IBM小型机硬件架构

上传人:pu****.1 文档编号:588608107 上传时间:2024-09-08 格式:PPT 页数:44 大小:3.14MB
返回 下载 相关 举报
IBM小型机硬件架构_第1页
第1页 / 共44页
IBM小型机硬件架构_第2页
第2页 / 共44页
IBM小型机硬件架构_第3页
第3页 / 共44页
IBM小型机硬件架构_第4页
第4页 / 共44页
IBM小型机硬件架构_第5页
第5页 / 共44页
点击查看更多>>
资源描述

《IBM小型机硬件架构》由会员分享,可在线阅读,更多相关《IBM小型机硬件架构(44页珍藏版)》请在金锄头文库上搜索。

1、IBM 小型机简介IBM Power6 p570General descriptionArchitecture and technical OverviewIBM Power5 p590General descriptionArchitecture and technical OverviewSystem specifications单个Central Electronics Complex(CEC) enclosure的规格:Physical package可以使用1-4个building block enclosures每个CEC drawer building blocks高4UFron

2、t ViewBack ViewPower5系列系列System featuresThe full system configuration is made of four CEC building blocks. It features:2-, 4-, 8-, 12-, 16-, and 32-core configurations utilizing the POWER6 chip on up to eight dual core processor cards, or eight dual-core POWER6 dual-chip processor cards.32 MB of L3

3、cache, 8 MB of L2 cache.3.5, 4.2, or 4.7 GHz. Up to 192 GB DDR2 memory per enclosure, 768 GB DDR2 max per system. Available memory features are 667 MHz, 533 MHz, or 400 MHz depending on memory density. Up to 6 SAS DASD disk drives per enclosure, 24 max per system.6 PCI slots per enclosure: 4 PCIe, 2

4、 PCI-X; 24 PCI per system: 16 PCIe, 8 PCI-X. Up to 2 GX+ adapters per enclosure; 8 per system One hot-plug slim-line media bay per enclosure, 4 max per system.IBM Power6 p570General descriptionArchitecture and technical OverviewIBM Power5 p590General descriptionArchitecture and technical OverviewOve

5、rviewThe POWER6 processorCompatibility of 64-bit architectureBinary compatibility for all POWER and PowerPC application code levelSupport of partition migrationSupport of virtualized partition memorySupport of four page sizes : 4 KB, 64 KB, 16 MB, and 16 GBHigh frequency optimizationDesigned to oper

6、ate at maximum speed of 5 GHzSuperscalar core organizationSimultaneous Multithreading: two threadsProcessor cardsIn the 570, the POWER6 processors, associated L3 cache chip, and memory DIMMs are packaged in processor cards.A single CEC may have one or two processor cards installed.They are interface

7、d to 12 memory slots, where as each memory DIMM has its own memory buffer chip and are interfaced in a point-to-point connection.I/O connects to the 570 processor module using the GX+ bus. Processor drawer interconnect cablesThe SMP fabric bus that connects the processors of separate 570 building bl

8、ocks is routed on the interconnect cable that is routed external to the building blocks. The flexible cable attaches directly to the processor cards, at the front of the 570 building block, and is routed behind the front covers (bezels) of the 570 building blocks.Memory subsystem-Fully buffered DIMM

9、FBD是一种用于提高可靠性、速度和密度的内存技术。Memory subsystem- Memory placements rulesFirst quad includes J0A, J0B, J0C, and J0D memory slotsSecond quad includes J1A, J1B, J1C, and J1D memory slotsThird quad includes J2A, J2B, J2C, and J2D memory slotsSystem buses - I/O buses and GX+ cardEach POWER6 processor provides

10、a GX+ bus which is used to connect to an I/O subsystem or Fabric Interface card.In a fully populated 570, there are two GX+ buses, one from each processor.Optional Dual port RIO-2 I/O Hub (FC 1800) and Dual port 12x Channel Attach (FC 1802)adapters that are installed in the GX+ slots are used for ex

11、ternal DASD and IO drawer expansion.System buses - Service processor busThe Service Processor (SP) flex cable is at the rear of the system and is used for SP communication between the system drawers.Internal I/O subsystemThe internal I/O subsystem resides on the system planar which supports a mixtur

