Evaluation of Objectivity Databases on the Sony HSM Software and Tape Robotics - PowerPoint PPT Presentation

evaluation of objectivity databases on the sony hsm software and tape robotics n.
Download
Skip this Video
Loading SlideShow in 5 Seconds..
Evaluation of Objectivity Databases on the Sony HSM Software and Tape Robotics PowerPoint Presentation
Download Presentation
Evaluation of Objectivity Databases on the Sony HSM Software and Tape Robotics

play fullscreen
1 / 29
Evaluation of Objectivity Databases on the Sony HSM Software and Tape Robotics
116 Views
Download Presentation
nodin
Download Presentation

Evaluation of Objectivity Databases on the Sony HSM Software and Tape Robotics

- - - - - - - - - - - - - - - - - - - - - - - - - - - E N D - - - - - - - - - - - - - - - - - - - - - - - - - - -
Presentation Transcript

  1. Evaluation of Objectivity Databases on the Sony HSM Software and Tape Robotics Michael Athanas Cornell Univ.* and Nobu Katayama, Atsushi Manabe KEK * on leave to Cereon Genomics

  2. Motivation • In upcoming HEP experiment, managing a huge volume of complex data is necessary. For handling such a data, Objectivity/DB is becoming a popular choice in HEP. • Better understanding of how the Objectivity/DB works with a rather conventional Hierarchical Storage Management system (HSM). Evaluattion of Objy/DB on Sony HSM. CHEP 98 CHICAGO

  3. HSM pros and cons • Popular software on many venders and affordable. • Unit of transfer between disk and tape is file • access time overhead • space inefficiency • User cannot assign any tapes to store the file. Evaluattion of Objy/DB on Sony HSM. CHEP 98 CHICAGO

  4. Prototyping Test • Schema: Based on Cornell Nile Proj*/objyDB. * presented in CHEP97 • Data Base software: Objectivity/DB 4.0.10 • Hardware: A part of the KEKB computer system. Evaluattion of Objy/DB on Sony HSM. CHEP 98 CHICAGO

  5. Test Bed System (1) Tape Lib. • Server:SUN UE 6000 (US1 166 MHz x4CPU) • Tape Library: Sony PetaSite* : 30TBSony DTF (8MB/s) x 8 • Disk: MSS RAID 80GB • HSM soft: Sony PetaServe* (OSM ver.2.1 base) SCSI2/W WS SCSI2/W RAID * http://www.sony.com/professional Evaluattion of Objy/DB on Sony HSM. CHEP 98 CHICAGO

  6. Test Bed System(2) • HSM disk Layer: 10GBytes • HSM tape Layer: 100GBytes • Tape Media Capacity: 10GB/tape • HSM Water Mark:(adjustable param.) High (force data move to tape) =8GB Low (move until this size) =4GB Shadow =1GB Evaluattion of Objy/DB on Sony HSM. CHEP 98 CHICAGO

  7. Test Bed System(3) • Disk • Max. read/write rate: 8/8 MB/s • Tape • Max. read/write rate: 12MB/s • media loading time: 30~60 sec (with positioning & mounting) • Network • DB Clients & Server are on the same machine • Disk and Tape are attached locally Evaluattion of Objy/DB on Sony HSM. CHEP 98 CHICAGO

  8. Class association DataSetCollection DataSet(Hadron) ……….... DataSet(Taus) x 100 RunCollection(1) ……….. RunCollection(100) x10 Run(1) ……….. Run(9) x1000 Event(1) ………….. Event(1000) ……….. EventRecord EventRecord Evaluattion of Objy/DB on Sony HSM. CHEP 98 CHICAGO

  9. Evaluattion of Objy/DB on Sony HSM. CHEP 98 CHICAGO

  10. Each DB is a UNIX file and Unit of HSM migration

  11. Data Size • Event Inf. ~50B • Event Record 4k~16kB • Run Inf. ~50B • RunCollection DB size = ~ 120MB • Total DataSet size = ~12GB • Tape Media Size = 10GB (small type media ) Evaluattion of Objy/DB on Sony HSM. CHEP 98 CHICAGO

  12. Population (object creation) • PageSize = 64kB • DataSet = 12GB population. • Single process / Multi process concurrent population Evaluattion of Objy/DB on Sony HSM. CHEP 98 CHICAGO

  13. Population flow diagram Define DB file disk 10GB Begin Run Obj write Run Loop 50B Commit RunCollection Loop Event header write ~12kB >Over HighWM Event Record write & close 50B End Run Obj write Commit Tape 100GB Evaluattion of Objy/DB on Sony HSM. CHEP 98 CHICAGO

