BlueGene/L. Computer Science, University of Warwick. Source: IBM

Size: px
Start display at page:

Download "BlueGene/L. Computer Science, University of Warwick. Source: IBM"

Transcription

1 BlueGene/L Source: IBM 1

2 BlueGene/L networking BlueGene system employs various network types. Central is the torus interconnection network: 3D torus with wrap-around. Each node connects to six neighbours (bidirectional). Routing achieved in hardware. each link with 1.4 Gbit/s. 1.4 x 6 x 2= 16.8 Gbit/s aggregate bandwidth 2

3 BlueGene/L Other three networks: Binary combining tree Used for global operations - reductions, sums, products, barriers etc. Low latency (2μS) Gigabit Ethernet I/O network Support file I/O Diagnostic & control network Booting nodes, monitoring processors. Each chip has the above four network interfaces (torus, tree, i/o, diagnostics) Note specialised networks are used for different purposes - quite different from many other HPC cluster architectures. 3

4 BlueGene/L Message Passing: The BlueGene focussed a good deal of energy developing an efficient MPI implementation to reduce latency in the software stack. Using the MPICH code-base as a start-point: MPI library was enhanced with respect to machine architecture. For example, using the combining tree for reductions & broadcasts. Reading paper: Filtering Failure Logs for a BlueGene/L Prototype 4

5 ASCI Q The Q supercomputing system at Los Alamos National Laboratory (LANL) Product of Advanced Simulation and Computing (ASCI) program Used for simulation and computational modelling Now No.90 (last year No. 40) in Top500 supercomputer list 5

6 ASCI Q Classical cluster architecture SMPs (AlphaServer ES45s from HP) are put in one segment Each with four EV Ghz CPUs with 16-MB cache the whole system has 3 segments The three segments can operate independently or as a single system Aggregate 60 TeraFLOPS capability. 33 Terabytes of memory 664 TB of global storage Interconnection using Quadrics dual-rail switch interconnect (QSNet) High bandwidth (250MB/s/rail) and Low latency (5us) network. Top500 list: 6

7 Earth Simulator Built by NEC, located in the Earth Simulator Centre in Japan Used for running global climate models to evaluate the effects of global warming The fastest supercomputer from Now No.30 in the Top500 supercomputer list 7

8 Earth Simulator 640 nodes, each with 8 vector processors and 16GB memory Two nodes are installed in one cabinet In total: 5120 processors (NEC SX-5) 10 TeraByte memory 700 TeraByte of disk storage and 1.6 PetaByte of Tape storage Computing capacity: 36 TFlop/s Networking: Crossbar interconnection (very expensive) Bandwidth: 16GB/s between any two nodes Latency: 5us Dual level parallelism: OpenMP in-node, MPI out of node Physical installation: Machine resides on 3th floor; Cables on 2nd ; Power generation & cooling on 1st and ground floor. 8

9 UK systems Cambridge PowerEdge 576 Dell PowerEdge 1950 compute servers Computing capability: 28TFlop/s Each server has two Dual- Core Intel Xeon 5160 processors 3GHz and 8GB of memory InfiniBand network Bandwidth: 10GBit Latency: 7us 60 TeraByte of disk storage 9

10 Cluster Networks Introduction Communication has significant impact on application performance. Interconnection networks therefore have a vital role in cluster systems. As usual, the driver is performance An increase in compute power typically demands proportional increases in lower latency / higher bandwidth communication services. 10

11 Cluster Networks Issues with cluster interconnections are similar to those with normal networks: Latency & Bandwidth Latency= sender overhead + switching overhead + (message size / Bandwidth) + receiver overhead. Topology type (bus, ring, torus, hypercube etc). Routing, switching. Direct connections (point-to-point) or indirect connections. Balance performance and cost NIC (Network Interface Card) capabilities. Physical media (wiring density, reliability). 11

12 Interconnection Topologies In standard LANs we have two general structures: Shared network (bus) As used by classic Ethernet networks. All messages are broadcast each processor listens to every message. Requires complex access control (e.g. CSMA/CD). Collisions can occur: requires back-off policies and retransmissions. Suitable when the offered load is low - inappropriate for high performance applications. Very little reason to use this form of network today. Switched network Permits point-to-point communications between sender & receiver. Fast internal transport provides high aggregate bandwidth. Multiple messages are sent simultaneously. 12

13 Interconnection Topologies For switched networks, first consider node connectivity: Each node usually has one link (connection) to the switch. So if the node is a 2-way SMP, both processors compete for capacity. Can be improved by allowing multiple links per SMP. Useful quantities for switched networks: Scalability : the network s switch scalability with nodes. Degree: number of links to / from a node. Diameter: the shortest path between the furthest nodes. Bisection width: the minimum number of links that must be cut in order to divide the topology into two independent networks of the same size (+/- one node). Essentially a measure of bottleneck bandwidth - if higher, the network will perform better under load. 13

14 Interconnection Topologies Crossbar switch: Low latency and high throughput. Switch scalability is poor - O(N 2 ) Lots of wiring 14

15 Interconnection Topologies Linear Arrays and Rings Consider networks with switch scaling costs better than O(N 2 ). In one dimension, we have simple linear arrays. Direct topology (the number of switches : the number of nodes = 1:1) O(N) switches. These can wrap around to make a ring or 1D torus. High overall bandwidth but latency is high. So 2D/3D Cartesian applications will perform poorly with this network. 15

16 Interconnection Topologies 2D Meshes Can wrap-around as a 2D torus. Switch scaling: O(N) Average degree: 4 (as node count increases) Diameter: O(2n 1/2 ) Bisection width: O(n 1/2 ) 16

