standby power
Soft Error Rate 250x lower than SRAM ( Better availability ) 1.5 Billion reduction in transistors
IBM is effectively doubling microprocessor performance beyond what classical scaling alone can achieve,” said Dr. Subramanian Iyer, DE (Distinguished Engineer) DT
EDRAM Cell
© 2010 IBM Corporation
IBM Power Systems
POWER7 Core64-bit PowerPC architecture v2.07
Execution Units• 2 Fixed Point Units• 2 Load Store Units• 4 Double Precision Floating Point Units• 1 Branch• 1 Condition Register• 1 Vector Unit• 1 Decimal Floating Point Unit• 6 Wide Dispatch• Units include distributed Recovery Function
Out of Order Execution
Modes: POWER6, POWER6+ and POWER7L2 Cache
IFUCRU/BRU
ISU
DFU
FXU
VSXFPU
LSU
• POWER7 continues to support VMX / Extends SIMD support with VSX– 2 VSX units that can each handle 2 Double-Precision FP instructions
– 8 FLOPS per cycles
– VSX units can also handle 4 Single Precision instructions per cycle
– VSX instruction set support for vector and scalar instructions
© 2010 IBM Corporation
IBM Power Systems
AIX
Release/TL
Max Cores
& Threads Supported
POWER6 Mode
POWER7 Mode
AIX 5.3
(All TLs Supported
64 / 128 N/A
AIX 6.1 TL2, TL3
64 / 128 N/A
AIX 6.1 TL4 64 / 128 64 / 256
AIX 6.1 TL5 64 / 128 64 / 256
AIX 7.1 64 / 128 256 / 1024
POWER7 Modes: IBM i and AIX
IBM i
Release
Max Cores
& Threads Supported
POWER6 Mode
POWER7 Mode
IBM i 6.1 32 / 64 32 / 128
Special Support
64/128 32 / 128
IBM i 7.1 32 / 64 32 / 128
Special Support
64/128 64 / 256
© 2010 IBM Corporation
IBM Power Systems
21
Linux Max Processors
& Threads Supported
POWER6 Mode POWER7 Mode
RHEL 5 ( Updates newer than U4) 64 / 128 N/A
SLES 10 SP# and newer 64 / 128 N/A
SLES 11 ( All Service Packs) 64 / 128 256 / 1024
RHEL 6 (Next major RHEL version) 64 / 128 256 / 1024
POWER7 Modes: Linux
© 2010 IBM Corporation
IBM Power Systems
POWER7 Design
Physical Design:• 8 cores with integrated cache and
memory controllers• 4 / 6 / 8 Core options• 45nm technology
Features:• 4th Generation SMP Fabric Bus• 3rd Generation Multi-Threading• New Power Bus• Energy Optimized Design• Multiple Memory Controllers• DDR3 memory support• Enhanced GX System Buses• On-Chip L2/L3 Cache• eDRAM L3 Cache• Industry Standard IO
Core
L2
Core
L2
Memory Interface
Core
L2
Core
L2
Core
L2
Core
L2
Core
L2
Core
L2
GX
POWER
BUS
SMP
FABRIC
L3 Cache
© 2010 IBM Corporation
IBM Power Systems
Transition from POWER6POWER6
Memory+
GX+ Bridge
Memory+
GX Bus Cntrl Mem
ory
Cn
trl
Mem
ory
Cn
trl
Fabric BusController
CoreAltiVec
L3Ctrl L3
L3CtrlL3
CoreAltiVec
4 MB L2
4 MB L2
Core
L2
Core
L2
Memory Interface
Core
L2
Core
L2
Core
L2
Core
L2
Core
L2
Core
L2
GX
SMP
FABRIC
POWER
BUS
POWER7
Memory++
L3 Cache
© 2010 IBM Corporation
IBM Power Systems
Memory Channel Bandwidth Evolution
DDR2 @ 553 / 667 MHzEffective Bandwidth:
2.6 GB/sec
DDR3 @ 1066 MHzEffective Bandwidth:
6.4 GB/sec
DDR2 @ 553 MHzEffective Bandwidth:
1.1 GB/s
POWER5 POWER6 POWER7
Memory Performance:
2x DIMM
Memory Performance:
4x DIMM
Memory Performance:
6x DIMM
DDR3
DDR3
DDR3
DDR3
DDR3
DDR3
DDR3
DDR3
DDR3
DDR3
© 2010 IBM Corporation
IBM Power Systems
Multi-threading Evolution
Thread 1 ExecutingThread 0 ExecutingNo Thread Executing
FX0FX1FP0FP1LS0LS1BRXCRL
Single thread Out of Order
FX0FX1FP0FP1LS0LS1BRXCRL
S80 Hardware Multi-thread
FX0FX1FP0FP1LS0LS1BRXCRL
POWER5 2 Way SMT
FX0FX1FP0FP1LS0LS1BRXCRL
POWER7 4 Way SMT
Thread 3 ExecutingThread 2 Executing
© 2010 IBM Corporation
IBM Power Systems
POWER7 TurboCore Mode
• TurboCore Chips: 4 available cores
• Aggregation of L3 Caches of unused cores.
•TurboCore chips have a 2X the L3 Cache per Chip available
–4 TurboCore Chips L3 = 32 MB
• Performance gain over POWER6.–Provides up to 1.5X per core to core
• Chips run at higher frequency:–Power reduction of unused cores.
• With “Reboot”, System can be reconfigured to 8 core mode.
–ASM Menus
Unused CoreTurboCores
Core
L2
Core
L2
Memory Interface
Core
L2
Core
L2
Core
L2
Core
L2
Core
L2
Core
L2
GX
SMP
FABRIC
POWER
BUS
32 MB L3 Cache
POWER7 Chip
Power 780 TurboCore Chip
© 2010 IBM Corporation
IBM Power Systems
Core
L2
Core
L2
Memory Interface
Core
L2
Core
L2
Core
L2
Core
L2
Core
L2
Core
L2
GX
SMP
FABRIC
POWER
BUS
24 MB L3 Cache
POWER7 Core / Cache options
6-Core Chip
Power 750 & Power 770
© 2010 IBM Corporation
IBM Power Systems
POWER7 SMT4• Standard Cache Option
• All cores active•Requires POWER7 Mode
– POWER6 Mode supports SMT1 and SMT2
•Operating System Support– AIX 6.1 and AIX 7.1
– IBM i 6.1 and 7.1
– Linux
•Dynamic Runtime SMT scheduling– Spread work among cores to execute in
appropriate threaded mode
– Can dynamical shift between modes as required: SMT1 / SMT2 / SMT4
•LPAR-wide SMT controls– ST, SMT2, SMT4 modes
– smtctl / mpstat commands
•Mixed SMT modes supported within same LPAR
– Requires use of “Resource Groups”0
0.5
1
1.5
2
SMT1 SMT2 SMT4
© 2010 IBM Corporation
IBM Power Systems
POWER7 Multi-threading Options •TurboCore option
•50% of the cores active
0
0.5
1
1.5
2
2.5
SMT4 SMT2 Single0
0.5
1
1.5
2
2.5
SMT4 SMT2 Single
•MaxCore option
•All cores active
Based of rPerf workload
© 2010 IBM Corporation
IBM Power Systems
Fully Optimized POWER7 Module Packaging•Distributed Enterprise/SMB, Infrastructure Consolidation
– Low-cost organic module designed for 2/4 socket platforms
– Targets infrastructure consolidation, distributed enterprise and SMB
– Targets sweet spot of performance, scalability and reliability
– High-density, low-power options
•Large Scale Enterprise and Server Consolidation– Glass ceramic targets leadership performance, scalability and reliability
– Designed for enterprise database, ERP, CRM and decision support
– Ideal for mission-critical and highly virtualized environments
– Supports scalability up to 32 sockets
•Massive Scale-Out– Quad-chip MCM targets high octane MFLOP engines
– Targets unparalleled capacity for modeling complex systems and compute-intensive research
– Scales up to 256 w eight-core processors per server and networked clusters of thousands of servers driving >PFLOP capability
2/4s Blades and RacksSingle Chip Organic
High-End and Mid-RangeSingle Chip Glass Ceramic
Compute IntensiveQuad-chip MCM
© 2010 IBM Corporation
IBM Power Systems
High-End / Mid-Range Packaging
High-End and Mid-RangeSingle Chip Glass Ceramic (61mm)
3363 Pins
© 2010 IBM Corporation
IBM Power Systems
Processor Offerings for Blades / Rack / HPC
POWER7 Processor Offerings
Cores / Socket 4 6 8
PS700 / 701 / 702 Yes - Yes
Power 750 - Yes Yes (3)
Power 755 - - Yes
Configuration Options
Sockets 1 2 3 4
6 Core Chips 6 Cores 12 Cores 18 Cores 24 Cores
8 Core Chips 8 Cores 16 Cores 24 Cores 32 Cores
1-4 Socket System
© 2010 IBM Corporation
IBM Power Systems
Processor Offerings for Modular Systems
POWER7 TurboCore / CoD
Processor Offerings
Cores / Socket 4 TurboCore 6 8 Base 8 Enhanced
Power 770 - Yes Yes -
Power 780 Yes - - Yes
Configuration Options
Enclosures 1 2 3 4
4 Core Chips 8 Cores 16 Cores 24 Cores 32 Cores
6 Core Chips 12 Cores 24 Cores 36 Cores 48 Cores
8 Core Chips 16 Cores 32 Cores 48 Cores 64 Cores
© 2010 IBM Corporation
IBM Power Systems
EnergyScale
•EnergyScale is IBM Trademark. It consists of a built-in Thermal Power Management Device (TPMD) card and Power Executive software.
