Technical computing- Observations on an ever changing, .ppt

《Technical computing- Observations on an ever changing, .ppt》由会员分享，可在线阅读，更多相关《Technical computing- Observations on an ever changing, .ppt（36页珍藏版）》请在麦多课文档分享上搜索。

1、Technical computing: Observations on an ever changing, occasionally repetitious, environment,Los Alamos National Laboratory 17 May 2002,A brief, simplified history of HPC,Sequential & data parallelism using shared memory, Crays Fortran computers 60-02 (US:90) 1978: VAXen threaten general purpose cen

2、ters NSF response: form many centers 1988 - present SCI: Search for parallelism to exploit micros 85-95 Scalability: “bet the farm” on clusters. Users “adapt” to clusters aka multi-computers with LCD program model, MPI. 95 Beowulf Clusters adopt standardized hardware and Linuss software to create a

3、standard! 1995 “Do-it-yourself” Beowulfs impede new structures and threaten g.p. centers 2000 1997-2002: Lets tell NEC they arent “in step”. High speed networking enables peer2peer computing and the Grid. Will this really work?,Outline,Retracing scientific computing evolution: Cray, SCI BRC on Cyber

4、Infrastructure urges 650M/year Role of Grid and Peer-to-peer Will commodities drive out or enable new ideas?,High performance architecture/program timeline,1950 . 1960 . 1970 . 1980 . 1990 . 2000 Vtubes Trans. MSI(mini) Micro RISC nMicrSequential programming-(single execution stream)SIMD Vector-/-Pa

5、rallelization-Parallel programs aka Cluster Computing - multicomputers -MPP era- ultracomputers 10X in size & price! 10x MPP“in situ” resources 100x in /sm NOW CCgeographically dispersed Grid,DARPA SCI: c1985-1995; prelude to DOEs ASCI,Motivated by Japanese 5th Generation note the creation of MCC Re

6、alization that “killer micros” were Custom VLSI and its potential Lots of ideas to build various high performance computers Threat and potential sale to military,Steve Squires & G Bell at our “Cray” at the start of DARPAs SCI c1984.,What Is the System Architecture? (GB c1990),X,X,X,GRID,SIMD,X,Taxon

7、omy:The Architectural Alternatives for scalability & high performance c1991,MIMD,multicomputers (mC) (message passing),multiprocessors (mP) (shared memory),all are scalable,multi, mainframe, super (limited-scalable) smP - scalable, with Distd. Shared Memory network mP? (scalable, with DSM)multicompu

8、ters clusters workstations (ATM) workstations (LAN),SymmAsymm,Processor Architectures?,VECTORS,VECTORS,OR,CS View MISC CISC Language directed RISC Super-scalar Extra-Long Instruction Word Caches: mostly alleviate need for memory B/W,SC Designers View RISC VCISC (vectors) Massively parallel (SIMD) (m

9、ultiple pipelines) Memory B/W = perf.,The Bell-Hillis Bet c1991 Massive (1000) Parallelism in 1995,TMC World-wide Supers,TMC World-wideSupers,TMCWorld-wide Supers,Applications,Revenue,Petaflops / mo.,Results from DARPAs SCI c1983,Many research and construction efforts virtually all new hardware effo

10、rts failed except Intel and Cray. DARPA directed purchases screwed up the market, including the many VC funded efforts. No Software funding! Users responded to the massive power potential with LCD software. Clusters, clusters, clusters using MPI. Its not scalar vs vector, its memory bandwidth! 6-10

11、scalar processors = 1 vector unit 16-64 scalars = a 2 6 processor SMP,Dead Supercomputer Society,ACRI Alliant American Supercomputer Ametek Applied Dynamics Astronautics BBN CDC Convex Cray Computer Cray Research Culler-Harris Culler Scientific Cydrome Dana/Ardent/Stellar/Stardent Denelcor Elexsi ET

12、A Systems Evans and Sutherland Computer Floating Point Systems Galaxy YH-1,Goodyear Aerospace MPP Gould NPL Guiltech Intel Scientific Computers International Parallel Machines Kendall Square Research Key Computer Laboratories MasPar Meiko Multiflow Myrias Numerix Prisma Tera Thinking Machines Saxpy

13、Scientific Computer Systems (SCS) Soviet Supercomputers Supertek Supercomputer Systems Suprenum Vitesse Electronics,The evolution of Cray Inc.,SELL SX5s,What a difference 25 years AND spending 10x makes!,LLNL 150 Mflops machine room c1978,ESRDC: 40 Tflops. 640 nodes (8 - 8GFl P.vec/node),Computer ty

14、pes,Netwrked Supers,Legion CondorBeowulfNT clusters,VPPuni,T3ESP2(mP) NOW,NEC mP,SGI DSM clusters & SGI DSM,NEC super Cray XT (all mPv),Mainframes Multis WSs PCs,- Connectivity- WAN/LAN SAN DSM SM,micros vector,Clusters,GRID & P2P,Old World,Top500 taxonomy everything is a cluster aka multicomputer,C

15、lusters are the ONLY scalable structure Cluster: n, inter-connected computer nodes operating as one system. Nodes: uni- or SMP. Processor types: scalar or vector. MPP= miscellaneous, not massive (1000), SIMD or something we couldnt name Cluster types. Implied message passing. Constellations = cluste

