Report on the HPSS Database Benchmark
Concerns about the long-term viability of SFS as the metadata store for HPSS have been increasing. A concern that Transarc may discontinue support for SFS motivates us to consider alternative means to store HPSS metadata. The obvious alternative is a commercial database. Commercial databases have th...
Main Author: | |
---|---|
Other Authors: | |
Format: | Report |
Language: | English |
Published: |
Lawrence Livermore National Laboratory
2001
|
Subjects: | |
Online Access: | https://doi.org/10.2172/15006190 https://digital.library.unt.edu/ark:/67531/metadc1405330/ |
id |
ftunivnotexas:info:ark/67531/metadc1405330 |
---|---|
record_format |
openpolar |
spelling |
ftunivnotexas:info:ark/67531/metadc1405330 2023-05-15T16:02:04+02:00 Report on the HPSS Database Benchmark Fisher,D United States. Department of Energy. 2001-06-19 PDF-FILE: 11 SIZE: 0.6 MBYTES pages Text https://doi.org/10.2172/15006190 https://digital.library.unt.edu/ark:/67531/metadc1405330/ English eng Lawrence Livermore National Laboratory rep-no: UCRL-ID-144274 grantno: W-7405-ENG-48 doi:10.2172/15006190 osti: 15006190 https://digital.library.unt.edu/ark:/67531/metadc1405330/ ark: ark:/67531/metadc1405330 Other Information: PBD: 19 Jun 2001 Management Storage 99 General And Miscellaneous//Mathematics Computing And Information Science Competition Programming Viability Maintenance Benchmarks Performance Market Report 2001 ftunivnotexas https://doi.org/10.2172/15006190 2019-02-09T23:08:25Z Concerns about the long-term viability of SFS as the metadata store for HPSS have been increasing. A concern that Transarc may discontinue support for SFS motivates us to consider alternative means to store HPSS metadata. The obvious alternative is a commercial database. Commercial databases have the necessary characteristics for storage of HPSS metadata records. They are robust and scalable and can easily accommodate the volume of data that must be stored. They provide programming interfaces, transactional semantics and a full set of maintenance and performance enhancement tools. A team was organized within the HPSS project to study and recommend an approach for the replacement of SFS. Members of the team are David Fisher, Jim Minton, Donna Mecozzi, Danny Cook, Bart Parliman and Lynn Jones. We examined several possible solutions to the problem of replacing SFS, and recommended on May 22, 2000, in a report to the HPSS Technical and Executive Committees, to change HPSS into a database application over either Oracle or DB2. We recommended either Oracle or DB2 on the basis of market share and technical suitability. Oracle and DB2 are dominant offerings in the market, and it is in the best interest of HPSS to use a major player's product. Both databases provide a suitable programming interface. Transaction management functions, support for multi-threaded clients and data manipulation languages (DML) are available. These findings were supported in meetings held with technical experts from both companies. In both cases, the evidence indicated that either database would provide the features needed to host HPSS. Report DML University of North Texas: UNT Digital Library |
institution |
Open Polar |
collection |
University of North Texas: UNT Digital Library |
op_collection_id |
ftunivnotexas |
language |
English |
topic |
Management Storage 99 General And Miscellaneous//Mathematics Computing And Information Science Competition Programming Viability Maintenance Benchmarks Performance Market |
spellingShingle |
Management Storage 99 General And Miscellaneous//Mathematics Computing And Information Science Competition Programming Viability Maintenance Benchmarks Performance Market Fisher,D Report on the HPSS Database Benchmark |
topic_facet |
Management Storage 99 General And Miscellaneous//Mathematics Computing And Information Science Competition Programming Viability Maintenance Benchmarks Performance Market |
description |
Concerns about the long-term viability of SFS as the metadata store for HPSS have been increasing. A concern that Transarc may discontinue support for SFS motivates us to consider alternative means to store HPSS metadata. The obvious alternative is a commercial database. Commercial databases have the necessary characteristics for storage of HPSS metadata records. They are robust and scalable and can easily accommodate the volume of data that must be stored. They provide programming interfaces, transactional semantics and a full set of maintenance and performance enhancement tools. A team was organized within the HPSS project to study and recommend an approach for the replacement of SFS. Members of the team are David Fisher, Jim Minton, Donna Mecozzi, Danny Cook, Bart Parliman and Lynn Jones. We examined several possible solutions to the problem of replacing SFS, and recommended on May 22, 2000, in a report to the HPSS Technical and Executive Committees, to change HPSS into a database application over either Oracle or DB2. We recommended either Oracle or DB2 on the basis of market share and technical suitability. Oracle and DB2 are dominant offerings in the market, and it is in the best interest of HPSS to use a major player's product. Both databases provide a suitable programming interface. Transaction management functions, support for multi-threaded clients and data manipulation languages (DML) are available. These findings were supported in meetings held with technical experts from both companies. In both cases, the evidence indicated that either database would provide the features needed to host HPSS. |
author2 |
United States. Department of Energy. |
format |
Report |
author |
Fisher,D |
author_facet |
Fisher,D |
author_sort |
Fisher,D |
title |
Report on the HPSS Database Benchmark |
title_short |
Report on the HPSS Database Benchmark |
title_full |
Report on the HPSS Database Benchmark |
title_fullStr |
Report on the HPSS Database Benchmark |
title_full_unstemmed |
Report on the HPSS Database Benchmark |
title_sort |
report on the hpss database benchmark |
publisher |
Lawrence Livermore National Laboratory |
publishDate |
2001 |
url |
https://doi.org/10.2172/15006190 https://digital.library.unt.edu/ark:/67531/metadc1405330/ |
genre |
DML |
genre_facet |
DML |
op_source |
Other Information: PBD: 19 Jun 2001 |
op_relation |
rep-no: UCRL-ID-144274 grantno: W-7405-ENG-48 doi:10.2172/15006190 osti: 15006190 https://digital.library.unt.edu/ark:/67531/metadc1405330/ ark: ark:/67531/metadc1405330 |
op_doi |
https://doi.org/10.2172/15006190 |
_version_ |
1766397695078957056 |