1. CSCI-455/522 Introduction to High Performance Computing Lecture 2

2. Slides for Parallel Programming Techniques & Applications Using Networked Workstations & Parallel Computers 2nd Edition, by B. Wilkinson & M. Allen, © 2004 Pearson Education Inc. All rights reserved. 1.2 Types of Parallel Computers Two principal types: Shared memory multiprocessor Distributed memory multicomputer

3. PPPPPP BUS Memory M P M P M P M P M P M P Network Shared memory - single address space. All processors have access to a pool of shared memory. (Ex: SGI Origin, Sun E10000) Distributed memory - each processor has it’s own local memory. Must do message passing to exchange data between processors. (Ex: CRAY T3E, IBM SP, clusters) Shared vs. Distributed Memory

4. Slides for Parallel Programming Techniques & Applications Using Networked Workstations & Parallel Computers 2nd Edition, by B. Wilkinson & M. Allen, © 2004 Pearson Education Inc. All rights reserved. 1.4 Shared Memory Multiprocessor

5. Slides for Parallel Programming Techniques & Applications Using Networked Workstations & Parallel Computers 2nd Edition, by B. Wilkinson & M. Allen, © 2004 Pearson Education Inc. All rights reserved. 1.5 Conventional Computer Consists of a processor executing a program stored in a (main) memory: Each main memory location located by its address. Addresses start at 0 and extend to 2 b - 1 when there are b bits (binary digits) in address. Main memory Processor Instructions (to processor) Data (to or from processor)

6. Slides for Parallel Programming Techniques & Applications Using Networked Workstations & Parallel Computers 2nd Edition, by B. Wilkinson & M. Allen, © 2004 Pearson Education Inc. All rights reserved. 1.6 Shared Memory Multiprocessor System Natural way to extend single processor model - have multiple processors connected to multiple memory modules, such that each processor can access any memory module : Processors Interconnection network Memory module One address space

7. Slides for Parallel Programming Techniques & Applications Using Networked Workstations & Parallel Computers 2nd Edition, by B. Wilkinson & M. Allen, © 2004 Pearson Education Inc. All rights reserved. 1.7 Simplistic View of a Small Shared Memory Multiprocessor Examples: Dual Pentiums Quad Pentiums ProcessorsShared memory Bus

8. Slides for Parallel Programming Techniques & Applications Using Networked Workstations & Parallel Computers 2nd Edition, by B. Wilkinson & M. Allen, © 2004 Pearson Education Inc. All rights reserved. 1.8 Quad Pentium Shared Memory Multiprocessor Processor L2 Cache Bus interface L1 cache Processor L2 Cache Bus interface L1 cache Processor L2 Cache Bus interface L1 cache Processor L2 Cache Bus interface L1 cache Memory controller Memory I/O interface I/O bus Processor/ memory bus Shared memory

9. Slides for Parallel Programming Techniques & Applications Using Networked Workstations & Parallel Computers 2nd Edition, by B. Wilkinson & M. Allen, © 2004 Pearson Education Inc. All rights reserved. 1.9 Programming Shared Memory Multiprocessors Threads - programmer decomposes program into individual parallel sequences, (threads), each being able to access variables declared outside threads. Example Pthreads Sequential programming language with preprocessor compiler directives to declare shared variables and specify parallelism. Example OpenMP - industry standard - needs OpenMP compiler

10. Slides for Parallel Programming Techniques & Applications Using Networked Workstations & Parallel Computers 2nd Edition, by B. Wilkinson & M. Allen, © 2004 Pearson Education Inc. All rights reserved. 1.10 Sequential programming language with added syntax to declare shared variables and specify parallelism. Example UPC (Unified Parallel C) - needs a UPC compiler. Parallel programming language with syntax to express parallelism - compiler creates executable code for each processor (not now common) Sequential programming language and ask parallelizing compiler to convert it into parallel executable code. - also not now common

11. PPPPPP BUS Memory Uniform memory access (UMA): Each processor has uniform access to memory. Also known as symmetric multiprocessors, or SMPs (Sun E10000) PPPP BUS Memory PPPP BUS Memory Network Non-uniform memory access (NUMA): Time for memory access depends on location of data. Local access is faster than non-local access. Easier to scale than SMPs (SGI Origin) Shared Memory: UMA vs. NUMA

12. Slides for Parallel Programming Techniques & Applications Using Networked Workstations & Parallel Computers 2nd Edition, by B. Wilkinson & M. Allen, © 2004 Pearson Education Inc. All rights reserved. 1.12 Distributed Memory /Message-Passing Multicomputers

13. Slides for Parallel Programming Techniques & Applications Using Networked Workstations & Parallel Computers 2nd Edition, by B. Wilkinson & M. Allen, © 2004 Pearson Education Inc. All rights reserved. 1.13 Message-Passing Multicomputer Complete computers connected through an interconnection network: Processor Interconnection network Local Computers Messages memory

