Grid Computing 7700 Fall 2005 Lecture 4 Scientific

Grid Computing 7700 Fall 2005 Lecture 4: Scientific Computing and Hardware Gabrielle Allen allen@bit. csc. lsu. edu http: //www. cct. lsu. edu/~gallen

Basic Elements Wide Area Network Machine Network CPU CPU DISK Campus Network (LAN)

Basic Elements q Distributed systems built from – Computing elements (processors) – Communication elements (networks) – Storage elements (disk, attached or networked) q New elements – Visualization/interactive devices – Experimental and operational devices

Distributed Resources q q q Local workstations CCT Resources Campus/OCS Resources State/LONI Resources National Centers International Colleagues

Laws q q q Moores Law – – Number of transistors on an integrated circuit will double every 18 months http: //en. wikipedia. org/wiki/Moores_law – – Hard disk capacity grows quicker than transistors http: //www. sciam. com/article. cfm? chan. ID=sa 006&col. ID=30&article. ID=000 B 0 C 22 -0805 -12 D 8 -BDFD 83414 B 7 F 0000 “Kryders Law” Gilders Law – q Total bandwidth of communication systems doubles every six months Metcalfe’s Law – Value of a network is proportional to the square of the number of nodes q Amdahl’s Law – Law of diminishing returns, maximum speedup restricted by slowest parts – http: //en. wikipedia. org/wiki/Amdahls_law q Question: So what about applications?

Compute Elements q Moore’s Law: #transistors on a chip (and clock speed) increase exponentially (double every 18 months) – Transistors = 20*2^[(year-1965)/1. 5] – 1975 Intel 8080 has 4500 transistors, 100 K intructions/sec – 2003 Pentium IV has 221, 000, 8 billion instructions/sec q q Corollary: Price of a given level of supercomputing power halves every 18 months Price decrease means that supercomputers now usually built from “commodity” processors – IA 32, Power. PC, “emotion engine”

Compute Elements q q q q Clock speed Cache hierarchy Floating point registers Main memory Internal bandwidths Etc, etc Need powerful operating systems, compilers, applications to leverage all this

Communication Elements q q Links, routers, switches, name servers, protocols Infrastructure evolves slowly (politics, large scale changes, money) Gilder's Law: total bandwidth of communication systems doubles every six months Change in LAN to desktops – – q 100 mbps shared 100 mbps switched 1 gbps 10 gbps Clusters: Gig. E (TCP/IP and MPICH/LAM) standard, Myricom/Quadrics (own MPI drivers) better performance, infiniband/fibrechannel different architecture

Network Speeds q q q q q Analog modem: 57 kbps GPRS: 114 kbps Bluetooth: 723 kbps T-1: 1. 5 Mbps Eth 10 Base-X: 10 Mbps 802. 11 b (Wi. Fi) 11 Mbps T-3: 45 Mbps OC-1: 52 Mbps Fast Eth 100 Base-X: 100 Mbps q OC-12: 622 Mbps Gig. Eth 1000 Base-X: 1 Gbps OC-24: 1. 2 Gbps OC-48: 2. 5 Gbps OC-192: 10 Gbps 10 Gig. Eth: 10 Gbps OC-3072: 160 Gbps q My Cox Cable q q q – Upload: 35 KB/s – Download 250 KB/s q CCT “is” to supermike – Up/down: 5000 KB/s

Communication Elements Interconnect Type Short Message Latency (microsec) Peak Bandwidth (mbps) Bidirectional Bandwidth (mbps) Approximate cost per port Gigabit Ethernet 100 ~65 ~130 $100 Myrinet 9 280 500 $1000 Quadrics 5 300 500 $3000

Storage Elements q q Magnetic tape/Magnetic disk – – – q Properties: density/rotation/cost 1970 -1988 density improvements 29% per year 1988 -now density improvements 60% per year Standard in PCs: 500 mb (1995), 2 gb(1997), 100 gb (2002) Performance not increasing so fast • Peak transfer (~100 mbs) • Seek times (3 -5 ms) [bottleneck] Grids: cost of storage neglibable, high speed networks make large data libraries attractive