17 Interconnection Topologies Hypercubes: Or binary n-cube K dimension, Switches N= 2 K. Diameter: O(log 2 N). Good bisectional width (O(N/2)). 17

18 Interconnection Topologies Binary Tree: Indirect topology (there is not a 1:1 ratio between switch and node counts). Scaling: n = 2 d processor nodes (where d = depth) 2 n -1 switches Degree: 3 Diameter: O(2 log n) Bisection width: O(1) 18

19 Interconnection Topologies Fat trees: Similar in diameter to a binary tree. Bisection width (which equates to bottleneck) is greatly improved due to additional dimensions. For a fat quad-tree (such as used by Quadrics QSNet) bisectional bandwidth scales linearly with size. 19

20 Interconnection Topologies Summary of topologies: Topology Degree Diameter Bisection 1D Array 2 N-1 1 1D Ring 2 N/2 2 2D Mesh 4 2N 1/2 N 1/2 2D Torus 4 N 1/2 2N 1/2 Hypercube n=log(n) n N/2 There are others - we saw a 3D torus in the BlueGene/L section for instance. 20

21 Switching Operational modes: Store-and-forward: Each switch receives an entire packet before it forwards it onto the next switch - useful in a non-dedicated environment (I.e. a LAN). usually, there is a finite buffer size so it is possible that packets will be dropped under heavy load. Also impose a larger in-switch latency. Can detect errors in the packets Worm hole routing (Also called cut-through switching): Packet is divided into small flits (flow units). Switch examines the first flit (header) which contains the destination address, sets up a circuit and forwards the flit immediately. Subsequent flits of the message are forwarded as they arrive (near wirespeed). Reduces latency and buffer overhead. Messaging occurs at a speed close to the processors being directly connected. Less error detection 21

22 Cluster Network Technologies The performance of (affordable) interconnects in the past 5 years has dramatically increased. Current choices include: Gigabit Ethernet. Myricom s Myrinet. SCI - Dolphin. Quadrics QSNet. InfiniBand. 22

23 Cluster Network Technologies Gigabit Ethernet: The technology has matured and now offers very good performance at a very low cost. Small 1000BaseT switches are cheap and available at below $100/port. Larger switches are still quite expensive (from $1000). Bandwidth achievable is about 90% of peak (compared to about 35% for shared-bus Ethernet). Latency performance is moderate - many Ethernet switches are designed for general LANs (store & forward) where latency reduction is not necessary the primary incentive (the latency is order of ms). Zero-copy OS-bypass message passing can be supported with programmable NIC and direct memory access. 23

24 Cluster Network Technologies Myrinet: using fibre optic cable Uses a fat-tree structure that can accommodate large numbers of nodes. Low latency (7-10 µsec) with a peak bandwidth of 1800+Mbps. Provides zero-copy message passing and can offload packet processing to the NIC. Uses cut-through/worm-hole switching to reduce latency. More expensive than Ethernet but current market leader in HPC. (a) Twisted pair cable in Ethernet (b) Fibre optic cable 24

25 Cluster Network Technologies SCI: Scalable Cluster Interconect offers 500MB/s per port Nodes can be arranged in 2D (or 3D) wrapped mesh without switches. Messages not intended for the recipient pass straight through intelligent NICs. Low latency (~4 µsec) and high peak bandwidth (1800 Mbps). Wrapped mesh topology is well suited to applications that decompose data onto a Cartesian grid. 25

26 Cluster Network Technologies Quadrics: product of a strategic partnership between Quadrics & Compaq (used in ASCI/Q). Very low latency of 2-5 µsec due to fast interconnects and highly tuned software stack (MPI libraries); bandwidth is about 2Gbps Uses a quad-tree arrangement. Overall scaling is known to be good as the number of nodes is increased. 26

27 Cluster Network Technologies InfiniBand: by Intel. Initially developed for cluster networks, InfiniBand may eventually be used internally (for inter-processor communication) and I/O subsystems. Basic link speed of 2.5Gb/s. Cut-through/worm-hole switches are used. Current installations are achieving latencies of less than 7 µsec, but this is expected to improve. 27

Communication has significant impact on application performance. Interconnection networks therefore have a vital role in cluster systems.

Communication has significant impact on application performance. Interconnection networks therefore have a vital role in cluster systems. Cluster Networks Introduction Communication has significant impact on application performance. Interconnection networks therefore have a vital role in cluster systems. As usual, the driver is performance

More information

Cluster Network Products

Cluster Network Products Cluster Network Products Cluster interconnects include, among others: Gigabit Ethernet Myrinet Quadrics InfiniBand 1 Interconnects in Top500 list 11/2009 2 Interconnects in Top500 list 11/2008 3 Cluster

More information

Reduces latency and buffer overhead. Messaging occurs at a speed close to the processors being directly connected. Less error detection

Reduces latency and buffer overhead. Messaging occurs at a speed close to the processors being directly connected. Less error detection Switching Operational modes: Store-and-forward: Each switch receives an entire packet before it forwards it onto the next switch - useful in a general purpose network (I.e. a LAN). usually, there is a

More information

Parallel Computer Architecture II

Parallel Computer Architecture II Parallel Computer Architecture II Stefan Lang Interdisciplinary Center for Scientific Computing (IWR) University of Heidelberg INF 368, Room 532 D-692 Heidelberg phone: 622/54-8264 email: Stefan.Lang@iwr.uni-heidelberg.de

More information

Outline. Execution Environments for Parallel Applications. Supercomputers. Supercomputers

