180 likes | 272 Views
Advanced User Support. Amit Majumdar 8/13/09. Outline. Three categories of AUS Operational Activities AUS.ASTA AUS.ASP ASTA example. Three Categories of AUS. A dvanced S upport for T eraGrid A pplications ( AUS . ASTA )
E N D
Advanced User Support Amit Majumdar 8/13/09
Outline • Three categories of AUS • Operational Activities • AUS.ASTA • AUS.ASP • ASTA example
Three Categories of AUS • Advanced Support for TeraGrid Applications (AUS.ASTA) • AUS staff work on a particular user’s code (some community codes) • Guided/initiated by the allocations process • Advanced Support for Projects (AUS.ASP) • Project initiated by AUS staff (jointly with users) – impacts many users • Optimally install software, maintain, train users • Advanced Support for EOT (AUS.ASEOT) • Advanced HPC/CI trainings • Participate in workshops, tutorials (TG09, PetaScale, SCXY) • Interaction with other CI programs PlantCI, DataNet etc.
Update on Operational Activities • Bi-weekly telecon among AUS POCs from every RP site • Matching of AUS staff to ASTA projects, discussion about ASP, EOT activities • Program plans, annual reports etc. • Web/Tele-con among AUS technical staff • Biweekly technical tele/web-conference on ASTA and other projects - using readytalk • Since we started (early October, 2008) 20 tele/web-con sessions and 31 technical presentations (including today)
ASTAs New Supplemental ASTAs Since last Round Table • Total ASTAs: 44 • Two more Startup/Supplemental ASTAs approved
Advanced Support Projects • Three projects (two completed first phase) • Benchmarking Molecular Dynamics codes (SDSC, PSC) – completed first phase • Benchmarking Materials Science codes (PSC, TACC, LONI, NCSA, NICS) – completed first phase • Hybrid Programming model for multi-core (TACC, SDSC, PSC, NICS) jointly with XS-WG - continuing
ASP Benchmark Webpage Doc group has helped in setting up of AUS benchmark pages Moved from earlier staging area webpage to production doc page: http://www.teragrid.org/userinfo/aus/benchmark.php
ASTA Example:As a part of Adrian Roitberg ASTA GPU Implementation of AMBER by Ross Walker (SDSC) jointly with NVIDIA experts Other ASTA team members SudhakarPamidighantam (NCSA ), Christian Halloy (NICS )
GPU Support • Collaboration with NVIDIA to produce CUDA version of AMBER. • PMEMD Engine • Implicit Solvent GB (V1.0 complete) • Explicit Solvent PME (in progress) • Focus on accuracy. • It MUST pass the AMBER regression tests. • Energy conservation MUST be comparable to double precision CPU code.
GPU Accuracy • Use of double precision in all places severely limits performance. • Make careful use of double precision where needed. • Calculate in single precision. • Accumulate in double precision. • Avoid large dynamic ranges in arithmetic expressions. • Switch over to double precision automatically if the dynamic range is too high.
Provisional Performance • 1 x Tesla C1060 (NCSA Lincoln S1070 – 4X) Timings ------- System Wall Time (s) NS/day ACE-ALA3-NME(GPU) 48.57 355.92 ACE-ALA3-NME(4xCPU)* 34.78 496.84 (*Note 4 cpu since we need 10x > atoms than processors) TRPCage(GPU) 62.48 276.57 TRPCage(8xCPU) 150.84 114.56 Myoglobin(GPU) 62.22 27.72 Myoglobin(8xCPU) 392.34 4.04 Nucleosome(GPU) 332.03 0.520 Nucleosome(8xCPU) 2877.60 0.060 Cray XT5 (1024xCPU) 175.49 0.985