The Future (? ? ) Machine Compute Memory Disk Network 2003 PC 8 g-op/s 512 mb 128 gb 1 gb/s 2003 SC 80 t-op/s 50 tb 1280 pb 10 tb/s 2008 PC 64 g-op/s 16 gb 2 tb 10 gb/s 2008 SC 640 t-op/s 160 tb 20 pb 100 tb/s 2013 PC 512 g-op/s 256 gb 32 tb 100 gb/s 2013 SC 5 p-op/s 2. 6 pb 320 pb 1 pb/s 1 mega = 10^6 1 giga = 10^9 1 tera = 10^12 1 peta = 10^15 DOE Blue. Gene: 367 TFlop/s 16 TB memory 400 Terabyte storage Earth Simulator: 40 TFlop/s 10 TB memory 2. 5 Petabytes storage 13 Gigabits/s Tera. Grid: 40 TFlop/s 6 TB memory 1 Petabytes storage 10 Gigabits/s

Supercomputers q Definition of supercomputer – Machine on top 500. org ? • http: //www. top 500. org/lists/plists. php? Y=2005&M=06 – Machine costing over $1 M ? – Basically highest end machines q Top 3 (2005) – DOE Blue. Gene/L (USA) 66 K procs/137 TF – IBM BGW (USA) 41 K procs/91 TF – NASA Columbia (USA) 10 K procs/52 TF q Top 3 (2003) – Earth Simulator (JAPAN) 5 K procs/36 TF (6) – ASCI Q (USA) 8 K procs/14 TF (12) – G 5 Cluster (USA) 2 k procs/12 TF (14) q Others – 18 IBM (China) – 147 Supermike (LSU !!!) www. webopedia. com The fastest type of computer. Supercomputers are very expensive and are employed for specializedapplications that require immense amounts of mathematical calculations. For example, weather forecasting requires a supercomputer. Other uses of supercomputers include animated graphics, fluid dynamic calculations, nuclear energy research, and petroleum exploration. The chief difference between a supercomputer and a mainframe is that a supercomputer channels all its power into executing a few programs as fast as possible, whereas a mainframe uses its power to execute many programs concurrently.

Architectural Classes Flynn (1972): classification based on the way system manipulates instruction and data streams: q SISD Single Instruction Single Data – One instruction stream executed serially. – Conventional workstations q SIMD Single Instruction Multiple Data – Large (many thousands) number of processing units – All execute same instruction on different data in lockstep – Vector processors (NEC SX-6 i) acting on arrays of data q MISD Multiple Instruction Single Data – No machines built q MIMD Multiple Instruction Multiple Data – Different to SISD because instructions/data are related

More Classification q Shared Memory Systems – – – q Multiple CPUs sharing same address space One memory accessed by all processors equally Location of data not important to user Can be SIMD (single processor vector processor) or MIMD Open. MP http: //www. openmp. org/index. cgi? faq Distributed Memory Systems – – Each CPU has own memory CPUs are connected by network Location of data important Can be SIMD (lock step example before) or MIMD (large variety of network topologies) – Distributed processing takes DM-MIMD to extreme

Message Passing q Essential for DM machines, but often also used for SM machines for compatibility – MPI Message Passing interface – PVM Parallel Virtual Machine

DM-MIMD q q q q Fast growing section, best performance. Need to balance computation and communication performance in machine design (and upgrades) User has to distribute data between processors User has to perform data exchange between processors explicitly Slow compared to SM machines to access data on other processors Programming models/algorithms important Programming environments can make this easier (e. g. Cactus Framework http: //www. cactuscode. org handles data distribution, communications, IO, …) Same programming models need to be extended to Grid computing

cc. NUMA q q q Cache Coherent Non Uniform Memory Access Build systems from SMPs (symmetric multiprocessing nodes) SMPs consist of up to ~16 processors connected by a crossbar which share same memory Each node is a SM-MIMD, but with different memory access times for different processors (memory is physically distributed) Nodes then connecting in a different way Computational scientists like these machines