Outline. Execution Environments for Parallel Applications. Supercomputers. Supercomputers Outline Execution Environments for Parallel Applications Master CANS 2007/2008 Departament d Arquitectura de Computadors Universitat Politècnica de Catalunya Supercomputers OS abstractions Extended OS

More information

EE 4683/5683: COMPUTER ARCHITECTURE

EE 4683/5683: COMPUTER ARCHITECTURE 3/3/205 EE 4683/5683: COMPUTER ARCHITECTURE Lecture 8: Interconnection Networks Avinash Kodi, kodi@ohio.edu Agenda 2 Interconnection Networks Performance Metrics Topology 3/3/205 IN Performance Metrics

More information

1/5/2012. Overview of Interconnects. Presentation Outline. Myrinet and Quadrics. Interconnects. Switch-Based Interconnects

1/5/2012. Overview of Interconnects. Presentation Outline. Myrinet and Quadrics. Interconnects. Switch-Based Interconnects Overview of Interconnects Myrinet and Quadrics Leading Modern Interconnects Presentation Outline General Concepts of Interconnects Myrinet Latest Products Quadrics Latest Release Our Research Interconnects

More information

TDT Appendix E Interconnection Networks

TDT Appendix E Interconnection Networks TDT 4260 Appendix E Interconnection Networks Review Advantages of a snooping coherency protocol? Disadvantages of a snooping coherency protocol? Advantages of a directory coherency protocol? Disadvantages

More information

Physical Organization of Parallel Platforms. Alexandre David

Physical Organization of Parallel Platforms. Alexandre David Physical Organization of Parallel Platforms Alexandre David 1.2.05 1 Static vs. Dynamic Networks 13-02-2008 Alexandre David, MVP'08 2 Interconnection networks built using links and switches. How to connect:

More information

Lecture 2 Parallel Programming Platforms

Lecture 2 Parallel Programming Platforms Lecture 2 Parallel Programming Platforms Flynn s Taxonomy In 1966, Michael Flynn classified systems according to numbers of instruction streams and the number of data stream. Data stream Single Multiple

More information

Scalability and Classifications

Scalability and Classifications Scalability and Classifications 1 Types of Parallel Computers MIMD and SIMD classifications shared and distributed memory multicomputers distributed shared memory computers 2 Network Topologies static

More information

Real Parallel Computers

Real Parallel Computers Real Parallel Computers Modular data centers Background Information Recent trends in the marketplace of high performance computing Strohmaier, Dongarra, Meuer, Simon Parallel Computing 2005 Short history

More information

COSC 6374 Parallel Computation. Parallel Computer Architectures

COSC 6374 Parallel Computation. Parallel Computer Architectures OS 6374 Parallel omputation Parallel omputer Architectures Some slides on network topologies based on a similar presentation by Michael Resch, University of Stuttgart Spring 2010 Flynn s Taxonomy SISD:

More information

Interconnection Network. Jinkyu Jeong Computer Systems Laboratory Sungkyunkwan University

Interconnection Network. Jinkyu Jeong Computer Systems Laboratory Sungkyunkwan University Interconnection Network Jinkyu Jeong (jinkyu@skku.edu) Computer Systems Laboratory Sungkyunkwan University http://csl.skku.edu Topics Taxonomy Metric Topologies Characteristics Cost Performance 2 Interconnection

More information

High Performance Computing Programming Paradigms and Scalability Part 2: High-Performance Networks

High Performance Computing Programming Paradigms and Scalability Part 2: High-Performance Networks High Performance Computing Programming Paradigms and Scalability Part 2: High-Performance Networks PD Dr. rer. nat. habil. Ralf-Peter Mundani Computation in Engineering (CiE) Scientific Computing (SCCS)

More information

COSC 6374 Parallel Computation. Parallel Computer Architectures

COSC 6374 Parallel Computation. Parallel Computer Architectures OS 6374 Parallel omputation Parallel omputer Architectures Some slides on network topologies based on a similar presentation by Michael Resch, University of Stuttgart Edgar Gabriel Fall 2015 Flynn s Taxonomy

More information

Lecture 12: Interconnection Networks. Topics: communication latency, centralized and decentralized switches, routing, deadlocks (Appendix E)

Lecture 12: Interconnection Networks. Topics: communication latency, centralized and decentralized switches, routing, deadlocks (Appendix E) Lecture 12: Interconnection Networks Topics: communication latency, centralized and decentralized switches, routing, deadlocks (Appendix E) 1 Topologies Internet topologies are not very regular they grew

More information

Interconnection Network

Interconnection Network Interconnection Network Jinkyu Jeong (jinkyu@skku.edu) Computer Systems Laboratory Sungkyunkwan University http://csl.skku.edu SSE3054: Multicore Systems, Spring 2017, Jinkyu Jeong (jinkyu@skku.edu) Topics

More information

4. Networks. in parallel computers. Advances in Computer Architecture

4. Networks. in parallel computers. Advances in Computer Architecture 4. Networks in parallel computers Advances in Computer Architecture System architectures for parallel computers Control organization Single Instruction stream Multiple Data stream (SIMD) All processors

More information

EN2910A: Advanced Computer Architecture Topic 06: Supercomputers & Data Centers Prof. Sherief Reda School of Engineering Brown University

EN2910A: Advanced Computer Architecture Topic 06: Supercomputers & Data Centers Prof. Sherief Reda School of Engineering Brown University EN2910A: Advanced Computer Architecture Topic 06: Supercomputers & Data Centers Prof. Sherief Reda School of Engineering Brown University Material from: The Datacenter as a Computer: An Introduction to

More information

Lecture: Interconnection Networks

Lecture: Interconnection Networks Lecture: Interconnection Networks Topics: Router microarchitecture, topologies Final exam next Tuesday: same rules as the first midterm 1 Packets/Flits A message is broken into multiple packets (each packet

More information

Interconnect Technology and Computational Speed

Interconnect Technology and Computational Speed Interconnect Technology and Computational Speed From Chapter 1 of B. Wilkinson et al., PARAL- LEL PROGRAMMING. Techniques and Applications Using Networked Workstations and Parallel Computers, augmented

More information

What is Parallel Computing?

