Adaptive multi-level cache allocation in distributed storage architectures

Ramya Prabhakar, Shekhar Srikantaiah, Mahmut Kandemir, Christina Patrick

Research output: Chapter in Book/Report/Conference proceedingConference contribution

8 Citations (Scopus)

Abstract

Increasing complexity of large-scale applications and continuous increases in data set sizes of such applications combined with slow improvements in disk access latencies has resulted in I/O becoming a performance bottleneck. While there are several ways of improving I/O access latencies of dataintensive applications, one of the promising approaches has been using different layers of the I/O subsystem to cache recently and/or frequently used data so that the number of I/O requests accessing the disk is reduced. These different layers of caches across the storage hierarchy introduce the need for efficient cache management schemes to derive maximum performance benefits. Several state-of-the-art multi-level storage cache management schemes focus on optimizing aggregate hit rate or overall I/O latency, while being agnostic to Service Level Objectives (SLOs). Also, most of the existing works focus on different cache replacement algorithms for managing storage caches and discuss different exclusive caching techniques in the context of multilevel cache hierarchy. However, the orthogonal problem of storage cache space allocation to multiple, simultaneously-running applications in a multi-level hierarchy of storage caches with multiple storage servers has remained an open research problem. In this work, using a combination of per-application latency model and a linear programming model, we proportion storage caches dynamically among multiple concurrently-executing applications across the different levels of the storage hierarchy and across multiple servers to provide isolation to applications while satisfying the application level SLOs. Further, our algorithm improves the overall system performance significantly.

Original languageEnglish (US)
Title of host publicationICS'10 - 2010 International Conference on Supercomputing
Pages211-221
Number of pages11
DOIs
StatePublished - Jul 23 2010
Event24th ACM International Conference on Supercomputing, ICS'10 - Tsukuba, Ibaraki, Japan
Duration: Jun 2 2010Jun 4 2010

Publication series

NameProceedings of the International Conference on Supercomputing

Other

Other24th ACM International Conference on Supercomputing, ICS'10
CountryJapan
CityTsukuba, Ibaraki
Period6/2/106/4/10

Fingerprint

Servers
Linear programming

All Science Journal Classification (ASJC) codes

  • Computer Science(all)

Cite this

Prabhakar, R., Srikantaiah, S., Kandemir, M., & Patrick, C. (2010). Adaptive multi-level cache allocation in distributed storage architectures. In ICS'10 - 2010 International Conference on Supercomputing (pp. 211-221). (Proceedings of the International Conference on Supercomputing). https://doi.org/10.1145/1810085.1810115
Prabhakar, Ramya ; Srikantaiah, Shekhar ; Kandemir, Mahmut ; Patrick, Christina. / Adaptive multi-level cache allocation in distributed storage architectures. ICS'10 - 2010 International Conference on Supercomputing. 2010. pp. 211-221 (Proceedings of the International Conference on Supercomputing).
@inproceedings{e070119436764828834d5e47da44ae81,
title = "Adaptive multi-level cache allocation in distributed storage architectures",
abstract = "Increasing complexity of large-scale applications and continuous increases in data set sizes of such applications combined with slow improvements in disk access latencies has resulted in I/O becoming a performance bottleneck. While there are several ways of improving I/O access latencies of dataintensive applications, one of the promising approaches has been using different layers of the I/O subsystem to cache recently and/or frequently used data so that the number of I/O requests accessing the disk is reduced. These different layers of caches across the storage hierarchy introduce the need for efficient cache management schemes to derive maximum performance benefits. Several state-of-the-art multi-level storage cache management schemes focus on optimizing aggregate hit rate or overall I/O latency, while being agnostic to Service Level Objectives (SLOs). Also, most of the existing works focus on different cache replacement algorithms for managing storage caches and discuss different exclusive caching techniques in the context of multilevel cache hierarchy. However, the orthogonal problem of storage cache space allocation to multiple, simultaneously-running applications in a multi-level hierarchy of storage caches with multiple storage servers has remained an open research problem. In this work, using a combination of per-application latency model and a linear programming model, we proportion storage caches dynamically among multiple concurrently-executing applications across the different levels of the storage hierarchy and across multiple servers to provide isolation to applications while satisfying the application level SLOs. Further, our algorithm improves the overall system performance significantly.",
author = "Ramya Prabhakar and Shekhar Srikantaiah and Mahmut Kandemir and Christina Patrick",
year = "2010",
month = "7",
day = "23",
doi = "10.1145/1810085.1810115",
language = "English (US)",
isbn = "9781450300186",
series = "Proceedings of the International Conference on Supercomputing",
pages = "211--221",
booktitle = "ICS'10 - 2010 International Conference on Supercomputing",