•IBM Systems Director is also required to manage Energy-Scale functions.
•EnergyScale is used to dynamically optimizes the processor performance versus processor power and system workload.
•IBM Systems Director is also required to manage AEM functions and supports the following functions:
–Power Trending
–Thermal Reporting
–Static Energy Saver Mode
–Dynamic Energy Saver Mode
–Energy Capping
–Soft Energy Capping
–Processor Nap
–Energy Optimized Fan Control
–Altitude Input
–Processor Folding
37
© 2010 IBM Corporation
IBM Power Systems
Definition of Terms
•Energy Trending
–EnergyScale provides continuous collection of real-time server energy consumption. This energy usage data may be displayed or exported by IBM Systems Director Active Energy Manager.
–Administrators may use such information to predict data center energy consumption at various times of the day, week, or month.
•Thermal Reporting
–A measured ambient temperature and a calculated exhaust heat index temperature can be displayed from Active Energy Manager.
–This information can help identify data center “hot-spots” that need attention.
38
© 2010 IBM Corporation
IBM Power Systems
Definition of Terms (continued)
•StaticEnergy Saver Mode
–Static Energy Saver lowers the processor frequency and voltage on an Power 750 a fixed amount, reducing the energy consumption of the system while still delivering predictable performance.
–This percentage is predetermined to be within a safe operating limit and is not user configurable.
–Active Energy Manager is the recommended user interface to enable/disable Energy Saver mode.
–Energy Saver could be enabled based on regular variations in workloads, such as predictable dips in utilization over night, or over weekends.
–Energy Saver can be used to reduce peak energy consumption, which can lower the cost of all power used.
39
© 2010 IBM Corporation
IBM Power Systems
Definition of Terms (continued)
•Dynamic Energy Saver Mode
–Dynamic Energy Saver varies processor frequency and voltage based on the utilization of the Power 750 POWER7 processors.
–The user must configure this setting from Active Energy Manager.
–Processor frequency and utilization are inversely proportional for most workloads, implying that as the frequency of a processor increases, its utilization decreases, given a constant workload.
–Dynamic Energy Saver takes advantage of this relationship to detect opportunities to save power, based on measured real-time system utilization.
–When a system is idle, the system firmware will lower the frequency and voltage to Static Energy Saver values.
–When fully utilized, the maximum frequency will vary, depending on whether the user favors power savings or system performance.
•If an administrator prefers energy savings and a system is fully-utilized, the system will reduce the maximum frequency to 95% of nominal values.
•If performance is favored over energy consumption, the maximum frequency will be at least 100% of nominal.
–Dynamic Energy Saver is mutually exclusive with Static Power Saver mode.
•Only one of these modes may be enabled at a given time.
40
© 2010 IBM Corporation
IBM Power Systems
Definition of Terms (continued)
•Energy Capping
–Power Capping enforces a user specified limit on energy consumption.
–The user must set and enable an energy cap from the Active Energy Manager user interface.
–In most data centers and other installations, when a machine is installed, a certain amount of energy is allocated to it.
–Generally, the amount is what is considered to be a “safe” value, and it typically has a large margin of reserved, extra energy that is never used. This is called the margined power.
–The main purpose of the energy cap is not to save energy but rather to allow a data center operator the ability to reallocate energy from current systems to new systems by reducing the margin assigned to the existing machines.
41
© 2010 IBM Corporation
IBM Power Systems
Definition of Terms (continued)
•“Soft” Energy Capping
–There are two power ranges into which the power cap may be set .
–When a power cap is set in the guaranteed range (described above), the system is
guaranteed to use less energy than the cap setting.
–Setting a energy cap in this region allows for the recovery of the margined power, but in many
cases cannot be used to save energy.
–Soft power capping extends the allowed energy capping range further, beyond a region that
can be guaranteed in all configurations and conditions.
–If the energy management goal is to meet a particular energy consumption limit, then soft
energy capping is the mechanism to use.
42
© 2010 IBM Corporation
IBM Power Systems
Definition of Terms (continued)
•Processor Nap
–The IBM POWER7 processor uses a low-power mode called Nap that stops processor
execution when there is no work to do on that processor core.
–The latency of exiting Nap falls within a partition dispatch (context switch) such that the
Hypervisor firmware can use it as a general purpose idle state.
–When the Operating System detects that a processor thread is idle, it yields control of a
hardware thread to the Hypervisor.
–The Hypervisor immediately puts the thread into Nap.
–When the operating system yields control of the second thread and the processor core
belongs to a dedicated processor partition, the second thread enters Nap mode
43
© 2010 IBM Corporation
IBM Power Systems
Definition of Terms (continued)
•Processor Nap
–If the processor core is in a shared processor pool (the set of cores being used for micro-partition dispatching) and there is no micro-partition to dispatch, the Hypervisor puts the second thread into Nap mode.
–By entering Nap mode, it allows the hardware to clock off most of the circuits inside the processor core.
•Reducing active energy consumption by turning off the clocks allows the temperature to fall, which further reduces leakage (static) power of the circuits causing a cumulative effect.
–Unlicensed cores are kept in core Nap until they are licensed and return to core Nap whenever they are unlicensed again.
44
© 2010 IBM Corporation
IBM Power Systems
Definition of Terms (continued)
•Energy-Optimized Fan Control and Altitude Input
–On the Power 750, firmware will dynamically adjust fan speed based on energy consumption, altitude, ambient
temperature, & energy savings modes.
–Systems are designed to operate in worst-case environments, in hot ambient temperatures, at high altitudes, & with
high power components.
–In a typical case, one or more of these constraints are not valid.
–When no power savings setting is enabled, fan speed is based on ambient temperature, and assumes a high-altitude
environment.
–When a power savings setting is enforced (either Static Energy Saver or Dynamic Energy Saver) fan speed will vary
based on power consumption, ambient temperature, & altitude (if available).
–System altitude may be set (Active Energy Manager). If no altitude is set, system will assume a default value of 350
meters above sea level.
45
© 2010 IBM Corporation
IBM Power Systems
Definition of Terms (continued)
•Processor Folding
–Processor Folding is a consolidation technique that dynamically adjusts, over the short-term, the number of processors available for dispatch to match the number of processors demanded by the workload.
–As the workload increases, the number of processors made available increases; as the workload decreases, the number of processors made available decreases.
–Processor Folding increases energy savings during periods of low to moderate workload because unavailable processors remain in low-power idle states longer.
46
© 2010 IBM Corporation
IBM Power Systems
TPMD: Thermal Power Management Device• TPMD card is part of the base hardware configuration.
• Residing on the processor planar
• TPMD function is comprised of a risk processor and data acquisition
• TPMD monitor power usage and temperatures in real time
• Responsible for thermal protection of the processor cards
• Can adjust the processor power and performance in real time.
• If the temperature exceeds an upper (functional) threshold, TPMD actively reduces power consumption by reducing processor voltage and frequency or throttling memory as needed.
• If the temperature is lower than upper (functional) threshold, TPMD will allows POWER7 cores to “Over clock” if workloads demands are present.
© 2010 IBM Corporation
IBM Power Systems
Active Energy Manager Benefits:
•Monitor energy consumption to allow better utilization of available energy resources.
•Can trend actual energy consumption and corresponding thermal loading of IBM Systems running in their environment with their
applications. :
–Allocate less power and cooling infrastructure to IBM servers
–Lower power usage on select IBM servers
–Plan for the future by viewing trends of power usage over time
–Determine power usage for all components of a rack
–Retrieve temperature and power information via wireless sensors
–Collect alerts and events from facility providers related to power and cooling equipment
•Better understand energy usage across your data center.