What is Parallel Computing? What is Parallel Computing? Parallel Computing is several processing elements working simultaneously to solve a problem faster. 1/33 What is Parallel Computing? Parallel Computing is several processing

More information

Interconnection Networks

Interconnection Networks Lecture 17: Interconnection Networks Parallel Computer Architecture and Programming A comment on web site comments It is okay to make a comment on a slide/topic that has already been commented on. In fact

More information

CS Parallel Algorithms in Scientific Computing

CS Parallel Algorithms in Scientific Computing CS 775 - arallel Algorithms in Scientific Computing arallel Architectures January 2, 2004 Lecture 2 References arallel Computer Architecture: A Hardware / Software Approach Culler, Singh, Gupta, Morgan

More information

Interconnection Network

Interconnection Network Interconnection Network Recap: Generic Parallel Architecture A generic modern multiprocessor Network Mem Communication assist (CA) $ P Node: processor(s), memory system, plus communication assist Network

More information

The Optimal CPU and Interconnect for an HPC Cluster

The Optimal CPU and Interconnect for an HPC Cluster 5. LS-DYNA Anwenderforum, Ulm 2006 Cluster / High Performance Computing I The Optimal CPU and Interconnect for an HPC Cluster Andreas Koch Transtec AG, Tübingen, Deutschland F - I - 15 Cluster / High Performance

More information

High Performance Computing Programming Paradigms and Scalability

High Performance Computing Programming Paradigms and Scalability High Performance Computing Programming Paradigms and Scalability PD Dr. rer. nat. habil. Ralf Peter Mundani Computation in Engineering / BGU Scientific Computing in Computer Science / INF Summer Term 208

More information

INTERCONNECTION NETWORKS LECTURE 4

INTERCONNECTION NETWORKS LECTURE 4 INTERCONNECTION NETWORKS LECTURE 4 DR. SAMMAN H. AMEEN 1 Topology Specifies way switches are wired Affects routing, reliability, throughput, latency, building ease Routing How does a message get from source

More information

Cray XC Scalability and the Aries Network Tony Ford

Cray XC Scalability and the Aries Network Tony Ford Cray XC Scalability and the Aries Network Tony Ford June 29, 2017 Exascale Scalability Which scalability metrics are important for Exascale? Performance (obviously!) What are the contributing factors?

More information

Lecture 26: Interconnects. James C. Hoe Department of ECE Carnegie Mellon University

Lecture 26: Interconnects. James C. Hoe Department of ECE Carnegie Mellon University 18 447 Lecture 26: Interconnects James C. Hoe Department of ECE Carnegie Mellon University 18 447 S18 L26 S1, James C. Hoe, CMU/ECE/CALCM, 2018 Housekeeping Your goal today get an overview of parallel

More information

MIMD Overview. Intel Paragon XP/S Overview. XP/S Usage. XP/S Nodes and Interconnection. ! Distributed-memory MIMD multicomputer

MIMD Overview. Intel Paragon XP/S Overview. XP/S Usage. XP/S Nodes and Interconnection. ! Distributed-memory MIMD multicomputer MIMD Overview Intel Paragon XP/S Overview! MIMDs in the 1980s and 1990s! Distributed-memory multicomputers! Intel Paragon XP/S! Thinking Machines CM-5! IBM SP2! Distributed-memory multicomputers with hardware

More information

Hubs. twisted pair. hub. 5: DataLink Layer 5-1

Hubs. twisted pair. hub. 5: DataLink Layer 5-1 Hubs Hubs are essentially physical-layer repeaters: bits coming from one link go out all other links at the same rate no frame buffering no CSMA/CD at : adapters detect collisions provides net management

More information

Non-Uniform Memory Access (NUMA) Architecture and Multicomputers

Non-Uniform Memory Access (NUMA) Architecture and Multicomputers Non-Uniform Memory Access (NUMA) Architecture and Multicomputers Parallel and Distributed Computing Department of Computer Science and Engineering (DEI) Instituto Superior Técnico February 29, 2016 CPD

More information

Networks. Distributed Systems. Philipp Kupferschmied. Universität Karlsruhe, System Architecture Group. May 6th, 2009

Networks. Distributed Systems. Philipp Kupferschmied. Universität Karlsruhe, System Architecture Group. May 6th, 2009 Networks Distributed Systems Philipp Kupferschmied Universität Karlsruhe, System Architecture Group May 6th, 2009 Philipp Kupferschmied Networks 1/ 41 1 Communication Basics Introduction Layered Communication

More information

Non-Uniform Memory Access (NUMA) Architecture and Multicomputers

Non-Uniform Memory Access (NUMA) Architecture and Multicomputers Non-Uniform Memory Access (NUMA) Architecture and Multicomputers Parallel and Distributed Computing Department of Computer Science and Engineering (DEI) Instituto Superior Técnico September 26, 2011 CPD

More information

Cray XD1 Supercomputer Release 1.3 CRAY XD1 DATASHEET

Cray XD1 Supercomputer Release 1.3 CRAY XD1 DATASHEET CRAY XD1 DATASHEET Cray XD1 Supercomputer Release 1.3 Purpose-built for HPC delivers exceptional application performance Affordable power designed for a broad range of HPC workloads and budgets Linux,

