VCacheshare: Automated server flash cache space management in a virtualization environment

Fei Meng, Li Zhou, Xiaosong Ma, Sandeep Uttamchandani, Deng Liu

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

23 Citations (Scopus)

Abstract

Server Flash Cache (SFC) is increasingly adopted in virtualization environments for IO acceleration. Deciding the optimal SFC allocation among VMs or VM disks is a major pain-point, dominantly handled manually by administrators. In this paper, we present vCacheShare, a dynamic, workload-aware, policy-driven framework for continuous and automated optimization of SFC space partitioning. Its decision-making is based on multiple IO access characteristics. In particular, vCacheShare adopts a cache utility model that captures both longer-term locality behavior and transient locality spikes. This paper validates the growing applicability of analytical programming techniques to solve real-time resource management problems, traditionally addressed using heuristics. We designed vCacheShare to coordinate with typical VM mobility events and implemented it within the widely used ESXi hypervisor. We performed extensive evaluation using 13 representative enterprise IO workloads, one IO benchmark, and two end-to-end deployment test cases targeting Virtual Desktop Infrastructure (VDI) and data warehousing scenarios respectively. Our results verified the advantage of vCacheShare over implicit management schemes such as global LRU, and confirmed its self-adaptive capability.

Original languageEnglish
Title of host publicationProceedings of the 2014 USENIX Annual Technical Conference, USENIX ATC 2014
PublisherUSENIX Association
Pages133-144
Number of pages12
ISBN (Electronic)9781931971102
Publication statusPublished - 1 Jan 2019
Event2014 USENIX Annual Technical Conference, USENIX ATC 2014 - Philadelphia, United States
Duration: 19 Jun 201420 Jun 2014

Publication series

NameProceedings of the 2014 USENIX Annual Technical Conference, USENIX ATC 2014

Conference

Conference2014 USENIX Annual Technical Conference, USENIX ATC 2014
CountryUnited States
CityPhiladelphia
Period19/6/1420/6/14

Fingerprint

Servers
Data warehouses
Decision making
Virtualization
Industry

ASJC Scopus subject areas

  • Computer Science(all)

Cite this

Meng, F., Zhou, L., Ma, X., Uttamchandani, S., & Liu, D. (2019). VCacheshare: Automated server flash cache space management in a virtualization environment. In Proceedings of the 2014 USENIX Annual Technical Conference, USENIX ATC 2014 (pp. 133-144). (Proceedings of the 2014 USENIX Annual Technical Conference, USENIX ATC 2014). USENIX Association.

VCacheshare : Automated server flash cache space management in a virtualization environment. / Meng, Fei; Zhou, Li; Ma, Xiaosong; Uttamchandani, Sandeep; Liu, Deng.

Proceedings of the 2014 USENIX Annual Technical Conference, USENIX ATC 2014. USENIX Association, 2019. p. 133-144 (Proceedings of the 2014 USENIX Annual Technical Conference, USENIX ATC 2014).