–Identify energy usage
–Measure cooling costs accurately
–Monitor IT costs across components
–Manage by department and/or user
•Active Energy Manager also provides a source of energy management data that can be exploited by Tivoli enterprise solutions
such as IBM Tivoli Monitoring and IBM Tivoli Usage and Accounting Manager.
50
© 2010 IBM Corporation
IBM Power Systems
Power 750 Product Features
•Features of the Power 750:8233-E8B…
–POWER7 processor with multiple cores
•32-ways (8 cores/processor card x 4 processor cards)
–Industry Standard RDIMM, DDR3 1066 Mbps with enhanced memory RAS features including 64-byte marking ECC code, and ChipKill detection and correction.
•512 GB maximum (16GB/DIMM x 8 DIMMs/processor card x 4 processor cards)
–8 hot plug and front access SFF SAS DASD.
–1 slim media bay for DVD.
–1 half high bay for tape drive.
–Hot plug 3 PCIe slots and two PCIX slots with Enhanced Error Handling.
–One GX+ slot and one GX++ slot (not hot pluggable)
–Hot plug and redundant power.
–Hot plug and redundant cooling.
–Support for Logical Partitioning (LPAR) and Dynamic LPAR (DLPAR).
–Embedded SAS and SATA
–Embedded four 1 Gigabit Ethernet devices or two 10 Gigabit Ethernet devices
–Embedded USB
–Service Processor FSP-1 for enhanced reliability and remote system management
–Rack mountable drawer
52
© 2010 IBM Corporation
IBM Power Systems
Power 750 System 8233-E8B
POWER7 Architecture 6 Cores @ 3.3 GHz8 Cores @ 3. 0, 3.3, 3.55 GHz Max: 4 Sockets
DDR3 Memory Up to 512 GBSystem Unit SAS SFF Bays
Up to 8 Drives (HDD or SSD)73 / 146 / 300GB @ 15k (2.4 TB)(Opt: cache & RAID-5/6)
System Unit IO Expansion Slots
PCIe x8: 3 Slots (2 shared)PCI-X DDR: 2 Slots 1 GX+ & Opt 1 GX++ 12X cards
Integrated SAS / SATA YesSystem Unit Integrated Ports
3 USB, 2 Serial, 2 HMC
Integrated Virtual Ethernet
Quad 10/100/1000 Optional: Dual 10 Gb
System Unit Media Bays 1 Slim-line DVD & 1 Half HeightIO Drawers w/ PCI slots PCIe = 4 Max: PCI-X = 8 MAXCluster 12X SDR / DDR (IB technology)Redundant Power andCooling
Yes (AC or DC Power)Single phase 240 VAC or -48 VDC
Certification (SoD) NEBS / ETSI for harsh environments
EnergyScale Active Thermal Power ManagementDynamic Energy Save & Capping
4UDepth: 28.8”
© 2010 IBM Corporation
IBM Power Systems
Power 750 System Overview
•8 SFF Bays•(Disk or SSD)
• Dual Power Supplies
•Half-High Bay •(tape or removable disk
• Up to 4 • Processor /
Memory Cards
• 3 PCIe & 2 PCI-X • Slots
• Fans
• TPMD
•DVD
© 2010 IBM Corporation
IBM Power Systems
55
PowerSupplies
Tape DriveRemove DASD Bay
DVD Drive
Operator Panel
8 SFF DASD / SSD
Power 750 Front View
© 2010 IBM Corporation
IBM Power Systems
56
SAS Port
SystemPort 1
SystemPort 2
USBPorts
HMCPorts
IVEEthernet
PCIeSlot 1
orGX++ Slot
PCIeSlot 2
orGX+ Slot
PCIeSlot 3
PCIXSlot 5
PCIXSlot 4
Power 750 Rear View
SPCN
© 2010 IBM Corporation
IBM Power Systems
57
POWER7 chip Enhanced Buffer controller
4 DIMM Slots
4 DIMM Slots
Processor Card
Processor Cards6-core 3.3 GHz #8335 – 1 to 4 per server8-core 3.0 GHz #8332 – 1 to 4 per server8-core 3.3 GHz #8334 – 1 to 4 per server8-core 3.55 GHz #8336 – 4 per server
All processor cards on the same server must be identical feature code
Processor VRM
Memory VRM
© 2010 IBM Corporation
IBM Power Systems
58
Power 750 System LayoutDIMMDIMM
SNDIMMDIMM
SN
DIMMDIMM SNDIMMDIMM SN
POWER7Chip
DIMMDIMM SNDIMMDIMM SN
DIMMDIMM SNDIMMDIMM SN
POWER7Chip
DIMMDIMM
SNDIMMDIMMSN
DIMMDIMMSNDIMMDIMMSN
POWER7Chip
DIMMDIMMSNDIMMDIMMSN
DIMMDIMMSNDIMMDIMMSN
POWER7Chip
PowerSupply 1
PowerSupply 2
8 SFF / SSDDASD
SLIMDVD
Tape Drive
Anchor Card
Cache RAID Card (opt)
Aux Write Cache (opt)
SPCN1SPCN2HMC1HMC2
S1S2
GX++ Slot
TPMD
SASController
DASD&
Media
BackPlane
IO ControllerUSBUSB
RJ45
RJ45ENETPHY
RJ45
RJ45ENETPHY
PCI-X S4PCIe S3
PCI-X S5
PCIe S2PCIe S1
USB
MUX
GX+ SlotFSP
Op-Panel
Ext SAS
USB
2nd Proc / Memory Card required for
GX++ Bus
© 2010 IBM Corporation
IBM Power Systems
Memory Options for Power 750 / 755
FeatureSize (2 DIMM)
DIMM
Size
Memory
Speed
750 Max
Memory
8 GB 4 GB 1066 MHz 128 GB
16 GB 8 GB 1066 MHz 256 GB
32 GB 16 GB 1066 MHz 512 GB
Power 755
Power 750
Feature
Size (2 DIMM)
DIMM
Size
Memory
Speed
755 Max
Memory
8 GB 4 GB 1066 MHz 128 GB
16 GB 8 GB 1066 MHz 256 GB
© 2010 IBM Corporation
IBM Power Systems
Power 750 Memory
• 8 DDR3 DIMM slots per processor card
• DIMMS: 4GB, 8GB and 16GB
• Plugged in pairs. 1 feature code = 1 pair
• Min = 1 feature per SERVER, but min 1 feat per Proc card recommended
• Can NOT mix different size DIMMs on same processor card
• Can have different size DIMMs on same server.