More information

Network-on-chip (NOC) Topologies

Network-on-chip (NOC) Topologies Network-on-chip (NOC) Topologies 1 Network Topology Static arrangement of channels and nodes in an interconnection network The roads over which packets travel Topology chosen based on cost and performance

More information

LS-DYNA Best-Practices: Networking, MPI and Parallel File System Effect on LS-DYNA Performance

LS-DYNA Best-Practices: Networking, MPI and Parallel File System Effect on LS-DYNA Performance 11 th International LS-DYNA Users Conference Computing Technology LS-DYNA Best-Practices: Networking, MPI and Parallel File System Effect on LS-DYNA Performance Gilad Shainer 1, Tong Liu 2, Jeff Layton

More information

Real Parallel Computers

Real Parallel Computers Real Parallel Computers Modular data centers Overview Short history of parallel machines Cluster computing Blue Gene supercomputer Performance development, top-500 DAS: Distributed supercomputing Short

More information

represent parallel computers, so distributed systems such as Does not consider storage or I/O issues

represent parallel computers, so distributed systems such as Does not consider storage or I/O issues Top500 Supercomputer list represent parallel computers, so distributed systems such as SETI@Home are not considered Does not consider storage or I/O issues Both custom designed machines and commodity machines

More information

Non-Uniform Memory Access (NUMA) Architecture and Multicomputers

Non-Uniform Memory Access (NUMA) Architecture and Multicomputers Non-Uniform Memory Access (NUMA) Architecture and Multicomputers Parallel and Distributed Computing MSc in Information Systems and Computer Engineering DEA in Computational Engineering Department of Computer

More information

High Performance Computing: Concepts, Methods & Means Enabling Technologies 2 : Cluster Networks

High Performance Computing: Concepts, Methods & Means Enabling Technologies 2 : Cluster Networks High Performance Computing: Concepts, Methods & Means Enabling Technologies 2 : Cluster Networks Prof. Amy Apon Department of Computer Science and Computer Engineering University of Arkansas March 15 th,

More information

CS575 Parallel Processing

CS575 Parallel Processing CS575 Parallel Processing Lecture three: Interconnection Networks Wim Bohm, CSU Except as otherwise noted, the content of this presentation is licensed under the Creative Commons Attribution 2.5 license.

More information

Overview. Processor organizations Types of parallel machines. Real machines

Overview. Processor organizations Types of parallel machines. Real machines Course Outline Introduction in algorithms and applications Parallel machines and architectures Overview of parallel machines, trends in top-500, clusters, DAS Programming methods, languages, and environments

More information

ECE/CS 757: Advanced Computer Architecture II Interconnects

ECE/CS 757: Advanced Computer Architecture II Interconnects ECE/CS 757: Advanced Computer Architecture II Interconnects Instructor:Mikko H Lipasti Spring 2017 University of Wisconsin-Madison Lecture notes created by Natalie Enright Jerger Lecture Outline Introduction

More information

The Tofu Interconnect D

The Tofu Interconnect D The Tofu Interconnect D 11 September 2018 Yuichiro Ajima, Takahiro Kawashima, Takayuki Okamoto, Naoyuki Shida, Kouichi Hirai, Toshiyuki Shimizu, Shinya Hiramoto, Yoshiro Ikeda, Takahide Yoshikawa, Kenji

More information

Scalable Ethernet Clos-Switches. Norbert Eicker John von Neumann-Institute for Computing Ferdinand Geier ParTec Cluster Competence Center GmbH

Scalable Ethernet Clos-Switches. Norbert Eicker John von Neumann-Institute for Computing Ferdinand Geier ParTec Cluster Competence Center GmbH Scalable Ethernet Clos-Switches Norbert Eicker John von Neumann-Institute for Computing Ferdinand Geier ParTec Cluster Competence Center GmbH Outline Motivation Clos-Switches Ethernet Crossbar Switches

More information

Parallel Architectures

Parallel Architectures Parallel Architectures Instructor: Tsung-Che Chiang tcchiang@ieee.org Department of Science and Information Engineering National Taiwan Normal University Introduction In the roughly three decades between

More information

CCS HPC. Interconnection Network. PC MPP (Massively Parallel Processor) MPP IBM

CCS HPC. Interconnection Network. PC MPP (Massively Parallel Processor) MPP IBM CCS HC taisuke@cs.tsukuba.ac.jp 1 2 CU memoryi/o 2 2 4single chipmulti-core CU 10 C CM (Massively arallel rocessor) M IBM BlueGene/L 65536 Interconnection Network 3 4 (distributed memory system) (shared

More information

The Impact of Optics on HPC System Interconnects

The Impact of Optics on HPC System Interconnects The Impact of Optics on HPC System Interconnects Mike Parker and Steve Scott Hot Interconnects 2009 Manhattan, NYC Will cost-effective optics fundamentally change the landscape of networking? Yes. Changes

More information

High Performance Computing - Parallel Computers and Networks. Prof Matt Probert

High Performance Computing - Parallel Computers and Networks. Prof Matt Probert High Performance Computing - Parallel Computers and Networks Prof Matt Probert http://www-users.york.ac.uk/~mijp1 Overview Parallel on a chip? Shared vs. distributed memory Latency & bandwidth Topology

More information

10-Gigabit iwarp Ethernet: Comparative Performance Analysis with InfiniBand and Myrinet-10G

