© 2005 Altera Corporation © 2006 Altera Corporation Batch Computing at Altera Condor, Quill and The Enterprise
© 2006 Altera Corporation 2 About Altera “The Programmable Solutions Company” Pioneer of SOPC technology Founded in 1983 $1.02 billion in 2004 sales 2,300 employees 14,000+ worldwide customers
© 2006 Altera Corporation 3 Programmable Logic Devices (PLDs) Intellectual Property (IP) Development Software About Programmable Solutions
© 2006 Altera Corporation 4 About Me Senior Software Engineer at the Toronto Technology Center B.A.Sc. in Engineering Science from the University of Toronto Joined Altera in 2001 Focus on distributed computing
© 2006 Altera Corporation 5 Where It All Began Developed in Toronto Centralized scheduling system Multiple queues Priority/FIFO execution No limit on resource claims Engineer-designed, custom API
© 2006 Altera Corporation 6 Change Is Good, Right? Multi-OS support Redundancy and fault tolerance Easy expansion beyond Toronto Easy-to-use API New features Improve matchmaking Capacity planning Really Important!
© 2006 Altera Corporation 7 Pain Free Migration TTC POOL USER TOOLS CONDOR POOL META SCHEDULER Priority Engine CONDOR TTC SOAP DB
© 2006 Altera Corporation 8 Time Stands Still Nice-style priorities [1:N] Use priority factor to ensure P N negotiates before P N+1, P N+2, etc. RUP(P N ) = 0.5 EUP(P N )/EUP(P N+1 ) = ½ Freeze RUP values in time PRIORITY_HALFLIFE = Let jobs at P N get all VMs in the system NEGOTIATOR_IGNORE_USER_PRIORITIES = True
© 2006 Altera Corporation 9 Translation Services 1 2 windows fitter META SCHEDULER +AlteraClusterID = 1 +AlteraGroup = fitter requirements = OpSys =... +AccountingGroup = P1 AlteraTargetOs = windows... +AlteraJobID =1... queue +AlteraJobID = 2...
© 2006 Altera Corporation 10 SQL! SQL! Everywhere! POSTGRESQL DBMS META SCHEDULER CONDOR QUILL CONDOR COLLECTOR USAGE HISTORY STATUS INFO SYSTEM AUDITS
© 2006 Altera Corporation 11 Better Planning Through Pictures
© 2006 Altera Corporation 12 From Here, Where? Roll out across the enterprise Scaling with multiple schedds Quill++ DBMS for configuration management (with R. Nordlund & J. Stowe from The Hartford)