One proc card GB memory capacity with
DIMM size
1 Pair 2 Pair 3 Pair 4 Pair
4 GB 8 16 24 32
8 GB 16 32 48 64
16 GB 32 64 96 128
Feature Code
Feature GB
#4526 16
#4527 32
#4528 64
The following is for ONE processor card in the Power 750
# Proc card
1 2 3 4
DIMM slots 8 16 24 32
Min/Max GB
8 / 128 8 / 256 8 / 384 8 / 512
© 2010 IBM Corporation
IBM Power Systems
POWER7 Memory Bandwidth (750 / 755 / Blades )
POWER7
Mem
Cntrl
DDR3
DDR3
DDR3
DDR3
DDR3
DDR3
DDR3
DDR3
DDR3
DDR3
Nova
Max Read Bandwidth: 51.168 GB/secMax Write Bandwidth: 25.584 GB/secMax Combined Bandwidth: 68.224 GB/sec
DDR3
DDR3
DDR3
DDR3
DDR3
DDR3
DDR3
DDR3
DDR3
DDR3
Nova
Nova
Nova
Each Nova Chip (Read/Write Buffer)
supports two DIMMS
Chip Bandwidth
© 2010 IBM Corporation
IBM Power Systems
Memory BandwidthL1 ( Data ) 170.4 GB/sec
L2 170.4 GB/sec
L3 113.6 GB/sec
Memory 68.224 GB/sec per Socket 272.896 GB/sec per System
Intra-NodeBuses
6.4 GB/sec
GX++ Bus (12X DDR)GX+ Bus (12X SDR)
20 GB/sec
10 GB/sec ( Shared )
GX Bus Slot 1GX Bus Slot 2
Internal IO SlotsTotal IO Bandwidth
20 GB/sec
5** GB/sec
5** GB/sec
30 GB/sec
Power 750 Bandwidth @ 3.55 GHz
** Pass thru bus
© 2010 IBM Corporation
IBM Power Systems
Power 750 Information….•Physical Specifications:
–Width: 440 mm (17.3 in)
–Depth: 730.8 mm (28.8 in)
–Height: 175 mm (6.89 in)
–Weight: 54.4 kg (120 lb)
•Operating voltage: –200 to 240 V
•Operating Frequency: 50/60 Hz
•Power Consumption: 1950 watts (maximum)
•Power Factor: 0.97
•Thermal Output: 4778 Btu/hour (maximum)
•Power-source Loading–1.443 KVA (maximum configuration)
•Noise Level and Sound –Rack-mount drawer: 7.0 Bels operating
© 2010 IBM Corporation
IBM Power Systems
Functional Differences
Power 550 Power 750
Up to 8 Cores (4 sockets) Up to 32 Cores (4 sockets)
Up to 256 GB Memory
32 DIMM slots
Up to 512 GB Memory
32 DIMM slots
DDR2 DIMMS DDR3 DIMMs
6 3.5 in or 8 SFF SAS disk/SSD 8 SFF SAS disk/SSD
3 PCIe & 2 PCI-X slots 3 PCIe & 2 PCI-X slots
Commercial focus Commercial & HPC focus
GX Bus & GX Passthru Slots GX Bus & GX Passthru Slots
IVE: Dual Gb
Optional: Quad Gb, or 10 Gb
IVE: Quad Gb
Optional: Dual 10 Gb
TPMD Enhanced TPMD
Guiding Light Light Path
© 2010 IBM Corporation
IBM Power Systems
Bandwidth Properties…
0
10
20
30
40
50
60
70
Memory Intra IO
Power 550 Power 750
© 2010 IBM Corporation
IBM Power Systems
POWER7 / POWER6 Comparison
0
1000
2000
3000
4000
5000
6000
7000
8000
9000
Energy Consumption Thermal
Power 750 Power 550 Power 560
Power 750: 32 Cores Power 550: 8 Cores Power 560: 16 Cores Active
© 2010 IBM Corporation
IBM Power Systems
Power 750 rPerf Performance …
Power 5508 Core @ 5GHz
Power 56016 Core
@ 3.6GHz
Power 75032 Core
@ 3.55GHz
© 2010 IBM Corporation
IBM Power Systems
Power 750 vs Power 550 / 560
Performance* / KW Performance* / K BTU
* Calculated on rPerf, CPW results siimilar
© 2010 IBM Corporation
IBM Power Systems
Power 755 HPC: 8236-E8C
•Power 755 / Power 750 Differences:1.Only an 8-core 3.3GHz will be offered
2.Valid configuration is 32-core 3.3GHz (i.e. 4 processor cards).
3.No 16GB DIMM - Maximum memory is 256GB.
4.No IBM i O/S support
5.No PowerVM features (i.e. no LPAR or DLPAR)
6.No RAID feature (CCIN 57B7 & 57B8)
7.No Split Disk feature
8.No tape drive
9.No external I/O Drawers (e.g. Tres 19 Drawers)
10.No IB 12x SDR adapter (CCIN 1817)
71
© 2010 IBM Corporation
IBM Power Systems
5.3 / 6.1 RHEL / SLES
Power 755 4-Socket HPC System
8236-E8CPOWER7 Architecture
4 Processor Sockets = 32 Cores8 Core @ 3.3 GHz
DDR3 Memory 128 GB / 256 GB, 32 DIMM SlotsSystem Unit SAS SFF Bays
Up to 8 disk or SSD 73 / 146 / 300GB @ 15K (up to 2.4TB)
System Unit Expansion
PCIe x8: 3 Slots (1 shared)PCI-X DDR: 2 Slots GX++ Bus
Integrated Ports 3 USB, 2 Serial, 2 HMCIntegrated Ethernet Quad 1Gb Copper
(Opt: Dual 10Gb Copper or Fiber)System Unit Media Bay
1 DVD-RAM ( No supported tape bay )
Cluster Up to 64 nodesEthernet or IB-DDR
Redundant Power Yes (AC or DC Power)Single phase 240vac or -48 VDC
Certifications (SoD) NEBS / ETSI for harsh environments
EnergyScale Active Thermal Power ManagementDynamic Energy Save & Capping
Up to 8.4 TFlops per Rack( 10 nodes per Rack )
4U x 28.8” depth
© 2010 IBM Corporation
IBM Power Systems
73
Power 755 System LayoutDIMMDIMM SNDIMMDIMM
SN
DIMMDIMM
SNDIMMDIMM SN
POWER7Core
DIMMDIMM SNDIMMDIMM SN
DIMMDIMM SNDIMMDIMM SN
POWER7Chip
DIMMDIMM
SNDIMMDIMMSN
DIMMDIMMSNDIMMDIMMSN
POWER7Chip
DIMMDIMM
SNDIMMDIMMSN
DIMMDIMMSNDIMMDIMM
SN
POWER7Chip
PowerSupply 1
PowerSupply 2
8 SFF / SSDDASD
SLIMDVD
Anchor Card
SPCN1SPCN2HMC1HMC2
S1S2
GX++ Slot
TPMD
SASController
DASD&
Media
BackPlane
IO ControllerUSBUSB
RJ45
RJ45ENETPHY
RJ45
RJ45ENETPHY
PCI-X S4PCIe S3
PCI-X S5
PCIe S2PCIe S1
USB
MUX
FSP
Op-Panel
Ext SAS
USB
© 2010 IBM Corporation
IBM Power Systems
1H / 2010
Scaling 64 nodes (32 Cores/node) 54 TFlops
Operating Systems AIX 6.1 TL 04 / 05
Linux
HPC Stack Levels xCAT v2.3.x
GPFS v3.3.x
PESSL v3.3.x
LL v4.1.x
PE v5.2.x
ESSL Beta (GA 06/2010)
ESSL v5.1
Compilers GA Levels
XLF v13.1
VAC/C++ v11.1
Power 755 HPC Cluster Node
IB-DDR Interconnect
Data Center in a RackUp to 10 Nodes per Rack
Air cooled
© 2010 IBM Corporation
IBM Power Systems
POWER5+ 575 (1.9GHz)
Power 575 (4.7GHz)
Power 755
(3.3GHz)
Latency (cycles/ns) 220 cycles / 110ns 420 cycles / 90ns 336 cycles / 102ns
Bus 2 X DRAM Freq 4 X DRAM Freq 6 X DRAM Freq
Memory Controllers
per chip1 per chip 1 (2 in HE) 1
Peak Bandwidth
per chip 25GB/s 34GB/s 68 GB/s
DRAM Technology DDR2 DDR2 DDR3
Power 755 Memory
© 2010 IBM Corporation
IBM Power Systems
Power 755 Power 575
Cores/chip 8 4
Total cores 32 32
Frequency 3.3 GHz 4.7 GHz
Memory (max) 256 GB 256 GB
Performance / TFlops .84 .6
Cooling Air Water
Cores/rack
Rack type
320
19”
448
24”
Power (Watts) 1650 5400
755 offers the same core count per node 40% better performance per node (Linpack) 1/3 the power per node 37% less floor space for a 64 node configuration.