10-Gigabit iwarp Ethernet: Comparative Performance Analysis with InfiniBand and Myrinet-10G 10-Gigabit iwarp Ethernet: Comparative Performance Analysis with InfiniBand and Myrinet-10G Mohammad J. Rashti and Ahmad Afsahi Queen s University Kingston, ON, Canada 2007 Workshop on Communication Architectures

More information

Interconnection Networks: Topology. Prof. Natalie Enright Jerger

Interconnection Networks: Topology. Prof. Natalie Enright Jerger Interconnection Networks: Topology Prof. Natalie Enright Jerger Topology Overview Definition: determines arrangement of channels and nodes in network Analogous to road map Often first step in network design

More information

Lecture 20: Distributed Memory Parallelism. William Gropp

Lecture 20: Distributed Memory Parallelism. William Gropp Lecture 20: Distributed Parallelism William Gropp www.cs.illinois.edu/~wgropp A Very Short, Very Introductory Introduction We start with a short introduction to parallel computing from scratch in order

More information

The NE010 iwarp Adapter

The NE010 iwarp Adapter The NE010 iwarp Adapter Gary Montry Senior Scientist +1-512-493-3241 GMontry@NetEffect.com Today s Data Center Users Applications networking adapter LAN Ethernet NAS block storage clustering adapter adapter

More information

Parallel Architectures

Parallel Architectures Parallel Architectures Part 1: The rise of parallel machines Intel Core i7 4 CPU cores 2 hardware thread per core (8 cores ) Lab Cluster Intel Xeon 4/10/16/18 CPU cores 2 hardware thread per core (8/20/32/36

More information

SMP and ccnuma Multiprocessor Systems. Sharing of Resources in Parallel and Distributed Computing Systems

SMP and ccnuma Multiprocessor Systems. Sharing of Resources in Parallel and Distributed Computing Systems Reference Papers on SMP/NUMA Systems: EE 657, Lecture 5 September 14, 2007 SMP and ccnuma Multiprocessor Systems Professor Kai Hwang USC Internet and Grid Computing Laboratory Email: kaihwang@usc.edu [1]

More information

InfiniBand SDR, DDR, and QDR Technology Guide

InfiniBand SDR, DDR, and QDR Technology Guide White Paper InfiniBand SDR, DDR, and QDR Technology Guide The InfiniBand standard supports single, double, and quadruple data rate that enables an InfiniBand link to transmit more data. This paper discusses

More information

Interconnection Networks

Interconnection Networks Lecture 18: Interconnection Networks Parallel Computer Architecture and Programming CMU 15-418/15-618, Spring 2015 Credit: many of these slides were created by Michael Papamichael This lecture is partially

More information

Use of the Internet SCSI (iscsi) protocol

Use of the Internet SCSI (iscsi) protocol A unified networking approach to iscsi storage with Broadcom controllers By Dhiraj Sehgal, Abhijit Aswath, and Srinivas Thodati In environments based on Internet SCSI (iscsi) and 10 Gigabit Ethernet, deploying

More information

Lecture 24: Interconnection Networks. Topics: topologies, routing, deadlocks, flow control

Lecture 24: Interconnection Networks. Topics: topologies, routing, deadlocks, flow control Lecture 24: Interconnection Networks Topics: topologies, routing, deadlocks, flow control 1 Topology Examples Grid Torus Hypercube Criteria Bus Ring 2Dtorus 6-cube Fully connected Performance Bisection

More information

CS 770G - Parallel Algorithms in Scientific Computing Parallel Architectures. May 7, 2001 Lecture 2

CS 770G - Parallel Algorithms in Scientific Computing Parallel Architectures. May 7, 2001 Lecture 2 CS 770G - arallel Algorithms in Scientific Computing arallel Architectures May 7, 2001 Lecture 2 References arallel Computer Architecture: A Hardware / Software Approach Culler, Singh, Gupta, Morgan Kaufmann

More information

Lecture 18: Communication Models and Architectures: Interconnection Networks

Lecture 18: Communication Models and Architectures: Interconnection Networks Design & Co-design of Embedded Systems Lecture 18: Communication Models and Architectures: Interconnection Networks Sharif University of Technology Computer Engineering g Dept. Winter-Spring 2008 Mehdi

More information

Cluster Computing. Interconnect Technologies for Clusters

Cluster Computing. Interconnect Technologies for Clusters Interconnect Technologies for Clusters Interconnect approaches WAN infinite distance LAN Few kilometers SAN Few meters Backplane Not scalable Physical Cluster Interconnects FastEther Gigabit EtherNet 10

More information

Interconnection Networks. Issues for Networks

Interconnection Networks. Issues for Networks Interconnection Networks Communications Among Processors Chris Nevison, Colgate University Issues for Networks Total Bandwidth amount of data which can be moved from somewhere to somewhere per unit time

More information

A LAN is a high-speed data network that covers a relatively small geographic area. It typically connects workstations, personal computers, printers,

A LAN is a high-speed data network that covers a relatively small geographic area. It typically connects workstations, personal computers, printers, CBCN4103 A LAN is a high-speed data network that covers a relatively small geographic area. It typically connects workstations, personal computers, printers, servers, and other devices. LANs offer computer

More information

CSC630/CSC730: Parallel Computing

CSC630/CSC730: Parallel Computing CSC630/CSC730: Parallel Computing Parallel Computing Platforms Chapter 2 (2.4.1 2.4.4) Dr. Joe Zhang PDC-4: Topology 1 Content Parallel computing platforms Logical organization (a programmer s view) Control