16、rs of =16 P, SMP Commodity clusters of uni or =4 Ps, SMP DSM: NUMA (and COMA) SMPs and constellations DMA clusters (direct memory access) vs msg. pass Uni- and SMPvector clusters: Vector Clusters and Vector Constellations,Linux - a web phenomenon,Linus Tovald - writes news reader for his PC Puts it

17、on the internet for others to play Others add to it contributing to open source software Beowulf adopts early Linux Beowulf adds Ethernet drivers for essentially all NICs Beowulf adds channel bonding to kernel Red Hat distributes Linux with Beowulf software Low level Beowulf cluster management tools

18、 added,The Challenge leading to Beowulf,NASA HPCC Program begun in 1992 Comprised Computational Aero-Science and Earth and Space Science (ESS) Driven by need for post processing data manipulation and visualization of large data sets Conventional techniques imposed long user response time and shared

19、resource contention Cost low enough for dedicated single-user platform Requirement: 1 Gflops peak, 10 Gbyte, $50K Commercial systems: $1000/Mflops or 1M/Gflops,Innovation,The Virtuous Economic Cycle drives the PC industry & Beowulf,Volume,Competition,Standards,Utility/value,DOJ,Greater availability

20、lower cost,Creates apps, tools, training,Attracts users,Attracts suppliers,Lessons from Beowulf,An experiment in parallel computing systems Established vision- low cost high end computing Demonstrated effectiveness of PC clusters for some (not all) classes of applications Provided networking softwar

21、e Provided cluster management tools Conveyed findings to broad community Tutorials and the book Provided design standard to rally community! Standards beget: books, trained people, software virtuous cycle that allowed apps to form Industry begins to form beyond a research project,Courtesy, Thomas St

22、erling, Caltech.,Clusters: Next Steps,Scalability They can exist at all levels: personal, group, centers Clusters challenge centers given that smaller users get small clusters,Disk Evolution,Capacity:100x in 10 years 1 TB 3.5” in 2005 20 TB? in 2012?! System on a chip High-speed SAN Disk replacing t

23、ape Disk is super computer!,Kilo Mega Giga Tera Peta Exa Zetta Yotta,Intermediate Step: Shared Logic,Brick with 8-12 disk drives 200 mips/arm (or more) 2xGbpsEthernet General purpose OS 10k$/TB to 100k$/TB Shared Sheet metal Power Support/Config Security Network ports These bricks could run applicat

24、ions e.g. SQL, Mail,Snap 1TB 12x80GB NAS,NetApp .5TB 8x70GB NAS,Maxstor 2TB 12x160GB NAS,IBM TotalStorage 360GB10x36GB NAS,SNAP Architecture-,RLX “cluster” in a cabinet,366 servers per 44U cabinet Single processor 2 - 30 GB/computer (24 TBytes) 2 - 100 Mbps Ethernets 10x perf*, power, disk, I/O per

25、cabinet 3x price/perf Network services Linux based*42, 2 processors, 84 Ethernet, 3 TBytes,Computing in small spaces LANL (RLX cluster in building with NO A/C),240 processors 2/3 GFlopsFill the 4 racks - gives a Teraflops,Beowulf Clusters: space,Beowulf clusters: power,“The networks becomes the syst

26、em.”- Bell 2/10/82 Ethernet announcement with Noyce (Intel), and Liddle (Xerox) “The network become the computer.” SUN Slogan 1982 “The network becomes the system.” GRID mantra c1999,Computing SNAP built entirely from PCs,Wide & Local Area Networks for: terminal, PC, workstation, & servers,Centraliz

27、ed & departmental uni- & mP servers (UNIX & NT),Legacy mainframes & minicomputers servers & terms,Wide-area global network,Legacy mainframe & minicomputer servers & terminals,Centralized & departmental servers buit from PCs,scalable computers built from PCs,TC=TV+PC home . (CATV or ATM or satellite)

28、,?,Portables,A space, time (bandwidth), & generation scalable environment,Person servers(PCs),Person servers(PCs),Mobile Nets,Telnet & FTP EMAIL,Standards,The virtuous cycle of bandwidth supply and demand,Incompence ?,Internet II concerns given $0.5B cost,Very high cost $(1 + 1) / GByte to send on t

29、he net; Fedex and 160 GByte shipments are cheaper DSL at home is $0.15 - $0.30 Disks cost $1/GByte to purchase! Low availability of fast links (last mile problem) Labs very poor communication links,Collaborative research sharing instrumentation, data, and programs,Weve talked about it for decades e.

30、g. accelerators to telelescopes and zoology Doer / “User, talker & meeter” = 4%. http:/www.all-species.org/ has the problem NSF focus has been and is on ops not bytes! E.g. Pittsburgh center funded with no storage Why have centers for computation at all? Dont we need datacenter? By having no storage

31、, re-compute everything Adding indexes i.e. databases, increases speed, lessens computation, and increases experimentation Computation centers become data centers since everyone/anyone builds a center Need for computational scientist database talent! Big question: Can distributed computing form to p

32、rovide something better than a “center” can provide?,Scalable computing: the effects,They come in all sizes; incremental growth 10 or 100 to 10,000 (100X for most users) debug vs run; problem growth Allows compatibility heretofore impossible 1978: VAX chose Cray Fortran 1987: The NSF centers went to

33、 UNIX Users chose sensible environment Acquisition and operational costs & environments Cost to use as measured by users time The role of gp centers e.g. NSF, statex is unclear. Necessity for support? Scientific Data for a given community Community programs and data Manage GRIDdiscipline Are clusters Greshams Law? Drive out alts.,The end,