Power 755 vs Power 575
© 2010 IBM Corporation
IBM Power Systems
Feature 755 750Processors 32-core @ 3.3 GHz 32-core @ 3.55 GHz
6 / 12 / 18 / 24-core @ 3.3 GHz 8 / 16 / 24 / 32-core @ 3.0 GHz
Memory 128GB OR 256GB 4GB & 8GB DIMMS
512GB Max. 4GB, 8GB, 16GB DIMMS
GX slot support Yes – IB clustering YesI/O Drawer support No YesDASD Backplane No Split Backplane Split Backplane supportIntegrated Ethernet Quad GbE or Dual 10GbE Quad GbE or Dual 10GbEVirtualization No PowerVM support PowerVM Std and EntDASD / Bays 8 SFF SAS HDD / SDD
10k and 15K SFF drives 8 SFF SAS HDD / SDD10k and 15K SFF drives Optional RAID
Internal Tape No YesPerformance Metric TFLOPS rPerfMisc. No IBM i Support
No H/W Raid CardsIBM i SupportH/W Raid Cards
Power 755 vs. 750 Offering Structures
© 2010 IBM Corporation
IBM Power Systems
Power 770 Power 770 Processor Technology 6 Cores @ 3.55 GHz
8 Cores @ 3.1 GHz
L3 Cache On Chip
Redundant Power & Cooling Yes
Redundant Server Processor Yes / Two Enclosure minimum
Redundant Clock Yes / Two Enclosure minimum
Concurrent Add Support Yes
Concurrent Service Yes
System Unit Single Enclosure 4 EnclosuresProcessors Up to 2 Sockets 8 SocketsDDR3 Memory (Buffered) Up to 512 GB Up to 2 TBSAS/SSD SFF Bays 6 24DVD-RAM Media Bays 1 Slim-line 4 Slim-line SAS / SATA Controller 2 / 1 8 / 4PCIe bays 6 PCIe 24 PCIe GX++ Slots (12X DDR) 2 8Integrated Ethernet Std: Quad 1Gb
Opt: Dual 10Gb + Dual 1 Gb
Std: Four Quad 1Gb Opt: Four x Dual 10Gb +
Dual 1 GbUSB 3 1212X I/O Drawers w/ PCI slots Max: 4 PCIe, 8 PCI-X Max: 16 PCIe, 32 PCI-X
Maint Coverage: 9 x 5
4U x 32 inches Depth
© 2010 IBM Corporation
IBM Power Systems
Power 780Power 780
Processor Technology 4 Cores @ 4.1 GHz TurboCore8 Cores @ 3.8 GHz
L3 Cache On Chip
Redundant Power & Cooling Yes
Redundant Server Processor Yes / Two Enclosure minimum
Redundant Clock Yes / Two Enclosure minimum
Concurrent Add Support Yes
Concurrent Service Yes
System Unit Single Enclosure 4 EnclosuresProcessors 2 Sockets 8 SocketsDDR3 Memory (Buffered) Up to 512 GB Up to 2 TBSAS/SSD SFF Bays (CEC) 6 24DVD-RAM Media Bays 1 Slim-line 4 Slim-line SAS / SATA Controller 2 / 1 8 / 4PCIe (CEC) 6 PCIe 24 PCIe GX++ Slots (12X DDR) 2 8Integrated Ethernet Std: Quad 1Gb
Opt: Dual 10Gb + Dual 1 Gb
Std: Four Quad 1Gb Opt: Four x Dual 10Gb +
Dual 1 GbUSB 3 1212X I/O Drawers w/ PCI slots Max: 4 PCIe, 8 PCI-X Max: 16 PCIe, 32 PCI-X
Maint Coverage 24 X 7
PowerCare Support
© 2010 IBM Corporation
IBM Power Systems
POWER7Processor
Chip
16 DIMM slots
PCIe Slots
FSP
GX Slots
6 SFFBays
POWER7 Processor
Chip
Interconnect
TPMD
POWER7 Modular Layout
© 2010 IBM Corporation
IBM Power Systems
POWER7 Modular Front View
FabricInterconnects
6 SFF Bays
DVD
Fans
Op Panel
© 2010 IBM Corporation
IBM Power Systems
POWER7 Modular Rear View
Two GX++ Bays
IVE Ports
Two PowerSupplies
FSP Connectors
HMCPorts
PCIe
PCIe
PCIe
PCIe
PCIe
PCIe
SPCNPorts
HMCPorts
SerialPort
USBPorts
© 2010 IBM Corporation
IBM Power Systems
POWER7 Modular System View
Socket
Socket
Fan
Fan
Fan
Fan
Fan
PCIe Slot
PCIe Slot
PCIe Slot
PCIe Slot
PCIe Slot
PCIe Slot
FSP & Clock
RegulatorMemory DIMMs
Qty: 8
Memory DIMMsQty: 4
Memory DIMMsQty: 4
TPMD
PowerGX++ (12X)
SFF
16 DIMM cards
© 2010 IBM Corporation
IBM Power Systems
Power 770 and Power 780 Processor Options
Socket
Socket
Memory
Memory
Memory
Power 780 Processor Options (2 Sockets per enclosure )
16-core 3.86 GHz #4982 – 1 to 4 per server 8-core 4.14 GHz #4982 – 1 to 4 per server - Turbo Core
Power 770 Processor Options (2 Sockets per enclosure )12-core 3.5 GHz #4980 – 1 to 4 per server16-core 3.1 GHz #4981 – 1 to 4 per server
© 2010 IBM Corporation
IBM Power Systems
Modular Block Diagram…..
RAIDBattery
RAIDBattery
SN
DRAM
DRAM
DRAM
DRAM
DRAM
DRAM
DRAM
DRAM
DRAM
SN
DRAM
DRAM
DRAM
DRAM
DRAM
DRAM
DRAM
DRAM
DRAM
8 SN Dimms
SN
DRAM
DRAM
DRAM
DRAM
DRAM
DRAM
DRAM
DRAM
DRAM
SN
DRAM
DRAM
DRAM
DRAM
DRAM
DRAM
DRAM
DRAM
DRAM
PSI
Media
X Bus
A/B Buses8 SN
Dimms
Write Cache
Y BusZ Bus
CPU CARD
DASD Backplane
PSI
I/OBackplane
PWR7
Ext SAS
GX++
A/B Buses
GX++
Write Cache
FSPCard
SAS
SAS EXP
EXP
PCIe BusesPCIe
BusesSATA
4 x 1Gb Eth
2 x HMC
AnchorCard
PWR7
GX++ Busses
GX++ Busses
IVE
IVE
USB
PCI-X
Serial
PCI-X
IVEUSB Serial
PCIe
PCIe
PCIe
PCIe
PCIe
PCIe
TPMD
IOC2IOC2
© 2010 IBM Corporation
IBM Power Systems
Cooling Domains
Air Flow Domain 1
Air Flow Domain 2
Fans (5)
Power Supply Fans(built-in)
Front Rear
POWER7 Modular supports Hot-Plug and Redundant cooling.There are five fans across the front of the box drawing in room air and is the primary cooling domain. Cool the processors, memory and I/O sub-system.TPMD controls this domain through the FSP. Control algorithm uses the processor, memory and I/O subsystem temperatures as input for fan control.
Second cooling domain, that uses fans inside of each power supplyCools the power supplies and DASD.SPCN controls the fan speed on the power supplies.If the DASD/SSD cage is not installed, SPCN relinquishes control of the fans to the power supplies.
Power supplies control fan speed based on internal power supply temperatures.Fan redundancy is limited to 1 fan fault per domain.More than one failing fan in each domain will force a drawer shutdown.
© 2010 IBM Corporation
IBM Power Systems
POWER7 Modular Memory Card Options
FeatureSize (4 DIMM card)
DIMM Size
MemorySpeed
MaxMemory
32 GB 8 GB 1066 MHz 512 GB
64 GB 16 GB 1066 MHz 1 TB
128 GB 32 GB 800 MHz 2 TB
SuperNova
DDR3 DDR3 DDR3 DDR3
DDR3 DDR3 DDR3 DDR3
DDR3 DDR3 DDR3 DDR3
DDR3 DDR3 DDR3 DDR3
DDR3 DDR3 DDR3 DDR3
DDR3 DDR3 DDR3 DDR3
DDR3 DDR3 DDR3 DDR3 DDR3 DDR3 DDR3 DDR3D
DR
3D
DR
3D
DR
3D
DR
3D
DR
3D
DR
3
DD
R3
DD
R3
DD
R3
DD
R3
4Q10 planned
© 2010 IBM Corporation
IBM Power Systems
POWER7Chip
DIMM 1 SN
DIMM 2 SN
DIMM 3 SN
DIMM 4 SN
DIMM 5 SN
DIMM 6 SN
DIMM 7 SN
DIMM 8 SN
MemCtrl 1
MemCtrl 0
D
CB
A
AB
C
D
POWER7 Modular Memory Layout
© 2010 IBM Corporation
IBM Power Systems
POWER7 Modular Memory Bandwidth
POWER7
Mem
Cntrl
DRAMs…
DRAMs…
Mem
Cntrl
Max Read Bandwidth: 102.336 GB/secMax Write Bandwidth: 51.168 GB/secMax Combined Bandwidth: 136.448 GB/sec
SuperNova
DDR3 DDR3 DDR3 DDR3
DDR3 DDR3 DDR3 DDR3
DDR3 DDR3 DDR3 DDR3
DDR3 DDR3 DDR3 DDR3
DDR3 DDR3 DDR3 DDR3
DDR3 DDR3 DDR3 DDR3
DDR3 DDR3 DDR3 DDR3 DDR3 DDR3 DDR3 DDR3
DD
R3
DD
R3
DD
R3
DD
R3
DD
R3
DD
R3
DD
R3
DD
R3
DD
R3
DD
R3
Nova
Nova
Nova
Nova
Nova
Nova
Nova
Nova
© 2010 IBM Corporation
IBM Power Systems
Three Enclosure Fabric Topology
3 2
3FC #3713
Two cables
FC #3712
© 2010 IBM Corporation
IBM Power Systems
Four Enclosure Fabric Topology
2
4
4
3
3
4
FC #3712
FC #3713Two cables
FC #3714Three cables
© 2010 IBM Corporation
IBM Power Systems
CEC Enclosure 1
FSP/ Clock
CEC Enclosure 2FSP/ Clock
CEC Enclosure 3
Drw to DrwConnection
CEC Enclosure 4
Drw to DrwConnection
Drw to DrwConnection
Point to Point Cabling Three cables
Hot Drawer Add SupportAdd cables to live systemsNo disruptions
Hot Failover supportFSPClock
Concurrent Service Support
FSP Cabling Configuration ( Logical View )
FrontRear
Drw to DrwConnection
© 2010 IBM Corporation
IBM Power Systems
FSP 4 Enclosure Configuration
Enclosure 1
Enclosure 2
Enclosure 3
Enclosure 4
Cable 2
Cable 1
Cable 3
© 2010 IBM Corporation
IBM Power Systems
19-inch Rack Considerations
21
Cables wider than CEC
Multi-enclosure configurations supported in IBM “Enterprise” racks: IBM 7014-T00, -T42, #0551, #0553 No problemS with a front door
(regular or acoustic), but if use rack trim, need new #6247 trim kit
For a Power 780 door with the pretty 780 label, order as feat code of 7014-T42 rack.
