"Practical Considerations in Building Beowulf Clusters" Lessons from Experience and Future Directions Arch Davis (GS*69) Davis Systems Engineering.

Slides:



Advertisements
Similar presentations
Elements of a Microprocessor system Central processing unit. This performs the arithmetic and logical operations, such as add/subtract, multiply/divide,
Advertisements

♦ Commodity processor with commodity inter- processor connection Clusters Pentium, Itanium, Opteron, Alpha GigE, Infiniband, Myrinet, Quadrics, SCI NEC.
Beowulf Supercomputer System Lee, Jung won CS843.
Teaching Parallel Computing using Beowulf Clusters: A Laboratory Approach Phil Prins Seattle Pacific University October 8, 2004
Chapter 5: Server Hardware and Availability. Hardware Reliability and LAN The more reliable a component, the more expensive it is. Server hardware is.
Types of Parallel Computers
Information Technology Center Introduction to High Performance Computing at KFUPM.
Linux Clustering A way to supercomputing. What is Cluster? A group of individual computers bundled together using hardware and software in order to make.
Presented by: Yash Gurung, ICFAI UNIVERSITY.Sikkim BUILDING of 3 R'sCLUSTER PARALLEL COMPUTER.
History of Distributed Systems Joseph Cordina
HELICS Petteri Johansson & Ilkka Uuhiniemi. HELICS COW –AMD Athlon MP 1.4Ghz –512 (2 in same computing node) –35 at top500.org –Linpack Benchmark 825.
Supercomputers Daniel Shin CS 147, Section 1 April 29, 2010.
A Comparative Study of Network Protocols & Interconnect for Cluster Computing Performance Evaluation of Fast Ethernet, Gigabit Ethernet and Myrinet.
Computational Astrophysics: Methodology 1.Identify astrophysical problem 2.Write down corresponding equations 3.Identify numerical algorithm 4.Find a computer.
Seminar on parallel computing Goal: provide environment for exploration of parallel computing Driven by participants Weekly hour for discussion, show &
Virtual Machines for HPC Paul Lu, Cam Macdonell Dept of Computing Science.
Real Parallel Computers. Background Information Recent trends in the marketplace of high performance computing Strohmaier, Dongarra, Meuer, Simon Parallel.
High Performance Computing (HPC) at Center for Information Communication and Technology in UTM.
Virtual Network Servers. What is a Server? 1. A software application that provides a specific one or more services to other computers  Example: Apache.
Chapter 4  Converts data into information  Control center  Set of electronic circuitry that executes stored program instructions  Two parts ◦ Control.
1b.1 Types of Parallel Computers Two principal approaches: Shared memory multiprocessor Distributed memory multicomputer ITCS 4/5145 Parallel Programming,
Prepared by Careene McCallum-Rodney Hardware specification of a computer system.
Real Parallel Computers. Modular data centers Background Information Recent trends in the marketplace of high performance computing Strohmaier, Dongarra,
Cluster Computers. Introduction Cluster computing –Standard PCs or workstations connected by a fast network –Good price/performance ratio –Exploit existing.
Digital Graphics and Computers. Hardware and Software Working with graphic images requires suitable hardware and software to produce the best results.
Chapter 01 Nell Dale & John Lewis.
XP Practical PC, 3e Chapter 17 1 Upgrading and Expanding your PC.
05/18/03Maurizio Davini Hepix2003 Department of Physics University of Pisa Site Report Maurizio Davini Department of Physics and INFN Pisa.
Parallel Computing The Bad News –Hardware is not getting faster fast enough –Too many architectures –Existing architectures are too specific –Programs.
Motivation “Every three minutes a woman is diagnosed with Breast cancer” (American Cancer Society, “Detailed Guide: Breast Cancer,” 2006) Explore the use.
GPU Programming with CUDA – Accelerated Architectures Mike Griffiths
