Presentation is loading. Please wait.

Presentation is loading. Please wait.

Developing a Cluster Strategy for NPACI All Hands Meeting Panel Feb 11, 2000 David E. Culler Computer Science Division University of California, Berkeley.

Similar presentations


Presentation on theme: "Developing a Cluster Strategy for NPACI All Hands Meeting Panel Feb 11, 2000 David E. Culler Computer Science Division University of California, Berkeley."— Presentation transcript:

1 Developing a Cluster Strategy for NPACI All Hands Meeting Panel Feb 11, 2000 David E. Culler Computer Science Division University of California, Berkeley http://www.cs.berkeley.edu/~culler

2 2/11/2000NPACI Clusters2 x86+Myrinet platforms w/ GbE inter-networking UCB Millennium Cluster of Clusters PIII-X 64x4 PII 8x2 PIII 32x2 ½ TB DLIB PII PIII Gigabit Ethernet (GbE) Ninja Math Bio CE Physics Astro NTON Internet-2 SuperNet Mobile Svcs Kiosks NOW Distributed ownership, allocation, and management

3 2/11/2000NPACI Clusters3 Vineyard Cluster Architecture Distributed resource utilization and management in a “Vineyard” of Clusters. - VIA / GM, GbE - Multicast Applications / Services (ISPACE/Kiosks) - NT / Linux (2.2.x) - Stride Scheduler MPIVEXEC PBS I/O Mgmt / Monitoring REXEC TOOLS Rootstock Distribution

4 2/11/2000NPACI Clusters4 Clusters “own” HPC

5 2/11/2000NPACI Clusters5 Fundamental Advantages of Clusters Cost Performance Performance / Cost Track leading edge of market technology Incremental scalability Availability Tremendous I/O performance Wide-Area Network performance –competitive internal network performance too Allow specialization of networked services

6 2/11/2000NPACI Clusters6 Fundamental Challenges Management Complete system on every node –need scalable administration Incremental scalability & availability => –heterogeneity –some parts inoperable at any time The Cluster projects are making great progress in this area –eg: Millennium rootstock Cluster tools are what you want for managing the desktops across your department

7 2/11/2000NPACI Clusters7 CS&E HPC hampered by “self-centered” usage model Have my own application for my studies Want the entire machine to myself Want it now Think “services” Think “software” The value is in your application. Make it a service and make it available to the scientific community. Put it on a cluster to deliver results 24x7 x 52

8 2/11/2000NPACI Clusters8 Example: TCAD Simulation Service star formation simulation earthquake simulations phylogeny, BLAST,... http://cuervo.eecs.berkeley.edu/Volcano/

9 2/11/2000NPACI Clusters9 Extreme Example UCB Millennium / NOW has deliver 70 CPU years! Simple special case, but... Engineered for portability, adaptability, availability

10 2/11/2000NPACI Clusters10 What should NPACI do? To be relevant: become a “Center of Expertise” for clusters draw expertise toward the center for ease of dissemination facilitate and encourage building clusters among the partners invest in an interesting cluster “close to home” –cheap! Graft Millennium invest in people to understand the implications To Lead: Pioneer widespread computational science and engineering services infiniband

11 2/11/2000NPACI Clusters11 from e-commerce to

12 2/11/2000NPACI Clusters12 Technical Backup Slides

13 2/11/2000NPACI Clusters13 Rootstock Mechanics K cluster stock - build - os - drvrs - mill SW - os mods leased builds cs CAN Cluster System Distribution Center... IP network 1. Cluster Stock - Rootstock build pages - Full Current Linux - all fixes and pckgs - SSL, SSH - Cluster Drivers - Cluster System Layers - rexec, mpe, pbs - Optional SW ($) - Cluster Kernal Mods 5. Cluster Update button (future) - 2nd dialtone, CF engine, rolling update 2. Make the CS “graft” - specify IP address - pckg removes - dchp, dns, nis,... sanity check and build - resolv.conf, /etc/hosts,... constructs cluster build (lease) download CS build floppy Cluster 3. CS power-on build - xfer and localize DT - add local admin scripts - node build floppy 4. Node power-on build - local stock from CS

14 2/11/2000NPACI Clusters14 REXEC / VEXEC Components –rexecd, rexec & vexecd rexecd vexecd (Policy A) rexec Cluster IP Multicast Channel %rexec –n 2 –r 3 indexer minimum $ vexecd (Policy B) Node ANode BNode CNode D “Nodes AB” run indexer on Nodes AB at 3 credits/min

15 2/11/2000NPACI Clusters15 Computational Economy Market-based approach to resource allocation –Optimizes for user value Resources Economic F.E. APIAPI APIAPI Access Modules Resource Managers Time Share Batch Queue Apps (Value)


Download ppt "Developing a Cluster Strategy for NPACI All Hands Meeting Panel Feb 11, 2000 David E. Culler Computer Science Division University of California, Berkeley."

Similar presentations


Ads by Google