© 2010 IBM Corporation
IBM Power Systems
POWER7 770 Bandwidth @ 3.1 GHz
Memory Bandwidth
L1 ( Data ) 148.8 GB/sec
L2 148.8 GB/sec
L3 99.2 GB/sec
Memory4 Nodes
136.448 GB/sec per socket1091.584 GB/sec
Inter-NodeBuses (4 Nodes) 158.016 GB/sec
Intra-NodeBuses (4 Nodes) 415.744 GB/sec
Int GX Bus 1 & 24 Enclosures
19.712 GB/sec78.848 GB/sec
Ext GX Bus 1 & 24 Enclosures
39.424 GB/sec157.696 GB/sec
Total IO( 4 Enclosures ) 236.544 GB /sec
© 2010 IBM Corporation
IBM Power Systems
POWER7 780 Bandwidth @ 3.86 GHz
Memory Bandwidth
L1 ( Data ) 185.28 GB/sec
L2 185.28GB/sec
L3 123.52 GB/sec
Memory4 Nodes
136.448 GB/sec per Socket1091.584 GB/sec
Inter-NodeBuses (4 Nodes) 158.016 GB/sec
Intra-NodeBuses (4 Nodes) 415.744 GB/sec
Int GX Bus 1 & 24 Enclosures
19.712 GB/sec78.848 GB/sec
Ext GX Bus 1 & 24 Enclosures
39.424 GB/sec157.696 GB/sec
Total IO( 4 Enclosures ) 236.544 GB /sec
© 2010 IBM Corporation
IBM Power Systems
POWER7 780 Bandwidth @ 4.14 GHz
Memory Bandwidth
L1 ( Data ) 198.72 GB/sec
L2 198.72 GB/sec
L3 132.48 GB/sec
Memory4 Nodes
136.448 GB/sec per Socket1091.584 GB/sec
Inter-NodeBuses (4 Nodes) 158.016 GB/sec
Intra-NodeBuses (4 Nodes) 415.744 GB/sec
Int GX Bus 1 & 24 Enclosures
19.712 GB/sec78.848 GB/sec
Ext GX Bus 1 & 24 Enclosures
39.424 GB/sec157.696 GB/sec
Total IO( 4 Enclosures ) 236.544 GB /sec
© 2010 IBM Corporation
IBM Power Systems
POWER7 Modular Information….
Physical Specifications (4 EIA units) Width: 483 mm (19.0 in.) Depth: 863 mm (32.0 in.) Height: 174 mm (6.85 in) Weight: 70.3 kg (155 lb)
Operating voltage: 200 to 240 V
Operating Frequency: 50/60 Hz Power Consumption: 1600 watts (maximum)
Per enclosure with 16 cores active Power Factor: 0.97 Thermal Output: 5461 Btu/hour (maximum)
Per enclosure with 16 cores activePower-source Loading
1.649 kva (maximum configuration) Noise Level and Sound
One enclosure with 16 active cores:6.8 bels / 6.3 bels with acoustic rack doors (operating/idle)
Four enclosures with 64 active cores:7.4 bels / 6.9 Bels with acoustic rack doors (operating/idle)
© 2010 IBM Corporation
IBM Power Systems
Power 770 & 780 vs Power 570 Differences
Power 570 Power 770 & 780
Up to 8 sockets, Up to 32 Cores Up to 8 Sockets, Up to 64 cores
Up to 768 GB Memory Up to 2 TB Memory ( Initial GA will be 1 TB)
DDR2 DIMMS DDR3 DIMMS
Six 3.5” SAS Bays / Enclosure Six SFF SAS Bays / Enclosure
4 PCIe & 2 PCI-X slots per Enclosure 6 PCIe slots per Enclosure
No write cache or RAID-5/6 support Write cache & RAID-5/6 support
Single integrated DASD / Media Cntlr Three integrated DASD / Media Controllers
Optional Split Backplane Standard Split backplaneOptional Tri-Split Backplane
No Power & Management Thermal Power & Thermal management TPMD support
Clock Cold FailoverNo Concurrent Maintenance of FSP/ClockConcurrent Drawer Maint restrictionsConcurrent Drawer Add cable restrictions
Clock Hot FailoverPlanned Concurrent MaintenanceNo Restrictions ( 4Q / 2010 )No Restrictions
© 2010 IBM Corporation
IBM Power Systems
Power 570/32 vs 770 Bandwidth Properties…
0
200
400
600
800
1000
1200
Memory Inter Intra IO
POWER6 POWER7
© 2010 IBM Corporation
IBM Power Systems
POWER7 / POWER6 Enclosure Comparison
0
1000
2000
3000
4000
5000
6000
Power Consumption Thermal
Power 770 Power 570
POWER7: 16 Cores active / POWER6: 8 Cores Active
© 2010 IBM Corporation
IBM Power Systems
770 780 595POWER6
Nodes 4 4 8
Processors 12 - 48 / 16 - 64 8 – 32 / 16 - 64 8 – 64( Upgradeable to 256 Cores)
Frequency 3.1 GHz3.5 GHz ( 6 Core )
3.55 GHz4.14 Core ( 4 Core )
4.2 GHz5.0 GHz
Memory / core 21 / 16 GB 32 / 16 GB 64 GB
rPerf 493.37 / 579.39 418.64 / 685.09 553
Memory Bandwidth
1088 GB/sec 1088 GB/sec 1376 GB / sec
IO Bandwidth 236 GB /sec 236 GB/sec 640 GB/sec
Warranty 9 x 5 24 x 7 24 x 7
PowerCare No Yes Yes
IBM Power 770 / 780 Positioning
© 2010 IBM Corporation
IBM Power Systems
Simplify Web Facing Application Deployment
1 JVM
AIX TL4
64 bit
16 threads
1 JVM
Windows
64 bit
16 threads
3920Transactions/sec
IBM Power 750
8 cores3.55GHz
Nehalem EP8 cores
2260Transactions/sec
73% more work per JVM image
Simpler configurationsBetter scale for software
built on application server
Extend WebSphere Application Server to fully exploit up to 32 threads in a single process, thereby reducing the number of images required
IBM WebSphere Application Server 7
Competitive application server
© 2010 IBM Corporation
IBM Power Systems
Simplify SAP Infrastructure And Reduce CostsCase Study - Support 15,000 Users
110
IBM Power 750 certification number not available at press time and can be found at sap.com/benchmarks. IBM Power System 750, 4p / 32–c / 128 – t, POWER7, 3.55 GHz, 256 GB memory, 15,600 SD users, dialog resp.: 0.98s, line items/hour: 1,704,330, Dialog steps/hour: 5,113,000, SAPS: 85,220, DB time (dialog/ update):0.015s / 0.028s, CPU utilization: 99%, OS: AIX 6.1, DB2 9.7;IBM Power 570 16p / 32-c / 64 –t, 256 GB memory, 14,432 SD users, POWER6 4.2 GHz, AIX 6.1, DB2 9.5, cert# 2008057All results are 2-tier, SAP EHP 4 for SAP ERP 6.0 (Unicode) and valid as of 2/9/2010.