  14. Population test result • Average write Rate: 1.2MB/sec • 1 DataSet (12GB) creation • Tape migration was negligible. • NO HSM overhead was seen. • Population in parallel (multi process) • Scalability: saturation was seen.(memory shortage ?) Evaluattion of Objy/DB on Sony HSM. CHEP 98 CHICAGO

  15. Population speed Evaluattion of Objy/DB on Sony HSM. CHEP 98 CHICAGO

  16. Evaluattion of Objy/DB on Sony HSM. CHEP 98 CHICAGO

  17. Data base access (1) • Access pattern can make big difference in its performance. • Data reloading time (Tape->Disk) affects much. Evaluattion of Objy/DB on Sony HSM. CHEP 98 CHICAGO

  18. Data base access (2) • Deep Scan • Load all data into memory • Light Scan • Just Event header data is scan. • Index Scan • 1/200 data access by event number Index search (ooEqualLookup). • At first all data are moved to Tape (migrate) then start access. Evaluattion of Objy/DB on Sony HSM. CHEP 98 CHICAGO

  19. Deep Scan • 6GB full scan (500 Runs) • On HSM • 3.4MB/sec 337 obj/sec 1482s • On normal disk (reference) • 7.3MB/sec 725 obj/sec 690s • Tape Reloading Time (reference) • 50x120MB file 1100s Evaluattion of Objy/DB on Sony HSM. CHEP 98 CHICAGO

  20. Light Scan • 250MB scan in 6GB (500 Runs) • On HSM • 28kB/sec 568 obj/sec 880s = 5.7MB/sec seq. access equiv. • On normal disk (reference) • 830kB/sec 16k obj/sec 30s = 166MB/sec seq. access equiv. • Tape Reloading Time (reference) • 50x120MB file 5.5MB/s 1100s Evaluattion of Objy/DB on Sony HSM. CHEP 98 CHICAGO

  21. Index Scan • 30MB (0.5%)scan in 6GB (500 Runs) • On HSM • 27kB/sec 2.7 obj/sec 950s = 5.3MB/sec seq. access equiv. • On normal disk (reference) • 116kB/sec 12 obj/sec 210s = 23MB/sec seq. access equiv. • Tape Reloading Time (reference) • 100x120MB file 6MB/s 1002s Evaluattion of Objy/DB on Sony HSM. CHEP 98 CHICAGO

  22. Seq. access equiv. speed • Required speed in using conventional sequential tape scan to fetch the data in the time. • Speed = Total Scaned object size /elapsed time • Seq. acc. equiv. speed = Total data size /elapsed time Evaluattion of Objy/DB on Sony HSM. CHEP 98 CHICAGO

  23. Event Search EventAtr= XXX RUN(I) Event Record Index table RUN(I+1) objyDB/HSM Conventional Tape access Evaluattion of Objy/DB on Sony HSM. CHEP 98 CHICAGO

  24. Index scan access speed on HSM on pure disk Evaluattion of Objy/DB on Sony HSM. CHEP 98 CHICAGO

  25. Conflicting tape access • Index Search with multi processes • 1 process/HSM31kB/sec 2.7 obj/sec 950s = 6.3MB/sec seq. access Equiv. • 2 process/HSM with access conflict7kB/sec 0.6 obj/sec 3900s = 1.5MB/sec seq. access Equiv. • 2 process/Disk (reference)1.7MB/sec 13 obj/sec 200s= 30MB/sec seq.access Equv. Evaluattion of Objy/DB on Sony HSM. CHEP 98 CHICAGO

  26. Typical conflict situationSimultaneous access to single tape Process 1 Back & Forth Process 2 Evaluattion of Objy/DB on Sony HSM. CHEP 98 CHICAGO

  27. Summary (1) • ObjectivityDB + PetaServ (HSM) generally works well. It could be a cost effective solution. • HSM overhead in DB write (populate) was negligible in our program. More efficiency was achieved by concurrent population with multiple processes. Evaluattion of Objy/DB on Sony HSM. CHEP 98 CHICAGO

  28. Summary(2) • The performance of accessing data directly from the HSM under three access pattern was measured. • The access tests showed that objyDB/HSM could get comparable speed with conventional sequential tape access in addition to the OODB merits. Evaluattion of Objy/DB on Sony HSM. CHEP 98 CHICAGO

  29. Summary(3) • Conflictive access to single tape among processes much degrades DB access speed. To avoid such a situation, access scheduling is one of a solution. • Strategy: • Access DB file in the order of tape file position. • Avoid concurrent accesses to DB files in single tape. Evaluattion of Objy/DB on Sony HSM. CHEP 98 CHICAGO