Managing response time tails by sharding
- Submitting institution
-
Imperial College of Science, Technology and Medicine
- Unit of assessment
- 11 - Computer Science and Informatics
- Output identifier
- 2416
- Type
- D - Journal article
- DOI
-
10.1145/3300143
- Title of journal
- ACM transactions on modeling and performance evaluation of computing systems
- Article number
- 5
- First page
- 1
- Volume
- 4
- Issue
- 1
- ISSN
- 2376-3647
- Open access status
- Compliant
- Month of publication
- March
- Year of publication
- 2019
- URL
-
-
- Supplementary information
-
10.1145/3300143
- Request cross-referral to
- -
- Output has been delayed by COVID-19
- No
- COVID-19 affected output statement
- -
- Forensic science
- No
- Criminology
- No
- Interdisciplinary
- No
- Number of additional authors
-
3
- Research group(s)
-
-
- Citation count
- 0
- Proposed double-weighted
- No
- Reserve for an output with double weighting
- No
- Additional information
- This work determines the performance advantage gained from sharing in striped storage systems in the case of phase-type access times for any number of replicas under two different deletion strategies for expired replicas. For the first time, our analysis yields an approximation that gives arbitrary accuracy. Our techniques are now used in NetApp products via co-author Patel (https://www.netapp.com/us/products/ontap-ai.aspx) and are equally applicable in parallel web searches. The work resulted in a keynote at MASCOTS’16 (https://san.ee.ic.ac.uk/mascots2016/) and led to Qiu’s employment by Google.
- Author contribution statement
- -
- Non-English
- No
- English abstract
- -