DB2 on Power 750
Oracle on Sun T5440
Oracle on Sun X4640
15,600SD Users
32 coresPOWER7
4,720SD Users
32 coresUltraSPAR
C T2+
10,000SD Users
48 coresAMD
Opteron
SAP Sales and DistributionERP 6.0 EHP 2-Tier performance
Support 3.3x more users on DB2 and Power 750 than Oracle on SPARC
Infrastructure costs per user DB2 on Power 750 $100/user Oracle on Sun T5440 $185/user Oracle on Sun x4640 $123/user
SAP and DB2 on Power 750
© 2010 IBM Corporation
IBM Power Systems
POWER4™p670
1.1 GHzrPerf: 24.46KWatts: 6.71
3.64
POWER4+™ p670
1.5 GHzrPerf: 46.79KWatts: 6.71
6.97
POWER5™p5-570
1.65 GHzrPerf: 68.4KWatts: 5.2
13.15
POWER5+™ p570
1.9 GHzrPerf: 85.20KWatts: 5.2
16.38
POWER6™ Power 570
4.7 GHzrPerf: 134.35KWatts: 5.6
23.99
POWER6™ Power 570
4.2 GHzrPerf: 193.25KWatts: 5.6
34.56
POWER7™ Power 780
3.8 GHzrPerf: 685.09KWatts: 6.4
107.04
Performance per Watt
© 2010 IBM Corporation
IBM Power Systems
POWER7 Virtualization SupportMaintain 1 to 10 ratio for Physical cores to LPARs. Power 750: Up to 160 (320) LPARS Power 755 Not Supported Power 770 / 780: Up to 160 (640) LPARs PS700 Up to 40 LPARS PS701 Up to 80 LPARs PS702 Up to 160 LPARs
Active Memory Expansion Active Memory Expansion compresses in-memory data to fit more data into
memory Increases the effective amount of memory capacity
Managed by the OS and hypervisor OS compresses and decompress data based on memory accesses
Is transparent to applications
© 2010 IBM Corporation
IBM Power Systems
HMC Support
HMC V7 R710 is the minimum level for POWER7 support
HMC used to manage any POWER7 processor based server, must be a CR3 or later model rack-mount HMC or C05 or later deskside HMC.
If IBM Systems Director is used to manage an HMC or if the HMC manages more than 254 partitions, the HMC should have 3GB of RAM minimum and be a CR3 model or later rack-mount, or C06 or later deskside.
© 2010 IBM Corporation
IBM Power Systems
POWER7 OS Software Support
AIX 5.3 with the 5300-11 Technology Level and SP2, or later
AIX 6.1 with the 6100-04 Technology Level and SP3, or later
IBM i 6.1 with 6.1.1 machine code, or later
SUSE Linux Enterprise Server 10 with SP3 for POWER
SUSE Linux Enterprise Server 11 for POWER, or later
RHEL SoD
VIOS 2.1.2.12 with Fix Pack 22.1 and Service Pack 2, or later
© 2010 IBM Corporation
IBM Power Systems
POWER7 Hardware Support
TL9TL8
TL10TL11
TL12
2009 – 2011 AIX TL Roadmap04/2010 10/2010
AIX 6.1
AIX 5.3
04/2011 10/201110/2009
AIX 7.1
TL0TL1
TL2
SP
Service PackPOWER7 Support
TL2TL1
TL3TL4
TL5TL6
TL7TL8
TL12
TL5
TL9TL10
TL2TL3
TL11
TL4
SPSP
SP
SPSP
SP
© 2010 IBM Corporation
IBM Power Systems
Partition Mobility
POWER6POWER6+
POWER7
Binary CompatibilityBinary Compatibility between POWER6 and POWER7 between POWER6 and POWER7
Leverage POWER6 / POWER6+ Compatibility ModeLeverage POWER6 / POWER6+ Compatibility Mode
Migrate partitions between POWER6 and POWER7 ServersMigrate partitions between POWER6 and POWER7 Servers Forward and BackwardForward and Backward
© 2010 IBM Corporation
IBM Power Systems
Active Memory Expansion
POWER7 AdvantageExpand memory beyond physical limitsMore effective server consolidation
Run more application workload / users per partition Run more partitions and more workload per server
ExpandedMemory
TrueMemory Effectively up
to 100% more memoryTrue
Memory
TrueMemory
TrueMemory
TrueMemory
TrueMemory
ExpandedMemory
ExpandedMemory
ExpandedMemory
ExpandedMemory
ExpandedMemory
© 2010 IBM Corporation
IBM Power Systems
Active Memory Expansion & Active Memory Sharing
Active Memory ExpansionEffectively gives more memory
capacity to the partition using compression / decompression of the contents in true memory
AIX partitions only
Active Memory SharingMoves memory from one partition
to anotherBest fit when one partition is not
busy when another partition is busy
AXI, IBM i, and Linux partitions
Active Memory Expansion Active Memory Sharing
Supported, potentially a very nice option
Considerations Only AIX partitions using Active Memory Expansion Active Memory Expansion value is dependent upon compressibility of data
and available CPU resource
0
5
10
15#10
#9
#8
#7
#6
#5
#4
#3
#2
#1
© 2010 IBM Corporation
IBM Power Systems
Active Memory Expansion
Innovative POWER7 technology For AIX 6.1 or later For POWER7 servers
Uses compression/decompression to effectively expand the true physical memory available for client workloads
Often a small amount of processor resource provides a significant increase in the effective memory maximum Processor resource part of AIX partition’s resource and licensing
Actual expansion results dependent upon how “compressible” the data being used in the application A SAP ERP sample workload shows up to 100% expansion, Your results will vary Estimator tool and free trial available
© 2010 IBM Corporation
IBM Power Systems
Active Memory Expansion – Client Deployment
1Planning Tool
A. Part of AIX 6.1 TL4B. Calculates data
compressibility & estimates CPU overhead due to Active Memory Expansion
C. Provides initial recommendations
260-Day Trial
A. One-time, temporarily enablement
B. Config LPAR based on planning tool
C. Use AIX tools to monitor Act Mem Exp environment
D. Tune based on actual results
3Deploy into Production
A. Permanently enable Active Memory Expansion
B. Deploy workload into production
C. Continue to monitor workload using AIX performance tools
Memory Expansion
CP
U U
tiliz
atio
n Estimated Results
CP
U U
tiliz
atio
n
Memory Expansion
Ap
p.
Per
form
ance
Memory Expansion Time
Pe
rfo
rman
ce
Actual Results
© 2010 IBM Corporation
IBM Power Systems
Active Memory Expansion - Planning Tool
Tool included in AIX 6.1 TL4 SP2 Run tool in the partition of interest for memory expansion. Input desired expanded memory size. Tool outputs different real memory
and CPU resource combinations to achieve the desired effective memory.
Active Memory Expansion Modeled Statistics:-----------------------Modeled Expanded Memory Size : 8.00 GB
Expansion True Memory Modeled Memory CPU Usage Factor Modeled Size Gain Estimate--------- -------------- ----------------- ----------- 1.21 6.75 GB 1.25 GB [ 19%] 0.00 1.31 6.25 GB 1.75 GB [ 28%] 0.20 1.41 5.75 GB 2.25 GB [ 39%] 0.35 1.51 5.50 GB 2.50 GB[ 45%] 0.58 1.61 5.00 GB 3.00 GB [ 60%] 1.46
Active Memory Expansion Recommendation:---------------------The recommended AME configuration for this workload is to configure the LPAR with a memory size of 5.50 GB and to configure a memory expansion factor of 1.51. This will result in a memory expansion of 45% from the LPAR's current memory size. With this configuration, the estimated CPU usage due to Active Memory Expansion is approximately 0.58 physical processors, and the estimated overall peak CPU resource required for the LPAR is 3.72 physical processors.
Sample outputThis sample partition has fairly good expansion potential
A nice “sweet” spot for this partition appears to be 45% expansion
• 2.5 GB gained memory
• Using about 0.58 cores additional CPU resource
© 2010 IBM Corporation
IBM Power Systems
Active Memory Expansion: Partition On / Off
With HMC, check Active Memory Expansion box and enterTrue and max memoryMemory expansion factor
To turn off expansion, unclick box
Partition IPL required to turn on or off
Active Memory Expansion Modeled Statistics:-----------------------Modeled Expanded Memory Size : 8.00 GB
Expansion True Memory Modeled Memory CPU Usage Factor Modeled Size Gain Estimate--------- -------------- ----------------- ----------- 1.21 6.75 GB 1.25 GB [ 19%] 0.00 1.31 6.25 GB 1.75 GB [ 28%] 0.20 1.41 5.75 GB 2.25 GB [ 39%] 0.35 1.51 5.50 GB 2.50 GB [ 45%] 0.58 1.61 5.00 GB 3.00 GB [ 60%] 1.46
Active Memory Expansion Recommendation:---------------------The recommended AME configuration for this workload is to configure the LPAR with a memory size of 5.50 GB and to configure a memory expansion factor of 1.51. This will result in a memory expansion of 45% from the LPAR's current memory size. With this configuration, the estimated CPU usage due to Active Memory Expansion is approximately 0.58 physical processors, and the estimated overall peak CPU resource required for the LPAR is 3.72 physical processors.
