1. 1 Grid File Replication using Storage Resource Management Presented By Alex Sim Contributors: JLAB: Bryan Hess, Andy Kowalski Fermi: Don Petravick, Timur Perelmutov, Rich Wellner LBNL: Junmin Gu, Vijaya Natarayan,Ekow Otoo, LBNL: Junmin Gu, Vijaya Natarayan, Ekow Otoo, Alex Romosan,Alex Sim, Arie Shoshani Alex Romosan, Alex Sim, Arie Shoshani WP2-EDG: Wolfgang Hoschek, Peter Kunszt, Heinz Stockinger, Kurt Stockinger WP5-EDG: Jean-Philippe Baud http://sdm.lbl.gov/srm

2. 2 How does file replication use SRMs (high level view)

3. 3 Main advantages of using SRMs for file replication Can work in front of MSSCan work in front of MSS To provide pre-staging To provide queued archiving Monitor transfer in/out of MSSMonitor transfer in/out of MSS Recover in case of transient failures Reorders pre-staging requests to minimize tape mounts Monitors GridFTP transfersMonitors GridFTP transfers Re-issues requests in case of failure Can control number of concurrent GridFTP transfers to optimize network use (future)Can control number of concurrent GridFTP transfers to optimize network use (future) SRMs role in the data replicationSRMs role in the data replication Storage resource coordination & scheduling SRMs do not perform file transfer SRMs do invoke file transfer service if needed (GridFTP)

4. 4 Brief history of SRM since GGF4 Agreed on single API for multiple storage systems Jlab has an SRM implementation based on SRM v1.1 spec on top of JASMine Fermi Lab has an SRM implementation based on SRM v1.1 spec on top of Enstore WP5-EDG is proceeding with SRM implementation on top of Castor LBNL Deployed HRM-HPSS (which accesses files in/out of HPSS) at BNL, ORNL, and NERSC (PDSF) HRM-NCAR that accesses MSS at NCAR is in progress

5. 5 SRM version 2.0 Joint design and specification of SRM v2.0 LBNL organized meeting to coordinate design, and summarized design conclusions SRM v2.0 spec draft version exists SRM v2.0 finalization to be done in Dec. Design uses OGSA service concept Define interface & behavior Select protocol binding (WSDL/SOAP) Permit multiple implementations Disk Resource Managers (DRMs) On top of multiple MSSs (HRMs)

6. 6 Brief Summary SRM main methods srmGet, srmPut, srmCopysrmGet, srmPut, srmCopy Multiple files srmGet from remote location to disk/tape srmPut from client to SRM disk/tape srmCopy from remote location to SRM disk/tape srmReleasesrmRelease Pinning automatic If not provided, apply pinning lifetime srmStatussrmStatus Per file, per request Time estimate srmAbortsrmAbort

7. 7 Main Design Points Interfaces to all types of SRMs to be uniformInterfaces to all types of SRMs to be uniform Any Clients, Middleware modules, other SRMsAny Clients, Middleware modules, other SRMs Will communicate with SRMs Support a “multiple files” requestSupport a “multiple files” request set of files, not ordered, no bundles Implies: queuing, status, time estimates, abort SRMs support asynchronous requestsSRMs support asynchronous requests Non-blocking, unlike FTP and other services Support long delays, multi-file requests Support call-backsSupport call-backs Plan to use “event notification service” Automatic Garbage CollectionAutomatic Garbage Collection In file replication, all files are “volatile” As soon as they are moved to target, SRMs perform “garbage collection” automatically.

8. 8 Current efforts on SRMs at LBNL Deployed HRM at BNL, LBNL (NERSC/PDSF), ORNL for HPSS accessDeployed HRM at BNL, LBNL (NERSC/PDSF), ORNL for HPSS access Developing HRM at NCAR for MSS accessDeveloping HRM at NCAR for MSS access Deployed GridFTP-HRM-HPSS connection daemonDeployed GridFTP-HRM-HPSS connection daemon Supports multiple transport protocolsSupports multiple transport protocols gsiftp, ftp, http, bbftp and hrm Deployed web-based File Monitoring Tool for HRMDeployed web-based File Monitoring Tool for HRM especially useful for large file replica requests Deployed HRM client command programsDeployed HRM client command programs User convenience Currently developing web-services gatewayCurrently developing web-services gateway Currently developing GSI-enabled requestsCurrently developing GSI-enabled requests

9. 9 Web-Based File Monitoring Tool

10. 10 HRMs in PPDG (high level view) Disk Cache Disk Cache HRM-COPY HRM-GET Replica Coordinator HRM (performs writes) HRM (performs reads) LBNLBNL GridFTP GET (pull mode)

11. 11 HRMs in ESG (high level view)

12. 12 File replication from NCAR to ORNL/LBNL HPSS controlled at NCAR

13. 13 File Replication from ORNL HPSS to NERSC HPSS controlled at NCAR : “Non-Blocking Calls”

14. 14 Recent Measurements of large multi-file replication Event time Event

15. 15 Recent Measurements of large multi-file replication (GridFTP transfer time)

16. 16 Recent Measurements of large multi-file replication (Archiving time)

17. 17 Recent Measurements of large multi-file replication Event time

18. 18 Recent Measurements of large multi-file replication (GridFTP transfer time)

19. 19 Recent Measurements of large multi-file replication (Archiving time)

20. 20 File Replication from ORNL HPSS to NERSC HPSS controlled at NCAR 0 Request_Arrive_at_LBNL 1 Request_Arrived_at_ORNL 2 Staging_requested_at_ORNL 3 Staging_started_ORNL 4 Staging_finished_ORNL 5 Callback_from_ORNL_to_LBNL 6 GridFTP_Start_by_LBNL 7 Transfered_from_ORNL_to_LBNL 8 Migration_Requested 9 Migration_Started 10 Migration_Finished 11 Notified_Client HRM Client on NCAR HRM on LBNL and ORNL Max Concurrent PFTPs=5 Cache Size at LBNL=30G Cache Size at ORNL=25G Max Concurrent Pinned File=5 Max Concurrent GridFTP=3 GridFTP –p=4 GridFTP –bs=1000000