Dynamic storage cache allocation in multi-server architectures

Ramya Prabhakar, Shekhar Srikantaiah, Christina Patrick, Mahmut Kandemir

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

5 Citations (Scopus)

Abstract

We introduce a dynamic and efficient shared cache management scheme, called Maxperf, that manages the aggregate cache space in multi-server storage architectures such that the service level objectives (SLOs) of concurrently executing applications are satisfied and any spare cache capacity is proportionately allocated according to the marginal gains of the applications to maximize performance. We use a combination of Neville's algorithm and linear-programming-model to discover the required storage cache partition size, on each server, for every application accessing that server. Experimental results show that our algorithm enforces partitions to provide stronger isolation to applications, meets application level SLOs even in the presence of dynamically changing storage cache requirements, and improves I/O latency of individual applications as well as the overall I/O latency significantly compared to two alternate storage cache management schemes, and a state-of-the-art single server storage cache management scheme extended to multi-server architecture.

Original languageEnglish (US)
Title of host publicationProceedings of the Conference on High Performance Computing Networking, Storage and Analysis, SC '09
DOIs
StatePublished - Dec 1 2009
EventConference on High Performance Computing Networking, Storage and Analysis, SC '09 - Portland, OR, United States
Duration: Nov 14 2009Nov 20 2009

Publication series

NameProceedings of the Conference on High Performance Computing Networking, Storage and Analysis, SC '09

Other

OtherConference on High Performance Computing Networking, Storage and Analysis, SC '09
CountryUnited States
CityPortland, OR
Period11/14/0911/20/09

Fingerprint

Servers
Linear programming

All Science Journal Classification (ASJC) codes

  • Computer Networks and Communications
  • Computer Science Applications

Cite this

Prabhakar, R., Srikantaiah, S., Patrick, C., & Kandemir, M. (2009). Dynamic storage cache allocation in multi-server architectures. In Proceedings of the Conference on High Performance Computing Networking, Storage and Analysis, SC '09 [1654068] (Proceedings of the Conference on High Performance Computing Networking, Storage and Analysis, SC '09). https://doi.org/10.1145/1654059.1654068
Prabhakar, Ramya ; Srikantaiah, Shekhar ; Patrick, Christina ; Kandemir, Mahmut. / Dynamic storage cache allocation in multi-server architectures. Proceedings of the Conference on High Performance Computing Networking, Storage and Analysis, SC '09. 2009. (Proceedings of the Conference on High Performance Computing Networking, Storage and Analysis, SC '09).
@inproceedings{4c2e0ae837d5483797fd93c9a65a9d13,
title = "Dynamic storage cache allocation in multi-server architectures",
abstract = "We introduce a dynamic and efficient shared cache management scheme, called Maxperf, that manages the aggregate cache space in multi-server storage architectures such that the service level objectives (SLOs) of concurrently executing applications are satisfied and any spare cache capacity is proportionately allocated according to the marginal gains of the applications to maximize performance. We use a combination of Neville's algorithm and linear-programming-model to discover the required storage cache partition size, on each server, for every application accessing that server. Experimental results show that our algorithm enforces partitions to provide stronger isolation to applications, meets application level SLOs even in the presence of dynamically changing storage cache requirements, and improves I/O latency of individual applications as well as the overall I/O latency significantly compared to two alternate storage cache management schemes, and a state-of-the-art single server storage cache management scheme extended to multi-server architecture.",
author = "Ramya Prabhakar and Shekhar Srikantaiah and Christina Patrick and Mahmut Kandemir",
year = "2009",
month = "12",
day = "1",
doi = "10.1145/1654059.1654068",
language = "English (US)",
isbn = "9781605587448",
series = "Proceedings of the Conference on High Performance Computing Networking, Storage and Analysis, SC '09",
booktitle = "Proceedings of the Conference on High Performance Computing Networking, Storage and Analysis, SC '09",

}

Prabhakar, R, Srikantaiah, S, Patrick, C & Kandemir, M 2009, Dynamic storage cache allocation in multi-server architectures. in Proceedings of the Conference on High Performance Computing Networking, Storage and Analysis, SC '09., 1654068, Proceedings of the Conference on High Performance Computing Networking, Storage and Analysis, SC '09, Conference on High Performance Computing Networking, Storage and Analysis, SC '09, Portland, OR, United States, 11/14/09. https://doi.org/10.1145/1654059.1654068

Dynamic storage cache allocation in multi-server architectures. / Prabhakar, Ramya; Srikantaiah, Shekhar; Patrick, Christina; Kandemir, Mahmut.

Proceedings of the Conference on High Performance Computing Networking, Storage and Analysis, SC '09. 2009. 1654068 (Proceedings of the Conference on High Performance Computing Networking, Storage and Analysis, SC '09).

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

TY - GEN

T1 - Dynamic storage cache allocation in multi-server architectures

AU - Prabhakar, Ramya

AU - Srikantaiah, Shekhar

AU - Patrick, Christina

AU - Kandemir, Mahmut

PY - 2009/12/1

Y1 - 2009/12/1

N2 - We introduce a dynamic and efficient shared cache management scheme, called Maxperf, that manages the aggregate cache space in multi-server storage architectures such that the service level objectives (SLOs) of concurrently executing applications are satisfied and any spare cache capacity is proportionately allocated according to the marginal gains of the applications to maximize performance. We use a combination of Neville's algorithm and linear-programming-model to discover the required storage cache partition size, on each server, for every application accessing that server. Experimental results show that our algorithm enforces partitions to provide stronger isolation to applications, meets application level SLOs even in the presence of dynamically changing storage cache requirements, and improves I/O latency of individual applications as well as the overall I/O latency significantly compared to two alternate storage cache management schemes, and a state-of-the-art single server storage cache management scheme extended to multi-server architecture.

AB - We introduce a dynamic and efficient shared cache management scheme, called Maxperf, that manages the aggregate cache space in multi-server storage architectures such that the service level objectives (SLOs) of concurrently executing applications are satisfied and any spare cache capacity is proportionately allocated according to the marginal gains of the applications to maximize performance. We use a combination of Neville's algorithm and linear-programming-model to discover the required storage cache partition size, on each server, for every application accessing that server. Experimental results show that our algorithm enforces partitions to provide stronger isolation to applications, meets application level SLOs even in the presence of dynamically changing storage cache requirements, and improves I/O latency of individual applications as well as the overall I/O latency significantly compared to two alternate storage cache management schemes, and a state-of-the-art single server storage cache management scheme extended to multi-server architecture.

UR - http://www.scopus.com/inward/record.url?scp=74049120372&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=74049120372&partnerID=8YFLogxK

U2 - 10.1145/1654059.1654068

DO - 10.1145/1654059.1654068

M3 - Conference contribution

SN - 9781605587448

T3 - Proceedings of the Conference on High Performance Computing Networking, Storage and Analysis, SC '09

BT - Proceedings of the Conference on High Performance Computing Networking, Storage and Analysis, SC '09

ER -

Prabhakar R, Srikantaiah S, Patrick C, Kandemir M. Dynamic storage cache allocation in multi-server architectures. In Proceedings of the Conference on High Performance Computing Networking, Storage and Analysis, SC '09. 2009. 1654068. (Proceedings of the Conference on High Performance Computing Networking, Storage and Analysis, SC '09). https://doi.org/10.1145/1654059.1654068