5.5 true8.0 max
Sample output
© 2010 IBM Corporation
IBM Power Systems
Upgrades from POWER6 and POWER6+
Power 7803.8 GHz / 4.1 GHz
POWER6+ 570/324.2 GHz
POWER6 5703.5, 4,2, 4.7 GHz
9117-MMA9117-MMB
All existing POWER6 570 systems can upgrade to POWER7
Power 7703.5 GHz
POWER6+ 5704.4, 5.0 GHz
9179-MHB
POWER6 upgrades to POWER7 POWER6+ upgrades to POWER7 Power 570/32 upgrades to POWER7
POWER7 System Upgrades
© 2010 IBM Corporation
IBM Power Systems
POWER6+ 570/329117-MMA 4.2 GHz
POWER6+ 5709117-MMA 4.4, 5.0 GHz
POWER6 5709117-MMA 3.5, 4.2, 4.7 GHz
POWER6 5709406-MMA 4.7 GHz
POWER7 7809179-MHB 3.8 / 4.1 GHz
POWER7 7709117-MMB 3.5 GHz
POWER7 7709117-MMB 3.1 GHz
• No direct POWER5 upgrades to POWER7. Use 2-step upgrade, first to POWER6 then to POWER7.
• Upgrades to POWER6 570 available as long as new box sales of POWER6 570 available
• Withdrawal planned end 2010
Power 770 and Power 780 Upgrades
© 2010 IBM Corporation
IBM Power Systems
I/O Upgrade ConsiderationsAll the newer IBM I/O drawers, disk, SSD and PCI adapters used on POWER6
supported on POWER7 servers May need to move 3.5-inch SAS drives and PCI-X adapters
Older I/O on POWER6 servers, but not on POWER7 servers RIO / RIO2 / HSL I/O drawers SCSI disk smaller than 69GB or SCSI drives slower than 15k rpm QIC tape drives IOPs and IOP-based PCI adapters (IBM i)
2749, 5702, 5712, 2757, 5581, 5591, 2790, 5580, 5590, 5704, 5761, 2787, 5760, 4801, 4805, 3709, 4746, 4812, 4813
Older LAN adapters: #5707, 1984, 5718, 1981, 5719, 1982 Older SCSI adapters: #5776, 5583, 5777 Telephony adapter: #6412
See planning web page www.ibm.com/systems/power/hardware/sod2.html
© 2010 IBM Corporation
IBM Power Systems
Power 520 SoD for Upgrade
SoD provided in February For Power 520 (8203-E4A) 2-core or 4-core servers Upgrade to a POWER7 product preserving the serial number
© 2010 IBM Corporation
IBM Power Systems
Power Systems: Hardware & OS RAS Leadership
Ref: ITIC 2009 Global Server Hardware and Server OS Reliability Survey
© 2010 IBM Corporation
IBM Power Systems
Availability / Reliability by Design for POWER7
ProcessorsDynamic De-Allocation
PackagingInstruction Retry
Alternate Processor Recovery
First Failure Data CaptureHelp eliminates intermittent failures
Hot Plug / RemovalFans & Power Supplies
Hot Plug / RemovalPCI-X & PCIe Adapters
IO Drawers
Hot Plug / RemovalDisks
MemoryChip Kill technology with
Bit-steering
Passive backplaneNo active components
HypervisorMainframe technology
MobilityPartition MobilityWPAR Mobility
Operating SystemHot patch Kernel
Storage KeysConcurrent Add: 770/780Eliminates Upgrade outages
Concurrent Service{ 770/780Eliminates Repair Outages
Hot AddI/O racks
Dual Clocks770/780
© 2010 IBM Corporation
IBM Power Systems
POWER7 Instruction Retry
Recovery Capability Array error
Error correction (ECC)Arrays with parity
o Processor restarts Instruction flow and Data flow Error
Processor restarts Control Error
Processor restarts
System Resiliency Processor states are check pointed and protected with ECC Processor states can be moved from one processor to another upon
unsuccessful recovery restart (CP Sparing)
Core
RecoveryUnit
Core restart
Core error collection
Execution Units
Load/Store
Instruction FetchDecode
© 2010 IBM Corporation
IBM Power Systems
PCI Adapter
Fabric Bus Interface to other Chips and NodesECC protectedNode hot add /repair
Core RecoveryLeverage speculative execution resources to enable recoveryError detected in GPRs FPRs VSR, flushed & retriedStacked latches to improve SERAlternate Processor Recovery Partition isolation for core checkstops
L3 eDRAM ECC protected SUE handling Purge and Line delete Spare rows and columns
GX IO Bus ECC protected Concurrent add/repair
InfiniBand® Interface Redundant paths
Retry/Freeze behavior options for Internal I/O Hub Faults PHB Errors
IO Hub
PCIBridge
Advanced 64 Byte ECC on Memory Multiple chip chipkill detections and
sparings HW assisted scrubbing SUE handling CRC with retry and Dynamic data bit-line sparing on channel
interface
OSC0 OSC1Dynamic Oscillator
FailoverConcurrent Repair
BUF
BUF
BUF
BUF
X8 Dimms
Fabric Interface
POWER7 770/780: RAS Features
© 2010 IBM Corporation
IBM Power Systems
Power 750 / 755 Memory RAS FeaturesMemory RASPower 750 supports memory Scrubbing, 64-byte Marking ECC and Chipkill. Memory errors are usually classified as either soft or hard. Hard errors can be caused by defects within the DRAM
package among other reasons (e.g. defect in the silicon), and are usually permanent once they occur.
Soft errors are caused by charged particles or radiation, and are usually transient.
Memory scrubbing corrects soft single bit errors in background while memory is idle preventing multiple bit errors.
Memory ECC is able to detect and correct single bit memory errors, which make up the majority of memory errors. It can also isolate a single Chipkill to a bad DRAM chip.
Memory Chipkill has the ability to correct the single bit errors that standard ECC memory can correct but also multi-bit (2, 3, 4 bits) memory errors and by doing so it increases server availability/reliability even further.
Selective Memory Mirroring is where an amount of memory is reserved and sections of the memory to select for mirroring in the reserved memory are dynamically determined. The selected sections of the memory contain critical areas. The selected sections of the memory are mirrored in the reserved memory.
137
© 2010 IBM Corporation
IBM Power Systems
Guiding Light vs. Light PathUser interface and repair action simplificationPOWER6 Guiding light: Console Monitored System
Used in POWER6 systemsTechnician used the HMC or ASMI to see what happened and what to replaceFRUs and their locations are found in the error logUsed HMC or ASMI to enter location code to activate identify LEDs from console to verify where FRUs are
Exchange FRUs (CM or dedicated) per procedure instructionsSystem Attention LED is persistent after power cycle until cleared by technician.
POWER7 Light Path: Console-less HW service simplicity (fast and easy)Used in POWER7 systemsNo need to view logs, just replace the FRUs with their LED on
o LEDs are activated automatically when error is detected.Friendly, color coded labels on Power 750 cover show how to pull and plug partsExternal LED indicates a failure inside the unitInternal LEDs per FRU indicate what to replace
o FRUs are either Hot Plug or system has to be powered offo Gold cap powers fault LEDs when system power is removed
Replacing a FRU automatically resets the fault LEDs
138
© 2010 IBM Corporation
IBM Power Systems
POWER7 Processor
POWER7 Servers Power 750 Power 755 Power 770 Power 780
Active Memory Expansion
I/O
Upgrades
Summary
© 2010 IBM Corporation
IBM Power Systems
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.
Current: 1Q 2010
Special notices
© 2010 IBM Corporation
IBM Power Systems
IBM, the IBM logo, ibm.com AIX, AIX (logo), AIX 6 (logo), AS/400, Active Memory, BladeCenter, Blue Gene, CacheFlow, 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, POWER7, pureScale, System i, System p, System p5, System Storage, System z, Tivoli Enterprise, TME 10, TurboCore, 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.
Current: 1Q 2010
Special notices (cont.)
© 2010 IBM Corporation
IBM Power Systems
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
Notes on benchmarks and values
Current: 1Q 2010
© 2010 IBM Corporation
IBM Power Systems
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 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
Current: 1Q 2010
Notes on HPC benchmarks and values
© 2010 IBM Corporation
IBM Power Systems
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
Current: 1Q 2010