12、e of both PCIe and PCI-X slots.Internal storageThe 570 internal disk subsystem is driven by the latest DASD interface technology Serial Attached SCSI (SAS). This interface provides enhancements over parallel SCSI with its point to point high frequency connections. The SAS controller has eight SAS po

13、rts, four of them are used to connect to the DASD drives and one to a media device.The DASD backplane implements two SAS port expanders that take four SAS ports from the SAS controller and expands it to 12 SAS ports. These 12 ports allow for redundant SAS ports to each of the six DASD devices.The DA

14、SD backplane provides the following functions :supports six 3.5 inches SAS DASD devices contains two SAS port expanders for redundant SAS paths to the SAS devicesSAS passthru connection to medias backplaneExternal I/O subsystemsEach GX+ bus can be populated with a GX+ adapter card that adds more RIO

15、-G ports to connect external I/O drawers.IBM Power6 p570General descriptionArchitecture and technical OverviewIBM Power5 p590General descriptionArchitecture and technical OverviewSystem framesp5-590 systems are based on the same 24-inch wide, 42 EIA height frameInside this frame all the server compo

16、nents are placed in predetermined positions.System specificationsPhysical packageIBM Power6 p570General descriptionArchitecture and technical OverviewIBM Power5 p590General descriptionArchitecture and technical OverviewOverviewSystem designBoth the p5-590 and p5-595 servers are based on a modular de

17、sign, where all components are mounted in 24-inch racks. Inside this rack, all the server components are placed in specific positions. This design and mechanical organization offer advantages in optimization of floor space usage.There are three major subsystems:The Central Electronics Complex (CEC)T

18、he power subsystemThe I/O subsystemCentral Electronics ComplexThe Central Electronics Complex is an 18 EIA unit drawer that houses:One to four processor books (nodes) The processor book contains the POWER5+ or POWER5 processors, the L3 cache located in multi-chip modules, memory, and RIO-2 attachmen

19、t cards.CEC backplane (double-sided passive backplane) that serves as the system component mounting unit Processor books plug into the front side of the backplane. The node distributed converter assemblies (DCA) plug into the back side of the backplane. The DCAs are the power supplies for the indivi

20、dual processor books.A fabric bus structure on the backplane board provides communication between books.Service processor unitLocated in the panel above the distributed converter assemblies (DCA). It contains redundant service processors and oscillator cards.Remote I/O (RIO) adapters to support atta

21、ched I/O drawersFans and blowers for CEC coolingLight strip (front, rear) to attenuate the system statusLogical view of the CEC componentsCEC backplane - top view of the p5-595 CEC backplaneIn the p5-590 configuration, book 2 and book 3 are not available.Service processorp5-590 and p5-595 have two s

22、ervice processors. one is considered the primary and the other secondary. The two service processor cards are in the same assembly with two redundant oscillator cards (OSC) shared by all processor books.Processor booksIn the p5-590 and p5-595 systems, the POWER5+ or POWER5 processors are packaged wi

23、th the L3 cache into a cost-effective multi-chip module package. The storage structure for the POWER5+ or POWER5 processors is a distributed memory architecture that provides high-memory bandwidth. Each processor can address all memory and sees a single shared memory resource. As such, two MCMs with

24、 their associated L3 cache and memory are packaged on a single processor book. The p5-590 supports up to two processor books (each book is a 16-core), and the p5-595 supports up to four processor books. Each processor book has dual MCMs containing POWER5+ or POWER5 processors and 36 MB L3 modules. E

25、ach 16-core processor book also includes 16 slots for memory cards (as shown in Figure 2-12 on page 30) and six remote I/O attachment cards (RIO-2) for connection of the system I/O drawers.The POWER5+ processorMulti-chip module and system interconnectPOWER5+ or POWER5 processors can be packaged in s

26、everal ways, such as a multi-chip module (MCM), dual-core module (DCM), or single-core module (SCM). MCMs are used for the basic building block for p5-590 and p5-595 systems.Multi-chip module and system interconnect-continueEach MCM has four POWER5+ or POWER5 processors (with a total of eight cores)

27、 and four L3 cache modules. The processors and their associated L3 cache are connected using processor-to-processor ports. There are separate communication buses between processors in the same MCM and processors in different MCMs.The POWER5+ or POWER5 processors are mounted on an MCM in order that t

28、hey are all rotated 90 degrees from one another. Figure 2-6 shows a picture of an MCM. This arrangement minimizes the interconnect distances, which improves the speed of the inter-processor communication.Multi-chip module and system interconnect-continueTwo POWER5+ or POWER5 MCMs can be tightly coup

