第7讲1大规模并行处理机系统MPP.ppt

第7讲 1 大规模并行处理机系统 MPP古志民千万亿次超级计算机-天河一号Tianhe-1 2009天河一号特点我国首台千万亿次超级计算机系统“天河一号”由国防科学技术大学研制成功。在今天中国高性能计在今天中国高性能计算机算机TOP100组织公布的组织公布的2009年度前年度前100强排名中，天强排名中，天河一号高居榜首。河一号高居榜首。有关专家认为，“天河一号”的诞生，是我国战略高技术和大型基础科技装备研制领域取得的又一重大创新成果，实现了我国自主研制超级计算机能力从百万亿次到千万亿次的跨越，使我国成我国成为继美国之后世界上第二个能够研制千万亿次超级计为继美国之后世界上第二个能够研制千万亿次超级计算机系统的国家。算机系统的国家。系统峰值性能达每秒系统峰值性能达每秒1206万亿次双精度浮点运算，内万亿次双精度浮点运算，内存总容量存总容量98TB，点点通信带宽每秒，点点通信带宽每秒40Gb，共享磁盘容，共享磁盘容量为量为1PB，具有高性能、高能效、高安全和易使用等显著特点，综合技术水平进入世界前列.IBM千万亿次超级计算机IBM千万亿次计算的超级计算机 IBM为美国洛斯阿拉莫斯国家实验室建造的计算机系统成为世界上首个突破每秒钟一千万亿次计算的超级计算机。排名前10名中有5个系统出自IBM；前50名中有17个系统出自IBM；前100名中有35个系统出自IBM，此外，上海超级计算中心的“曙光”5000A排名第15位。在500强榜单中，有188台超级计算机来自于IBM，却有212台超级计算机来自惠普。IBM for los national laboratory building computer system become the worlds first breakthrough one quadrillion times per second calculation of the super computer.The top 10 has five system from IBM;Top 50 has 17 system from IBM;In the first 100 has 35 system from IBM,in addition,Shanghai supercomputing center dawn 5000 a ranking 15th.In the 500 list,there are 188 sets of super computer from IBM,are 212 supercomputer from HP.1MPP(massively parallel processing)MPP(massively parallel processing)is the coordinated processing of a program by multiple processor s that work on different parts of the program,with each processor using its own operating system and memory.Typically,MPP processors communicate using some messaging interface.In some implementations,up to 200 or more processors can work on the same application.An interconnect arrangement of data paths allows messages to be sent between processors.Typically,the setup for MPP is more complicated,requiring thought about how to partition a common database among processors and how to assign work among the processors.An MPP system is also known as a loosely coupled or shared nothing system.An MPP system is considered better than a symmetrically Multi-processing system(SMP)for applications that allow a number of databases to be searched in parallel.These include decision support system and data warehouse applications.2 MPP Architecture 高速网络（高速网络（HSN）本地互连网络 NICP/C.P/C M磁盘和其他I/OSMP/SINGLE PROCESSORMPP with/without SMPSMP 2-64 processors today Shared-everything architecture All processors share all the global resources available Single copy of the OS runs on these systemsMPP A large parallel processing system with a shared-nothing architecture Consist of several hundred nodes with a high-speed interconnection network/switch Each node consists of a main memory&one or more processors Runs a separate copy of the OS3 可扩放性可扩放性scalability-If an application needs more MIPS or megabytes,additional processors can be added help solve the problem采用物理分布式主存结构distributed memory system；平衡的处理能力和主存与I/O能力，保证数据快速送入处理器；平衡的计算能力和并行性以及交互能力，保证进程/线程管理及通信与同步极小的开销；以上述条件为基础实施可扩放性。In a massively parallel processing system,current levels of technology allow for Thousands of processors per system Tens/Hundreds of Megabytes of RAM per processor Gigabytes of disk storage per processor Tens of Megabytes/sec global communication bandwidth per processor Hundreds of MIPS/MFLOPS per processor 4 系统成本System Cost需要控制MPP系统中每一部件成本，采取的措施：1利用Moor定律（每1824个月性能就翻一番）选用商用微处理器（为PC或小型系统或工作站设计）；2采用壳体系结构（用shell方法，系统其他部分无须改变），支持（微处理器）部件换代的可扩放性；然而也产生了问题：物理地址空间太小；TLB（Translation Look-aside Buffer）太小；单字（Single-Word Stride）存取效率很低等。The need to control MPP system in each unit cost,take measures:1 Use Moor law(every 18 24 months performance is doubling)choose commercial microprocessor(for PC or a small system or workstation design);2 The shell system structure(with shell method,system other part does not need to change)support(microprocessor)unit scalability;However also produced a problem:physical address space is too small;TLB(Translation Look-aside Buffer)is too small;Words(Single-Word Stride)access efficiency is very low.5 通用性和可用性支持MIMD；支持PVM、MPI、HPF；支持节点分区；高可用性；其他：支持通信需求；支持可扩放I/O性能；面临的问题(some difficulties)实际性能差:Rmax Rpeak；并行程序可编程性困难,need new programming tools；If the system is designed intelligently,the overall performance of the system(global communication bandwidth,MIPS,MFLOPS,etc.)will scale up linearly with the system size.It should be noted,though,that the degree to which performance can be extracted from a MPP system is very algorithm dependent.Undoubtedly the level of computing power available in a large MPP system will increase dramatically over time.Processor speeds and memory sizes are doubling approximately every eighteen months and this increase will be quickly adopted by MPP manufacturers.This means that the age of a Teraflop/Terabyte computer is not far off.Extremely large amounts of data will be able to be analyzed using this amount of processing power.7 实例1：Cray T3E体系结构NCC-NUMA+DSM 三维双向环网链接 I/O设备 千兆环通道Alpha21164主存控制和寄存器路由器shell8 Cray T3E性能300 MHZ ProcessorEach processor Rpeak=600Mflops62048 processorsSystem Rpeak=3.61228GflopsMemory size=14096GBMemory Rpeak=7.22450Gb/sNetwork Rpeak=600MB/s9 T3E系统软件与价格UNICOS/mk(64 bit UNIX)PVMMPIHPFC/C+Totalview并行程序调试器MPP Apprenti