14. Distributed Memory: MPPs vs. Clusters Processor-memory nodes are connected by some type of interconnect network –Massively Parallel Processor (MPP): tightly integrated, single system image. –Cluster: individual computers connected by s/w CPU MEM CPU MEM CPU MEM CPU MEM CPU MEM CPU MEM CPU MEM CPU MEM CPU MEM Interconnect Network

15. Distributed Memory: MPPs vs. Clusters Processor-memory nodes are connected by some type of interconnect network –Massively Parallel Processor (MPP): tightly integrated, single system image. –Cluster: individual computers connected by s/w CPU MEM CPU MEM CPU MEM CPU MEM CPU MEM CPU MEM CPU MEM CPU MEM CPU MEM Interconnect Network

16. Clusters Similar to MPPs –Commodity processors and memory Processor performance must be maximized –Memory hierarchy includes remote memory –No shared memory--message passing Communication overhead must be minimized Different from MPPs –All commodity, including interconnect and OS –Multiple independent systems: more robust –Separate I/O systems

17. Slides for Parallel Programming Techniques & Applications Using Networked Workstations & Parallel Computers 2nd Edition, by B. Wilkinson & M. Allen, © 2004 Pearson Education Inc. All rights reserved. 1.17 Interconnection Networks Limited and exhaustive interconnections 2- and 3-dimensional meshes Hypercube (not now common) Using Switches: –Crossbar –Trees –Multistage interconnection networks

18. Slides for Parallel Programming Techniques & Applications Using Networked Workstations & Parallel Computers 2nd Edition, by B. Wilkinson & M. Allen, © 2004 Pearson Education Inc. All rights reserved. Communications Networks Custom –Many vendors have custom interconnects that provide high performance for their system, specially MPP –CRAY T3E interconnect is the fastest for MPPs: lowest latency, highest bandwidth Commodity –Used in some MPPs and all clusters –Myrinet, Gigabit Ethernet, Fast Ethernet, etc.

19. Slides for Parallel Programming Techniques & Applications Using Networked Workstations & Parallel Computers 2nd Edition, by B. Wilkinson & M. Allen, © 2004 Pearson Education Inc. All rights reserved. Types of Interconnects Fully connected –not feasible Array and torus –Intel Paragon (2D array), CRAY T3E (3D torus) Crossbar –IBM SP (8 nodes) Hypercube –SGI Origin 2000 (hypercube), Meiko CS-2 (fat tree) Combinations of some of the above –IBM SP (crossbar & fully connected for 80 nodes) –IBM SP (fat tree for > 80 nodes)

20. Slides for Parallel Programming Techniques & Applications Using Networked Workstations & Parallel Computers 2nd Edition, by B. Wilkinson & M. Allen, © 2004 Pearson Education Inc. All rights reserved.

22. 1.22 Two-dimensional Array (Mesh) Also three-dimensional - used in some large high performance systems. Links Computer/ processor

23. Slides for Parallel Programming Techniques & Applications Using Networked Workstations & Parallel Computers 2nd Edition, by B. Wilkinson & M. Allen, © 2004 Pearson Education Inc. All rights reserved. 1.23 Three-dimensional Hypercube

24. Slides for Parallel Programming Techniques & Applications Using Networked Workstations & Parallel Computers 2nd Edition, by B. Wilkinson & M. Allen, © 2004 Pearson Education Inc. All rights reserved. 1.24 Four-dimensional Hypercube Hypercubes popular in 1980’s - not now

25. Slides for Parallel Programming Techniques & Applications Using Networked Workstations & Parallel Computers 2nd Edition, by B. Wilkinson & M. Allen, © 2004 Pearson Education Inc. All rights reserved. 1.25 Crossbar Switch Switches Processors Memories

26. Slides for Parallel Programming Techniques & Applications Using Networked Workstations & Parallel Computers 2nd Edition, by B. Wilkinson & M. Allen, © 2004 Pearson Education Inc. All rights reserved. 1.26 Tree Switch element Root Links Processors

27. Slides for Parallel Programming Techniques & Applications Using Networked Workstations & Parallel Computers 2nd Edition, by B. Wilkinson & M. Allen, © 2004 Pearson Education Inc. All rights reserved. 1.27 Multistage Interconnection Network Example: Omega network 000 001 010 011 100 101 110 111 000 001 010 011 100 101 110 111 Inputs Outputs 2´ 2 switch elements (straight-through or crossover connections)

28. Slides for Parallel Programming Techniques & Applications Using Networked Workstations & Parallel Computers 2nd Edition, by B. Wilkinson & M. Allen, © 2004 Pearson Education Inc. All rights reserved.

35. 1.35 Distributed Shared Memory Making main memory of group of interconnected computers look as though a single memory with single address space. Then can use shared memory programming techniques. Processor Interconnection network Shared Computers Messages memory

36. Slides for Parallel Programming Techniques & Applications Using Networked Workstations & Parallel Computers 2nd Edition, by B. Wilkinson & M. Allen, © 2004 Pearson Education Inc. All rights reserved. 1.36 Flynn’s Classifications Flynn (1966) created a classification for computers based upon instruction streams and data streams: –Single instruction stream-single data stream (SISD) computer Single processor computer - single stream of instructions generated from program. Instructions operate upon a single stream of data items.