Hardware. THE MOVES INSTITUTE Hardware So you want to build a cluster. What do you need to buy? Remember the definition of a beowulf cluster: Commodity.
Information and Communication Technology Fundamentals Credits Hours: 2+1 Instructor: Ayesha Bint Saleem.
Operational computing environment at EARS Jure Jerman Meteorological Office Environmental Agency of Slovenia (EARS)
Computer Hardware Mr. Richard Orr Technology Teacher Bednarcik Jr. High School.
CHAPTER 11: Modern Computer Systems
Translate the following message:
1b.1 Types of Parallel Computers Two principal approaches: Shared memory multiprocessor Distributed memory multicomputer ITCS 4/5145 Parallel Programming,
High-End Computing Systems EE380 State-of-the-Art Lecture Hank Dietz Professor & Hardymon Chair in Networking Electrical & Computer Engineering Dept. University.
Cluster Computers. Introduction Cluster computing –Standard PCs or workstations connected by a fast network –Good price/performance ratio –Exploit existing.
Ohio Supercomputer Center Cluster Computing Overview Summer Institute for Advanced Computing August 22, 2000 Doug Johnson, OSC.
Introduction, background, jargon Jakub Yaghob. Literature T.G.Mattson, B.A.Sanders, B.L.Massingill: Patterns for Parallel Programming, Addison- Wesley,
Cluster Workstations. Recently the distinction between parallel and distributed computers has become blurred with the advent of the network of workstations.
Hardware Trends. Contents Memory Hard Disks Processors Network Accessories Future.
Department of Computer Science University of the West Indies.
High Performance Computing Processors Felix Noble Mirayma V. Rodriguez Agnes Velez Electric and Computer Engineer Department August 25, 2004.
المحاضرة الاولى Operating Systems. The general objectives of this decision explain the concepts and the importance of operating systems and development.
Taking the Complexity out of Cluster Computing Vendor Update HPC User Forum Arend Dittmer Director Product Management HPC April,
Loosely Coupled Parallelism: Clusters. Context We have studied older archictures for loosely coupled parallelism, such as mesh’s, hypercubes etc, which.
The Guts. CPU CPU Socket The CPU is generally a 2 inch ceramic square with a silicon chip located inside. The chip usually about the size of a thumbnail.
Copyright © 2011 Curt Hill MIMD Multiple Instructions Multiple Data.
Computing Environment The computing environment rapidly evolving ‑ you need to know not only the methods, but also How and when to apply them, Which computers.
COMP381 by M. Hamdi 1 Clusters: Networks of WS/PC.
Building and managing production bioclusters Chris Dagdigian BIOSILICO Vol2, No. 5 September 2004 Ankur Dhanik.
3/12/2013Computer Engg, IIT(BHU)1 PARALLEL COMPUTERS- 2.
Copyright © 2006 by The McGraw-Hill Companies, Inc. All rights reserved. McGraw-Hill Technology Education Copyright © 2006 by The McGraw-Hill Companies,
3/12/2013Computer Engg, IIT(BHU)1 PARALLEL COMPUTERS- 1.
Computer Parts - Hardware There are many parts that work together to make a computer work.
Cluster Computers. Introduction Cluster computing –Standard PCs or workstations connected by a fast network –Good price/performance ratio –Exploit existing.
Background Computer System Architectures Computer System Software.
Computer System Evolution. Yesterday’s Computers filled Rooms IBM Selective Sequence Electroinic Calculator, 1948.
Lecture # 10 Processors Microcomputer Processors.
Constructing a system with multiple computers or processors 1 ITCS 4/5145 Parallel Programming, UNC-Charlotte, B. Wilkinson. Jan 13, 2016.
Constructing a system with multiple computers or processors
Web Server Administration
Constructing a system with multiple computers or processors
Constructing a system with multiple computers or processors
Constructing a system with multiple computers or processors
Types of Parallel Computers
Cluster Computers.
Presentation transcript:

"Practical Considerations in Building Beowulf Clusters" Lessons from Experience and Future Directions Arch Davis (GS*69) Davis Systems Engineering

Poor light-socket coordination

Parallel Computing Architectures 1. (not parallel) Fastest Possible serial –a. Make it complex –b. Limits 2. Old superscalar, vector Crays, etc. 3. Silicon graphics shared memory (<64 CPUs) 4. Intel shared memory: 2-32 processor servers 5. Distributed memory: “Beowulf” clusters 6. Biggest D.m.: NEC SS6 “Earth Simulator”

+ glue → Cluster ? Building a Beowulf cluster

Some Design Considerations 1. Processor type and speed 2. Single or Dual 3. Type of memory 4. Disk topology 5. Interconnection technology 6. Physical packaging 7. Reliability

“Just a bunch of ordinary PCs” But to be reliable, more must be watched. –Power supplies –Fans –Motherboard components –Packaging layout –Heat dissipation –Power quality To be cost effective, configure carefully. –Easy to overspecify and cost >2x what is necessary –Don’t overdo the connections, they cost a lot. –The old woman swallowed a fly. Be careful your budget doesn’t die.

1. Processor type & speed A. Pentium 4 Inexpensive if not leading edge speed B. Xeon =dual processor P4. Shares a motherboard. C. AMD Opteron 64-bit Needed for >2GB mem. D. (future) Intel 64-bit Will be AMD compatible! E. IBM 970 (G5) True 64-bit design Apple is using F. Intel Itanium “Ititanic” 64-bit long instruction word

Disk Topology 1. Disk per board 2. Diskless + RAID

Interconnect Options Always a desire for way more speed than possible Latency is ultimately an issue of light speed Existing options: 1. Ethernet, including Gigabit Switched Very Robust, by Dave Boggs EECS’72 Affordable, even at Gigabit 2. Infiniband Switched 3. Proprietary: Myrinet, Quadrics, Dolphin Various topologies, including 2&3-D meshes Remote DMA may be transfer method Assumes noise-free channel, may have CRC

Physical Packaging It’s not “rocket science,” but it takes care. A few equations now and then never hurt when you are doing heat transfer design. How convenient is it to service? How compact is the cluster? What about the little things? Lights & buttons? “Take care of yourself, you never know how long you will live.”

Reliability Quality is designed-in, not an accident. Many factors affect reliability. Truism: “All PCs are the same. Buy the cheapest and save.” Mil-spec spirit can be followed without gold plate. Many components and procedures affect the result. Early philosophy: triage of failing modules Later philosophy: Entire cluster uptime Consequence of long uptime: user confidence, greatly accelerated research

Benchmarks ● Not a synthetic ● 100 timesteps of Terra code (John R. Baumgardner, LANL) ● Computational fluid dynamics application ● Navier-Stokes equation with ∞ Prandtl number ● 3D spherical shell multi-grid solver ● Global elliptic problem with 174,000 elements ● Inverting and solving at each timestep Results are with Portland Group pf90 Fortran compiler on –fastsse option And with Intel release 8 Fortran: Compiler MachineIntel Portland baselineP s362 sec lowpowerP4M s358 sec Router2Xeon s305 sec epiphanyXeon s312 sec pntium28P4 2.8/ s209 sec opteron146AMD s164 sec Cray designNEC SX-6 ~50 sec

Software Usually is Linux with MPI for communication. Could be Windows, but not many. Compilers optimize. Management and monitoring software Scheduling software

Linux32-bit/64-bit Windows Pentium 4 Athlon Xeon Opteron PGI ® Workstation – 1 to 4 CPU Systems

Workstation Clusters PGI CDK ™ = PGI Compilers + Open Source Clustering Software A turn-key package for configuration of an HPC cluster from a group of networked Linux workstations or dedicated blades

What about the future? Always go Beowulf if you can. Work on source code to minimize communication. Compilers may never be smart enough to automatically parallelize or second-guess the programmer or the investigator. Components will get faster, but interconnects will always lag processors.

Future Hardware No existing boards are made for clustering. Better management firmware is needed. Blade designs may be proprietary. They may require common components to operate at all. Hard disks need more affordable reliability. Large, affordable Ethernet switches are needed.

General advice? Think of clusters as “personal supercomputers.” They are simplest if used as a departmental or small-group resource. Clusters too large may cost too much: –Overconfigured –Massive interconnect switches –Users can only exploit so many processors at once –Multiple runs may beat one massively parallel run. –Think “lean and mean.”

Opportunities 1. Test these machines with your code. 2. Get a consultation on configuration

More are Coming Peter Bunge sends his greetings In anticipation of a Deutsche Geowulf 256 Processors… And many more clusters here and there. Happy Computing! But, NOT The End