More information

Adaptive Routing Strategies for Modern High Performance Networks

Adaptive Routing Strategies for Modern High Performance Networks Adaptive Routing Strategies for Modern High Performance Networks Patrick Geoffray Myricom patrick@myri.com Torsten Hoefler Indiana University htor@cs.indiana.edu 28 August 2008 Hot Interconnect Stanford,

More information

Parallel Architectures

Parallel Architectures Parallel Architectures CPS343 Parallel and High Performance Computing Spring 2018 CPS343 (Parallel and HPC) Parallel Architectures Spring 2018 1 / 36 Outline 1 Parallel Computer Classification Flynn s

More information

The way toward peta-flops

The way toward peta-flops The way toward peta-flops ISC-2011 Dr. Pierre Lagier Chief Technology Officer Fujitsu Systems Europe Where things started from DESIGN CONCEPTS 2 New challenges and requirements! Optimal sustained flops

More information

More on LANS. LAN Wiring, Interface

More on LANS. LAN Wiring, Interface More on LANS Chapters 10-11 LAN Wiring, Interface Mostly covered this material already NIC = Network Interface Card Separate processor, buffers incoming/outgoing data CPU might not be able to keep up network

More information

Performance Optimizations via Connect-IB and Dynamically Connected Transport Service for Maximum Performance on LS-DYNA

Performance Optimizations via Connect-IB and Dynamically Connected Transport Service for Maximum Performance on LS-DYNA Performance Optimizations via Connect-IB and Dynamically Connected Transport Service for Maximum Performance on LS-DYNA Pak Lui, Gilad Shainer, Brian Klaff Mellanox Technologies Abstract From concept to

More information

Lecture 13: Interconnection Networks. Topics: lots of background, recent innovations for power and performance

Lecture 13: Interconnection Networks. Topics: lots of background, recent innovations for power and performance Lecture 13: Interconnection Networks Topics: lots of background, recent innovations for power and performance 1 Interconnection Networks Recall: fully connected network, arrays/rings, meshes/tori, trees,

More information

Comparing Linux Clusters for the Community Climate System Model

Comparing Linux Clusters for the Community Climate System Model Comparing Linux Clusters for the Community Climate System Model Matthew Woitaszek, Michael Oberg, and Henry M. Tufo Department of Computer Science University of Colorado, Boulder {matthew.woitaszek, michael.oberg}@colorado.edu,

More information

Lecture 28: Networks & Interconnect Architectural Issues Professor Randy H. Katz Computer Science 252 Spring 1996

Lecture 28: Networks & Interconnect Architectural Issues Professor Randy H. Katz Computer Science 252 Spring 1996 Lecture 28: Networks & Interconnect Architectural Issues Professor Randy H. Katz Computer Science 252 Spring 1996 RHK.S96 1 Review: ABCs of Networks Starting Point: Send bits between 2 computers Queue

More information

High bandwidth, Long distance. Where is my throughput? Robin Tasker CCLRC, Daresbury Laboratory, UK

High bandwidth, Long distance. Where is my throughput? Robin Tasker CCLRC, Daresbury Laboratory, UK High bandwidth, Long distance. Where is my throughput? Robin Tasker CCLRC, Daresbury Laboratory, UK [r.tasker@dl.ac.uk] DataTAG is a project sponsored by the European Commission - EU Grant IST-2001-32459

More information

What is the Future for High-Performance Networking?

What is the Future for High-Performance Networking? What is the Future for High-Performance Networking? Wu-chun (Wu) Feng feng@lanl.gov RADIANT: Research And Development in Advanced Network Technology http://www.lanl.gov/radiant Computer & Computational

More information

ABySS Performance Benchmark and Profiling. May 2010

ABySS Performance Benchmark and Profiling. May 2010 ABySS Performance Benchmark and Profiling May 2010 Note The following research was performed under the HPC Advisory Council activities Participating vendors: AMD, Dell, Mellanox Compute resource - HPC

More information

BlueGene/L (No. 4 in the Latest Top500 List)

BlueGene/L (No. 4 in the Latest Top500 List) BlueGene/L (No. 4 in the Latest Top500 List) first supercomputer in the Blue Gene project architecture. Individual PowerPC 440 processors at 700Mhz Two processors reside in a single chip. Two chips reside

More information

Message Passing Models and Multicomputer distributed system LECTURE 7

Message Passing Models and Multicomputer distributed system LECTURE 7 Message Passing Models and Multicomputer distributed system LECTURE 7 DR SAMMAN H AMEEN 1 Node Node Node Node Node Node Message-passing direct network interconnection Node Node Node Node Node Node PAGE

More information

Intel Many Integrated Core (MIC) Matt Kelly & Ryan Rawlins

Intel Many Integrated Core (MIC) Matt Kelly & Ryan Rawlins Intel Many Integrated Core (MIC) Matt Kelly & Ryan Rawlins Outline History & Motivation Architecture Core architecture Network Topology Memory hierarchy Brief comparison to GPU & Tilera Programming Applications

More information

CS/COE1541: Intro. to Computer Architecture

CS/COE1541: Intro. to Computer Architecture CS/COE1541: Intro. to Computer Architecture Multiprocessors Sangyeun Cho Computer Science Department Tilera TILE64 IBM BlueGene/L nvidia GPGPU Intel Core 2 Duo 2 Why multiprocessors? For improved latency

More information

Infiniband Fast Interconnect

Infiniband Fast Interconnect Infiniband Fast Interconnect Yuan Liu Institute of Information and Mathematical Sciences Massey University May 2009 Abstract Infiniband is the new generation fast interconnect provides bandwidths both

