Challenges and Solutions for Tracing Storage Systems: A Case Study with Spectrum Scale
- Publikationstyp:
- Zeitschriftenaufsatz
- Metadaten:
-
- Autoren
- Marc-Andre Vef
- Vasily Tarasov
- Dean Hildebrand
- Andre Brinkmann
- Autoren-URL
- https://www.webofscience.com/api/gateway?GWVersion=2&SrcApp=fis-test-1&SrcAuth=WosAPI&KeyUT=WOS:000434635800007&DestLinkType=FullRecord&DestApp=WOS_CPL
- DOI
- 10.1145/3149376
- eISSN
- 1553-3093
- Externe Identifier
- Clarivate Analytics Document Solution ID: GI6TM
- ISSN
- 1553-3077
- Ausgabe der Veröffentlichung
- 2
- Zeitschrift
- ACM TRANSACTIONS ON STORAGE
- Schlüsselwörter
- Parallel file system
- GPFS
- trace analysis
- performance
- Artikelnummer
- ARTN 18
- Datum der Veröffentlichung
- 2018
- Status
- Published
- Titel
- Challenges and Solutions for Tracing Storage Systems: A Case Study with Spectrum Scale
- Sub types
- Article
- Ausgabe der Zeitschrift
- 14
Datenquelle: Web of Science (Lite)
- Andere Metadatenquellen:
-
- Abstract
- <jats:p>IBM Spectrum Scale’s parallel file system General Parallel File System (GPFS) has a 20-year development history with over 100 contributing developers. Its ability to support strict POSIX semantics across more than 10K clients leads to a complex design with intricate interactions between the cluster nodes. Tracing has proven to be a vital tool to understand the behavior and the anomalies of such a complex software product. However, the necessary trace information is often buried in hundreds of gigabytes of by-product trace records. Further, the overhead of tracing can significantly impact running applications and file system performance, limiting the use of tracing in a production system.</jats:p> <jats:p> In this research article, we discuss the evolution of the mature and highly scalable GPFS tracing tool and present the exploratory study of GPFS’ new tracing interface, <jats:italic>FlexTrace</jats:italic> , which allows developers and users to accurately specify what to trace for the problem they are trying to solve. We evaluate our methodology and prototype, demonstrating that the proposed approach has negligible overhead, even under intensive I/O workloads and with low-latency storage devices. </jats:p>
- Autoren
- Marc-André Vef
- Vasily Tarasov
- Dean Hildebrand
- André Brinkmann
- DOI
- 10.1145/3149376
- eISSN
- 1553-3093
- ISSN
- 1553-3077
- Ausgabe der Veröffentlichung
- 2
- Zeitschrift
- ACM Transactions on Storage
- Sprache
- en
- Online publication date
- 2018
- Paginierung
- 1 - 24
- Datum der Veröffentlichung
- 2018
- Status
- Published
- Herausgeber
- Association for Computing Machinery (ACM)
- Herausgeber URL
- http://dx.doi.org/10.1145/3149376
- Datum der Datenerfassung
- 2022
- Titel
- Challenges and Solutions for Tracing Storage Systems
- Ausgabe der Zeitschrift
- 14
Datenquelle: Crossref
- Autoren
- Marc-Andre Vef
- Vasily Tarasov
- Dean Hildebrand
- André Brinkmann
- Zeitschrift
- ACM Trans. Storage
- Artikelnummer
- 2
- Paginierung
- 18:1 - 18:1
- Datum der Veröffentlichung
- 2018
- Titel
- Challenges and Solutions for Tracing Storage Systems: A Case Study with Spectrum Scale.
- Ausgabe der Zeitschrift
- 14
Datenquelle: DBLP
- Abstract
- IBM Spectrum Scale’s parallel file system General Parallel File System (GPFS) has a 20-year development history with over 100 contributing developers. Its ability to support strict POSIX semantics across more than 10K clients leads to a complex design with intricate interactions between the cluster nodes. Tracing has proven to be a vital tool to understand the behavior and the anomalies of such a complex software product. However, the necessary trace information is often buried in hundreds of gigabytes of by-product trace records. Further, the overhead of tracing can significantly impact running applications and file system performance, limiting the use of tracing in a production system. In this research article, we discuss the evolution of the mature and highly scalable GPFS tracing tool and present the exploratory study of GPFS’ new tracing interface, FlexTrace, which allows developers and users to accurately specify what to trace for the problem they are trying to solve. We evaluate our methodology and prototype, demonstrating that the proposed approach has negligible overhead, even under intensive I/O workloads and with low-latency storage devices.
- Autoren
- Marc-André Vef
- Vasily Tarasov
- Dean Hildebrand
- André Brinkmann
- Autoren-URL
- https://seafile.rlp.net/f/619df4b4b27e44318ead/
- DOI
- 10.1145/3149376
- Zeitschrift
- ACM Transactions on Storage (ToS)
- Artikelnummer
- 2
- Paginierung
- 18:1 - 18:24
- Datum der Veröffentlichung
- 2018
- Datum der Datenerfassung
- 2020
- Titel
- Challenges and Solutions for Tracing Storage Systems: A Case Study with Spectrum Scale
- Ausgabe der Zeitschrift
- 14
Datenquelle: Manual
- Beziehungen:
- Eigentum von