29、led to form a book, as shown in Figure 2-7. These books are interconnected again to form larger SMPs with up to 64-cores.The MCMs and books can be interconnected to form 8-core, 16-core, 32-core, 48-core, and 64-core SMPs with one, two, four, six, and eight MCMs, respectively.Light stripThere is no

30、operator panel on p5-590 and p5-595 servers. The p5-590 and p5-595 have a light strip on both the front and the rear of the system unit. The light strips contain several LEDs, each representing the status of a particular field replaceable unit (FRU) or component. Under normal system operating condit

31、ions:No amber LEDs are lit. An amber LED that is lit indicates a problem with the component associated with that LED. If an active component has a green LED associated with it, that LED is lit. Processor books, oscillator cards, SP cards, and the light strips themselves are FRUs for which both green

32、 and amber LEDS are assigned. If, for example, PU book 3 is not active (as would be the case in a 48-core system), both the green and amber LEDs would be unlit. If the System Attention LED is lit, a serviceable event has been detected and recorded by the system.Memory subsystemThe p5-590 and p5-595

33、memory controllers are internal to the POWER5+ or POWER5processor. The memory controller interfaces to four Synchronous Memory Interface II (SMI-II) buffers and eight DIMM cards per processor as shown in Figure 2-10. There are 16 memory card slots per processor book and each processor on an MCM owns

34、 a pair ofmemory cards. The p5-590 and p5-595 use Double Data Rate (DDR) DRAM memory cards. The two types of DDR memory used are the higher-speed DDR2 and DDR1. 电源子系统电源子系统BPA由BPR,BPD, BPC,BPE,BPF组成.1. BPR (bulk power regulator,电源整流器) 用于将380伏的交流电转换为380V的直流电2.BPD (bulk power distributor) 用于将380V的直流电分发

35、给主机的各个模块3.BPC(bulk power controller)是整个BPA中的电源控制器控制给各个部件的上电顺序4.BPF ( bulk power fan )负责给BPA制冷5.BPE (bulk power enclosure ) 一个金属笼子System buses - GX+ and RIO-2 busesThe processor module provides a GX+ bus that is used to connect to the I/O subsystem. Table 2-10 shows the relationship between RIO cards

36、, MCMs, and processors:Each processor book has eight GX+ slots that support communication with the I/O drawer.Each processor on an MCM can own one GX+ I/O card.Each GX+ I/O card has two RIO-2 ports. Remote I/O (RIO-2) links allow for connectivity to external I/O drawers and PCI-X technology.Internal

37、 I/O subsystemThe p5-590 and p5-595 use remote I/O drawers (that are 4U) for directly attached PCI or PCI-X adapters and SCSI disk capabilities. A minimum of one I/O drawer (FC 5791 or FC 5794) is required per system.The p5-590 supports up to eight I/O drawers, while the p5-595 supports up to 12 I/O

38、 drawers.Internal I/O subsystem - I/O drawer一个I/O柜是由two separate halves组成;每个I/O柜有20个PCI-X slots ,16块hot-swap DASD bays , 2块RIO-2卡 Each half of the I/O drawer is powered separately.HMC(Hardware Management Console)HMC is a dedicated workstation that provides a graphical user interface for configuring,

39、 operating, and performing basic system tasks for the POWER6 processor-based (as well as the POWER5 and POWER5+ processor-based) systems that function in either non-partitioned, partitioned, or clustered environments. In addition the HMC is used to configure and manage partitions. One HMC is capable

40、 of controlling multiple POWER5, POWER5+, and POWER6 processor-based systems.POWER5, POWER5+ and POWER6 processor-based system HMCs require Ethernet connectivity between the HMC and the servers service processor, moreover if dynamic LPAR operations are required, all AIX 5L, AIX V6, and Linux partitions must be enabled to communicate over the network to the HMC. Ensure that at least two Ethernet ports are available to enable public and private networks:HMC(Hardware Management Console)

展开阅读全文
相关资源
正为您匹配相似的精品文档
相关搜索

最新文档


当前位置:首页 > 商业/管理/HR > 营销创新

电脑版 |金锄头文库版权所有
经营许可证:蜀ICP备13022795号 | 川公网安备 51140202000112号