More information

QLogic TrueScale InfiniBand and Teraflop Simulations

QLogic TrueScale InfiniBand and Teraflop Simulations WHITE Paper QLogic TrueScale InfiniBand and Teraflop Simulations For ANSYS Mechanical v12 High Performance Interconnect for ANSYS Computer Aided Engineering Solutions Executive Summary Today s challenging

More information

Network Design Considerations for Grid Computing

Network Design Considerations for Grid Computing Network Design Considerations for Grid Computing Engineering Systems How Bandwidth, Latency, and Packet Size Impact Grid Job Performance by Erik Burrows, Engineering Systems Analyst, Principal, Broadcom

More information

Stockholm Brain Institute Blue Gene/L

Stockholm Brain Institute Blue Gene/L Stockholm Brain Institute Blue Gene/L 1 Stockholm Brain Institute Blue Gene/L 2 IBM Systems & Technology Group and IBM Research IBM Blue Gene /P - An Overview of a Petaflop Capable System Carl G. Tengwall

More information

Scaling to Petaflop. Ola Torudbakken Distinguished Engineer. Sun Microsystems, Inc

Scaling to Petaflop. Ola Torudbakken Distinguished Engineer. Sun Microsystems, Inc Scaling to Petaflop Ola Torudbakken Distinguished Engineer Sun Microsystems, Inc HPC Market growth is strong CAGR increased from 9.2% (2006) to 15.5% (2007) Market in 2007 doubled from 2003 (Source: IDC

More information

Resource allocation and utilization in the Blue Gene/L supercomputer

Resource allocation and utilization in the Blue Gene/L supercomputer Resource allocation and utilization in the Blue Gene/L supercomputer Tamar Domany, Y Aridor, O Goldshmidt, Y Kliteynik, EShmueli, U Silbershtein IBM Labs in Haifa Agenda Blue Gene/L Background Blue Gene/L

More information

Overview of Parallel Computing. Timothy H. Kaiser, PH.D.

Overview of Parallel Computing. Timothy H. Kaiser, PH.D. Overview of Parallel Computing Timothy H. Kaiser, PH.D. tkaiser@mines.edu Introduction What is parallel computing? Why go parallel? The best example of parallel computing Some Terminology Slides and examples

More information

vstart 50 VMware vsphere Solution Specification

vstart 50 VMware vsphere Solution Specification vstart 50 VMware vsphere Solution Specification Release 1.3 for 12 th Generation Servers Dell Virtualization Solutions Engineering Revision: A00 March 2012 THIS WHITE PAPER IS FOR INFORMATIONAL PURPOSES

More information

Parallel Architecture. Sathish Vadhiyar

Parallel Architecture. Sathish Vadhiyar Parallel Architecture Sathish Vadhiyar Motivations of Parallel Computing Faster execution times From days or months to hours or seconds E.g., climate modelling, bioinformatics Large amount of data dictate

More information

Local Area Network Overview

Local Area Network Overview Local Area Network Overview Chapter 15 CS420/520 Axel Krings Page 1 LAN Applications (1) Personal computer LANs Low cost Limited data rate Back end networks Interconnecting large systems (mainframes and

More information

EE/CSCI 451: Parallel and Distributed Computation

EE/CSCI 451: Parallel and Distributed Computation EE/CSCI 451: Parallel and Distributed Computation Lecture #11 2/21/2017 Xuehai Qian Xuehai.qian@usc.edu http://alchem.usc.edu/portal/xuehaiq.html University of Southern California 1 Outline Midterm 1:

More information

Internetworking is connecting two or more computer networks with some sort of routing device to exchange traffic back and forth, and guide traffic on

Internetworking is connecting two or more computer networks with some sort of routing device to exchange traffic back and forth, and guide traffic on CBCN4103 Internetworking is connecting two or more computer networks with some sort of routing device to exchange traffic back and forth, and guide traffic on the correct path across the complete network

More information

Parallel Programming with MPI

Parallel Programming with MPI Parallel Programming with MPI Science and Technology Support Ohio Supercomputer Center 1224 Kinnear Road. Columbus, OH 43212 (614) 292-1800 oschelp@osc.edu http://www.osc.edu/supercomputing/ Functions

More information

Introduction to Multiprocessors (Part I) Prof. Cristina Silvano Politecnico di Milano

Introduction to Multiprocessors (Part I) Prof. Cristina Silvano Politecnico di Milano Introduction to Multiprocessors (Part I) Prof. Cristina Silvano Politecnico di Milano Outline Key issues to design multiprocessors Interconnection network Centralized shared-memory architectures Distributed

More information

Clusters of SMP s. Sean Peisert

Clusters of SMP s. Sean Peisert Clusters of SMP s Sean Peisert What s Being Discussed Today SMP s Cluters of SMP s Programming Models/Languages Relevance to Commodity Computing Relevance to Supercomputing SMP s Symmetric Multiprocessors

More information

Chapter 4 NETWORK HARDWARE

Chapter 4 NETWORK HARDWARE Chapter 4 NETWORK HARDWARE 1 Network Devices As Organizations grow, so do their networks Growth in number of users Geographical Growth Network Devices : Are products used to expand or connect networks.

More information

CS4961 Parallel Programming. Lecture 4: Memory Systems and Interconnects 9/1/11. Administrative. Mary Hall September 1, Homework 2, cont.

CS4961 Parallel Programming. Lecture 4: Memory Systems and Interconnects 9/1/11. Administrative. Mary Hall September 1, Homework 2, cont. CS4961 Parallel Programming Lecture 4: Memory Systems and Interconnects Administrative Nikhil office hours: - Monday, 2-3PM - Lab hours on Tuesday afternoons during programming assignments First homework

More information