1 STAR STAR Upgrade Plans and R&D Open Meeting on RHIC Planning, December 4, 2003 R. Majka for STAR.
Grid Components in STAR: Experience, Needs and Plans
description
Transcript of Grid Components in STAR: Experience, Needs and Plans
![Page 1: Grid Components in STAR: Experience, Needs and Plans](https://reader036.fdocuments.in/reader036/viewer/2022083006/56813e66550346895da872a0/html5/thumbnails/1.jpg)
1
Grid in STAR: Needs and PlansA. Vaniachine
STAR
Grid Components in STAR: Grid Components in STAR: Experience, Needs and PlansExperience, Needs and Plans
A. Vaniachine
Lawrence Berkeley National Laboratory
July 14, 2000
PPDG Workshop, Argonne
![Page 2: Grid Components in STAR: Experience, Needs and Plans](https://reader036.fdocuments.in/reader036/viewer/2022083006/56813e66550346895da872a0/html5/thumbnails/2.jpg)
2
Grid in STAR: Needs and PlansA. Vaniachine
STAR
OutlineOutline
• STAR is taking data:
• Grid in analysis
• Grid for production?
• Plans
![Page 3: Grid Components in STAR: Experience, Needs and Plans](https://reader036.fdocuments.in/reader036/viewer/2022083006/56813e66550346895da872a0/html5/thumbnails/3.jpg)
3
Grid in STAR: Needs and PlansA. Vaniachine
STAR
Decisions TakenDecisions Taken
• Event split into components:– daq (raw data) or fzd (Geant3)– dst: legacy C-structures– event: same info in OO– tags: like an n-tuple– hist: for QA– ...
• ROOT-based Architecture– Persistency:
• STL containers - structural and reference
• schema evolution– Analysis:
“good-bye” PAW
![Page 4: Grid Components in STAR: Experience, Needs and Plans](https://reader036.fdocuments.in/reader036/viewer/2022083006/56813e66550346895da872a0/html5/thumbnails/4.jpg)
4
Grid in STAR: Needs and PlansA. Vaniachine
STAR
Grid Components Deployed for Grid Components Deployed for AnalysisAnalysis
• tag database: to speed up user response– data in ROOT TTree files (n-tuple):
• data description stored together with the data (solves the schema evolution problem)• comments preserved• 2-level hierarchy: 500 tags are in seven groups• split mode for structures (not objects: no need to
load classes to analyze tags)
• indexed tags (Grand Challenge deployed)– efficient cache management– transparent user access to HPSS
in principle, to the raw data as well:“the Dream” is coming true
![Page 5: Grid Components in STAR: Experience, Needs and Plans](https://reader036.fdocuments.in/reader036/viewer/2022083006/56813e66550346895da872a0/html5/thumbnails/5.jpg)
5
Grid in STAR: Needs and PlansA. Vaniachine
STAR
Component-based Component-based ArchitecturesArchitectures
• Grand Challenge:– large size components one component per developer
• NOVA:–smaller-size components one components per task
Needed: small components, easily adaptable to experiments
e.g., “ls” for the HPSS
![Page 6: Grid Components in STAR: Experience, Needs and Plans](https://reader036.fdocuments.in/reader036/viewer/2022083006/56813e66550346895da872a0/html5/thumbnails/6.jpg)
6
Grid in STAR: Needs and PlansA. Vaniachine
STAR
Grid Components for Production?Grid Components for Production?
$$$
constraints
have to use low-cost commodity components:• latency tolerant algorithms
have to load all distributed nodes:• sustained data transfer methods
have to use HPPS data storage:• efficient cache management• transparent user access
![Page 7: Grid Components in STAR: Experience, Needs and Plans](https://reader036.fdocuments.in/reader036/viewer/2022083006/56813e66550346895da872a0/html5/thumbnails/7.jpg)
7
Grid in STAR: Needs and PlansA. Vaniachine
STAR
Plans for PPDGPlans for PPDG
• Deploy file transfer methods for the BNL-LBNL linkcurrent rate: rcf pdsf = 100 GB/day
• Managed file caching for high-throughput computing–pdsf as a development test-bed–rcf as a target for deployment