Presentation on theme: "Building Beowulfs for High Performance Computing Duncan Grove Department of Computer Science University of Adelaide"— Presentation transcript:
Building Beowulfs for High Performance Computing Duncan Grove Department of Computer Science University of Adelaide
Anatomy of a “Beowulf” “Cluster” of networked PCs –Intel PentiumII or Compaq Alpha –Switched 100Mbit/s Ethernet or Myrinet –Linux –Parallel and batch software support Switching Infrastructure n1nNn2 Front-end Node Outside World Compute Nodes
Why build Beowulfs? Science/$ Some problems take lots of processing Many supercomputers are used as batch processing engines –Traditional supercomputers wasteful high throughput computing Beowulfs: –“ [useful] computational cycles at the lowest possible price.” –Suited to high throughput computing –Effective at an increasingly large set of parallel problems
Three Computational Paradigms Data Parallel –Regular grid based problems Parallelising compilers, eg HPF Eg physicists running lattice gauge calculations Message Passing –Unstructured parallel problems. MPI, PVM Eg chemists running molecular dynamics simulations. Task Farming –“High throughput computing” - batch jobs Queuing systems Eg chemists running Gaussian.
A Brief Cluster History Caltech Prehistory Berkeley NOW NASA Beowulf Stone SouperComputer USQ Topcat UIUC NT Supercluster LANL Avalon SNL Cplant AU Perseus?
Beowulf Wishlist Single System Image (SSI) –Unified process space –Distributed shared memory –Distributed file system Performance easily extensible –Just “add more bits” Is fault tolerant Is “simple” to administer and use
Current Sophistication? Shrinkwrapped “solutions” or do-it-yourself –Not much more than a nicely installed network of PCs –A few kernel hacks to improve performance –No magical software for making the cluster transparent to the user –Queuing software and parallel programming software can create the appearance of a more unified machine
Iofor Learning platform Program development Simple benchmarking Simple performance evaluation of real applcaions Teaching machine Money lever
iMacwulf Student lab by day, Beowulf by night? –MacOS with Appleseed –LinuxPPC 4.0, soon LinuxPPC 5.0 –MacOS/X
“Gigaflop harlotry” MachineCost# Processors~ Peak Speed Cray T3E10s million Gflop/s SGI Origin s million Gflop/s IBM SP210s million512400Gflop/s Sun HPC1s million6450Gflop/s TMC CM55 Million (1992)12820Gflop/s SGI PowerChallenge1 Million (1995)2020Gflop/s Beowulf cluster + myrinet1 Million256120Gflop/s Beowulf cluster300K256120Gflop/s
The obvious, but important In the past: –Commomdity processors way behind supercomputer processors –Commodity networks way, way, way behind supercomputer networks In the now: –Commomdity processors only just behind supercomputer processors –Commmodity networks still way, way behind supercomputer networks –More exotic networks still way behind supercomputer networks In the future: –Commodity processors will be supercomputer processors –Will the commodity networks catch up?
Open Source The good... –Lots of users, active development –Easy access to make your own tweaks –Aspects of Linux are still immature, but recently SGI has release xfs as open source Sun has released its HPC software as open source And the bad... –There’s a lot of bad code out there!
Network technologies So many choices! –Interfaces, cables, switches, hubs; ATM, Ethernet, Fast Ethernet, gigabit Ethernet, firewire, HiPPI, serial HiPPI, Myrinet, SCI… The important issues –latency –bandwidth –availability –price –price/performance –application type!
Disk subsystems I/O a problem in parallel systems –Data not local on compute nodes is a performance hit –Distributed file systems CacheFS CODA –Parallel file systems PVFS On-line bulk data is interesting in itself –Beowulf Bulk Data Server cf with slow, expensive tape silos...
Perseus Machine for chemistry simulations –Mainly high throughput computing –RIEF grant in excess of $300K –128 nodes. For < $2K per node Dual processor PII450 At least 256MB RAM –Some nodes up to 1GB 6GB local disk each –5x24 (+2x4) port Intel 100Mbit/s switches
Perseus: installing a node Switching Infrastructure n1nNn2 Front-end Node Outside World User node, administration, compilers, queues, nfs, dns, NIS, /etc/*, bootp/dhcp, kickstart,... Floppy disk or bootrom
Software on perseus Software to support the three computational paradigms –Data Parallel Portland Group HPF –Message Passing MPICH, LAM/MPI, PVM –High throughput computing Condor, GNU Queue Gaussian94, Gaussian98
Expected parallel performance Loki, 1996 –16 Pentium Pro processors, 10Mbit/s Ethernet –3.2 Gflop/s peak, achieved 1.2 real Gflop/s on Linpack benchmark Perseus, 1999 –256 PentiumII processors, 100Mbit/s Ethernet –115 Gflop/s peak ~40 Gflop/s on Linpack benchmark? Compare with top 500! –Would get us to about 200 currently –Other Australian machines? NEC BOM at #102 Sun HPC at #181, #182, #255 Fujitsi ANU at #400
Reliability in a large system Build it right! Is the operating system and software running ok? Is heat dissipation going to be a problem? –Monitoring daemon Normal features –CPU, network, memory, disk More exotic features –Power supply and CPU fan speeds –Motherboard and CPU temperatures Do we have any heisen-cabling? –Racks and lots of cable ties!
The limitations... Scalability Load balancing –Effects of machines capabilities –Desktop machines vs. dedicated machines –Resource allocation –Task Migration Distributed I/O System monitoring and control tools Maintenance requirements –Installation, upgrading, versioning Complicated scripts Parallel interactive shell?
… and the opportuntities A large proportion of the current limitations compared with traditional HPC solutions are merely systems integration problems Some contributions to be made in –HOWTOs –Monitoring and maintenance –Performance modelling and real benchmarking