}

Prabhakar, R, Srikantaiah, S, Kandemir, M & Patrick, C 2010, Adaptive multi-level cache allocation in distributed storage architectures. in ICS'10 - 2010 International Conference on Supercomputing. Proceedings of the International Conference on Supercomputing, pp. 211-221, 24th ACM International Conference on Supercomputing, ICS'10, Tsukuba, Ibaraki, Japan, 6/2/10. https://doi.org/10.1145/1810085.1810115

Adaptive multi-level cache allocation in distributed storage architectures. / Prabhakar, Ramya; Srikantaiah, Shekhar; Kandemir, Mahmut; Patrick, Christina.

ICS'10 - 2010 International Conference on Supercomputing. 2010. p. 211-221 (Proceedings of the International Conference on Supercomputing).

Research output: Chapter in Book/Report/Conference proceedingConference contribution

TY - GEN

T1 - Adaptive multi-level cache allocation in distributed storage architectures

AU - Prabhakar, Ramya

AU - Srikantaiah, Shekhar

AU - Kandemir, Mahmut

AU - Patrick, Christina

PY - 2010/7/23

Y1 - 2010/7/23

N2 - Increasing complexity of large-scale applications and continuous increases in data set sizes of such applications combined with slow improvements in disk access latencies has resulted in I/O becoming a performance bottleneck. While there are several ways of improving I/O access latencies of dataintensive applications, one of the promising approaches has been using different layers of the I/O subsystem to cache recently and/or frequently used data so that the number of I/O requests accessing the disk is reduced. These different layers of caches across the storage hierarchy introduce the need for efficient cache management schemes to derive maximum performance benefits. Several state-of-the-art multi-level storage cache management schemes focus on optimizing aggregate hit rate or overall I/O latency, while being agnostic to Service Level Objectives (SLOs). Also, most of the existing works focus on different cache replacement algorithms for managing storage caches and discuss different exclusive caching techniques in the context of multilevel cache hierarchy. However, the orthogonal problem of storage cache space allocation to multiple, simultaneously-running applications in a multi-level hierarchy of storage caches with multiple storage servers has remained an open research problem. In this work, using a combination of per-application latency model and a linear programming model, we proportion storage caches dynamically among multiple concurrently-executing applications across the different levels of the storage hierarchy and across multiple servers to provide isolation to applications while satisfying the application level SLOs. Further, our algorithm improves the overall system performance significantly.

AB - Increasing complexity of large-scale applications and continuous increases in data set sizes of such applications combined with slow improvements in disk access latencies has resulted in I/O becoming a performance bottleneck. While there are several ways of improving I/O access latencies of dataintensive applications, one of the promising approaches has been using different layers of the I/O subsystem to cache recently and/or frequently used data so that the number of I/O requests accessing the disk is reduced. These different layers of caches across the storage hierarchy introduce the need for efficient cache management schemes to derive maximum performance benefits. Several state-of-the-art multi-level storage cache management schemes focus on optimizing aggregate hit rate or overall I/O latency, while being agnostic to Service Level Objectives (SLOs). Also, most of the existing works focus on different cache replacement algorithms for managing storage caches and discuss different exclusive caching techniques in the context of multilevel cache hierarchy. However, the orthogonal problem of storage cache space allocation to multiple, simultaneously-running applications in a multi-level hierarchy of storage caches with multiple storage servers has remained an open research problem. In this work, using a combination of per-application latency model and a linear programming model, we proportion storage caches dynamically among multiple concurrently-executing applications across the different levels of the storage hierarchy and across multiple servers to provide isolation to applications while satisfying the application level SLOs. Further, our algorithm improves the overall system performance significantly.

UR - http://www.scopus.com/inward/record.url?scp=77954738649&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=77954738649&partnerID=8YFLogxK

U2 - 10.1145/1810085.1810115

DO - 10.1145/1810085.1810115

M3 - Conference contribution

AN - SCOPUS:77954738649

SN - 9781450300186

T3 - Proceedings of the International Conference on Supercomputing

SP - 211

EP - 221

BT - ICS'10 - 2010 International Conference on Supercomputing

ER -

Prabhakar R, Srikantaiah S, Kandemir M, Patrick C. Adaptive multi-level cache allocation in distributed storage architectures. In ICS'10 - 2010 International Conference on Supercomputing. 2010. p. 211-221. (Proceedings of the International Conference on Supercomputing). https://doi.org/10.1145/1810085.1810115