DM-MIMD q Processor topology and interconnects very important – Hypercube (with 2^d nodes number of steps between two nodes at most d, possible to simulate other topologies) – Fat tree (simple tree structure with more connections at higher levels to ease conjestion) – 2 D/3 D mesh structure (many apps map well to this, avoids expense) – Crossbars (connecting up to around 64 processors, can be hierarchical) q Details should be hidden from application programmers, but for performance need to be aware

Virtual Shared Memory q q Kendall Square Research Systems tried to implement at hardware level High Performance Fortran – – q HPF Specification 1993 Simulates a virtual shared memory at a software level Programming directives distribute data across processors Looks like shared memory machine to user Some vendors have propriety virtual shared memory programming models by providing global address space

Network Eras q Past (1969 -1988) – ARPANET/NSFNET q Current (1988 -2005) Future (2005 -) q Historical network maps q – http: //www. cybergeography. org/atlas/historical. html

Network Infrastructure q q Chapter 30 (The Grid 2) Network infrastructure is the foundation on which Grids are built Composition of local and wide area services, transport protocols and services, routing protocols and network services, link protocols and physical media One example of network infrastructure in the Internet (core protocols TCP/IP)

Protocol q Agreed-upon format for transmitting data between two devices which determines: – – q q The type of error checking to be used Any data compression method How sending device indicates it has finished sending a message How receiving device indicates it has received a message Various standard protocols: differ in simplicity, reliability, performance. Computer/device must support the right ones to communicate with other computers. Implemented either in hardware or in software http: //www. protocols. com/protocols. htm

Slow to Change q Internet has not changed much since 1983 (when TCP/IP deployed), which does make is stable, but still don’t really have envisaged services: – Multicast (one-to-many communication) – Network Reservation – Quality of Service q q New protocols peer-to-peer file sharing and instant messaging New technology coupled to applications drive change: e-mail, web/file-sharing, video streaming

Past: 1969 -1988 q ARPANET (1969) 56 -kbps lines – Experiment to investigate resource sharing and remote access – Added interface message processor (IMP) at each end of network (our routers), provided flexibility for lower levels and higher level applications – Success from: freely available documentation and source code; software bundled with new machines; use for teaching; community development vs. proprietary q NSFNET (1985) 45 -mpbs lines – Connect academic HPC centers

ARPANET: 1971

ARPANET: 1980

NSFNET: 1991

Past: 1969 -1988 q q Driving application: e-mail, remote file access, remote job control (drove basic protocols) Network technology: WAN links lines leased from telephone companies. Xerox Palo Alto Research Center (PARC) created Ethernet (3 mbps) (alternatives token ring (IBM), …). Workstations appear bundled with network protocols. PCs on the network as interface costs dropped and processors became more powerful.

Past: 1969 -1988 q Protocols and Services – telnet, file transfer protocol, e-mail – Underlying transport protocol TCP (stream of bytes which can be opened or closed, data can be sent or received) – Machine location: Domain Name System (DNS) (replaced list of named files) • Hierarchical, distributed, redundant

Past: 1969 -1988 q System Integration – ARPANET: assumed central network operations center – NSFNET: introduced hierarchical system, toplevel backbone network connecting to regional networks connecting to campuses q q Packet switching strategy was important (using computing power to optimize communication) Single communication model was important because it allowed so many people to be connected driving future development.

Present: 1988 -2005 q q Internet today: complex structure of backbone networks and regional networks Increased role of private sector (e. g. AT&T, Bell. South), who basically control our network now.

LSU Campus

LANet q q Louisiana statewide network: Office of Telecommunications Management, state agencies, higher education: 6 Mbps -> $2450 a month http: //www. state. la. us/otm/lanet/

Quest

Bell South Baton Rouge: 4 DS 3 to New Orleans, 1 DS 3 to Houston

Abeline (Internet 2) http: //abilene. internet 2. edu/maps-lists/ Traffic: http: //loadrunner. uits. iu. edu/weathermaps/abilene/

National Lambda Rail http: //www. nationallambdarail. org/architecture. html

National Lambda Rail

Global Terabit Research Network

Required Reading q Overview of Recent Supercomputers – http: //www. euroben. nl/reports/overview 05 a. pdf q Concentrate on pages 1 to 32, you do not need to learn this, just get an appreciation of the concepts.