37. Slides for Parallel Programming Techniques & Applications Using Networked Workstations & Parallel Computers 2nd Edition, by B. Wilkinson & M. Allen, © 2004 Pearson Education Inc. All rights reserved. 1.37 Single Instruction Stream-Multiple Data Stream (SIMD) Computer A specially designed computer - a single instruction stream from a single program, but multiple data streams exist. Instructions from program broadcast to more than one processor. Each processor executes same instruction in synchronism, but using different data. Developed because a number of important applications that mostly operate upon arrays of data.

38. Slides for Parallel Programming Techniques & Applications Using Networked Workstations & Parallel Computers 2nd Edition, by B. Wilkinson & M. Allen, © 2004 Pearson Education Inc. All rights reserved. 1.38 Multiple Instruction Stream-Multiple Data Stream (MIMD) Computer General-purpose multiprocessor system - each processor has a separate program and one instruction stream is generated from each program for each processor. Each instruction operates upon different data. Both the shared memory and the message- passing multiprocessors so far described are in the MIMD classification.

39. The Banking Analogy Tellers: Parallel Processors Customers: tasks Transactions: operations Accounts: data

40. Vector/Array Each teller/processor gets a very fine-grained task Use pipeline parallelism Good for handling batches when operations can be broken down into fine- grained stages

41. SIMD (Single-Instruction-Multiple- Data) All processors do the same things or idle Phase 1: data partitioning and distributed Phase 2: data-parallel processing Efficient for big, regular data- sets

42. Systolic Array Combination of SIMD and Pipeline parallelism 2-d array of processors with memory at the boundary Tighter coordination between processors Achieve very high speeds by circulating data among processors before returning to memory

43. MIMD(Multi-Instruction-Multiple-Data) Each processor (teller) operates independently Need synchronization mechanism –by message passing –or mutual exclusion (locks) Best suited for large-grained problems Less than data-flow parallelism

44. Slides for Parallel Programming Techniques & Applications Using Networked Workstations & Parallel Computers 2nd Edition, by B. Wilkinson & M. Allen, © 2004 Pearson Education Inc. All rights reserved. 1.44 Networked Computers as a Computing Platform A network of computers became a very attractive alternative to expensive supercomputers and parallel computer systems for high-performance computing in early 1990’s. Several early projects. Notable: – Berkeley NOW (network of workstations) project. –NASA Beowulf project.

45. Slides for Parallel Programming Techniques & Applications Using Networked Workstations & Parallel Computers 2nd Edition, by B. Wilkinson & M. Allen, © 2004 Pearson Education Inc. All rights reserved. 1.45 Key advantages: Very high performance workstations and PCs readily available at low cost. The latest processors can easily be incorporated into the system as they become available. Existing software can be used or modified.

46. Slides for Parallel Programming Techniques & Applications Using Networked Workstations & Parallel Computers 2nd Edition, by B. Wilkinson & M. Allen, © 2004 Pearson Education Inc. All rights reserved. 1.46 Software Tools for Clusters Based upon Message Passing Parallel Programming: Parallel Virtual Machine (PVM) - developed in late 1980’s. Became very popular. Message-Passing Interface (MPI) - standard defined in 1990s. Both provide a set of user-level libraries for message passing. Use with regular programming languages (C, C++,...).

47. Slides for Parallel Programming Techniques & Applications Using Networked Workstations & Parallel Computers 2nd Edition, by B. Wilkinson & M. Allen, © 2004 Pearson Education Inc. All rights reserved. 1.47 Beowulf Clusters* A group of interconnected “commodity” computers achieving high performance with low cost. Typically using commodity interconnects - high speed Ethernet, and Linux OS. * Beowulf comes from name given by NASA Goddard Space Flight Center cluster project.

48. Slides for Parallel Programming Techniques & Applications Using Networked Workstations & Parallel Computers 2nd Edition, by B. Wilkinson & M. Allen, © 2004 Pearson Education Inc. All rights reserved. 1.48 Cluster Interconnects Originally fast Ethernet on low cost clusters Gigabit Ethernet - easy upgrade path More Specialized/Higher Performance Myrinet - 2.4 Gbits/sec - disadvantage: single vendor cLan SCI (Scalable Coherent Interface) QNet Infiniband - may be important as infininband interfaces may be integrated on next generation PCs

49. Slides for Parallel Programming Techniques & Applications Using Networked Workstations & Parallel Computers 2nd Edition, by B. Wilkinson & M. Allen, © 2004 Pearson Education Inc. All rights reserved. 1.49 Dedicated cluster with a master node Dedicated Cluster User Switch Master node Compute nodes Up link 2nd Ethernet interface External network