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

Meng, F, Zhou, L, Ma, X, Uttamchandani, S & Liu, D 2019, VCacheshare: Automated server flash cache space management in a virtualization environment. in Proceedings of the 2014 USENIX Annual Technical Conference, USENIX ATC 2014. Proceedings of the 2014 USENIX Annual Technical Conference, USENIX ATC 2014, USENIX Association, pp. 133-144, 2014 USENIX Annual Technical Conference, USENIX ATC 2014, Philadelphia, United States, 19/6/14.
Meng F, Zhou L, Ma X, Uttamchandani S, Liu D. VCacheshare: Automated server flash cache space management in a virtualization environment. In Proceedings of the 2014 USENIX Annual Technical Conference, USENIX ATC 2014. USENIX Association. 2019. p. 133-144. (Proceedings of the 2014 USENIX Annual Technical Conference, USENIX ATC 2014).
Meng, Fei ; Zhou, Li ; Ma, Xiaosong ; Uttamchandani, Sandeep ; Liu, Deng. / VCacheshare : Automated server flash cache space management in a virtualization environment. Proceedings of the 2014 USENIX Annual Technical Conference, USENIX ATC 2014. USENIX Association, 2019. pp. 133-144 (Proceedings of the 2014 USENIX Annual Technical Conference, USENIX ATC 2014).
@inproceedings{e4e0ca7ac3424577b7d0c810e3be78c4,
title = "VCacheshare: Automated server flash cache space management in a virtualization environment",
abstract = "Server Flash Cache (SFC) is increasingly adopted in virtualization environments for IO acceleration. Deciding the optimal SFC allocation among VMs or VM disks is a major pain-point, dominantly handled manually by administrators. In this paper, we present vCacheShare, a dynamic, workload-aware, policy-driven framework for continuous and automated optimization of SFC space partitioning. Its decision-making is based on multiple IO access characteristics. In particular, vCacheShare adopts a cache utility model that captures both longer-term locality behavior and transient locality spikes. This paper validates the growing applicability of analytical programming techniques to solve real-time resource management problems, traditionally addressed using heuristics. We designed vCacheShare to coordinate with typical VM mobility events and implemented it within the widely used ESXi hypervisor. We performed extensive evaluation using 13 representative enterprise IO workloads, one IO benchmark, and two end-to-end deployment test cases targeting Virtual Desktop Infrastructure (VDI) and data warehousing scenarios respectively. Our results verified the advantage of vCacheShare over implicit management schemes such as global LRU, and confirmed its self-adaptive capability.",
author = "Fei Meng and Li Zhou and Xiaosong Ma and Sandeep Uttamchandani and Deng Liu",
year = "2019",
month = "1",
day = "1",
language = "English",
series = "Proceedings of the 2014 USENIX Annual Technical Conference, USENIX ATC 2014",
publisher = "USENIX Association",
pages = "133--144",
booktitle = "Proceedings of the 2014 USENIX Annual Technical Conference, USENIX ATC 2014",

}

TY - GEN

T1 - VCacheshare

T2 - Automated server flash cache space management in a virtualization environment

AU - Meng, Fei

AU - Zhou, Li

AU - Ma, Xiaosong

AU - Uttamchandani, Sandeep

AU - Liu, Deng

PY - 2019/1/1

Y1 - 2019/1/1

N2 - Server Flash Cache (SFC) is increasingly adopted in virtualization environments for IO acceleration. Deciding the optimal SFC allocation among VMs or VM disks is a major pain-point, dominantly handled manually by administrators. In this paper, we present vCacheShare, a dynamic, workload-aware, policy-driven framework for continuous and automated optimization of SFC space partitioning. Its decision-making is based on multiple IO access characteristics. In particular, vCacheShare adopts a cache utility model that captures both longer-term locality behavior and transient locality spikes. This paper validates the growing applicability of analytical programming techniques to solve real-time resource management problems, traditionally addressed using heuristics. We designed vCacheShare to coordinate with typical VM mobility events and implemented it within the widely used ESXi hypervisor. We performed extensive evaluation using 13 representative enterprise IO workloads, one IO benchmark, and two end-to-end deployment test cases targeting Virtual Desktop Infrastructure (VDI) and data warehousing scenarios respectively. Our results verified the advantage of vCacheShare over implicit management schemes such as global LRU, and confirmed its self-adaptive capability.

AB - Server Flash Cache (SFC) is increasingly adopted in virtualization environments for IO acceleration. Deciding the optimal SFC allocation among VMs or VM disks is a major pain-point, dominantly handled manually by administrators. In this paper, we present vCacheShare, a dynamic, workload-aware, policy-driven framework for continuous and automated optimization of SFC space partitioning. Its decision-making is based on multiple IO access characteristics. In particular, vCacheShare adopts a cache utility model that captures both longer-term locality behavior and transient locality spikes. This paper validates the growing applicability of analytical programming techniques to solve real-time resource management problems, traditionally addressed using heuristics. We designed vCacheShare to coordinate with typical VM mobility events and implemented it within the widely used ESXi hypervisor. We performed extensive evaluation using 13 representative enterprise IO workloads, one IO benchmark, and two end-to-end deployment test cases targeting Virtual Desktop Infrastructure (VDI) and data warehousing scenarios respectively. Our results verified the advantage of vCacheShare over implicit management schemes such as global LRU, and confirmed its self-adaptive capability.

UR - http://www.scopus.com/inward/record.url?scp=85077451961&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85077451961&partnerID=8YFLogxK

M3 - Conference contribution

AN - SCOPUS:85077451961

T3 - Proceedings of the 2014 USENIX Annual Technical Conference, USENIX ATC 2014

SP - 133

EP - 144

BT - Proceedings of the 2014 USENIX Annual Technical Conference, USENIX ATC 2014

PB - USENIX Association

ER -