PFC

Transparent optimization of existing prefetching strategies for multi-level storage systems

Zhe Zhang, Kyuhyung Lee, Xiaosong Ma, Yuanyuan Zhou

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

16 Citations (Scopus)

Abstract

The multi-level storage architecture has been widely adopted in servers and data centers. However, while prefetching has been shown as a crucial technique to exploit the sequentiality in accesses common for such systems and hide the increasing relative cost of disk I/O, existing multi-level storage studies have focused mostly on cache replacement strategies. In this paper, we show that prefetching algorithms designed for single-level systems may have their limitations magnified when applied to multi-level systems. Overly conservative prefetching will not be able to effectively use the lower-level cache space, while overly aggressive prefetching will be compounded across levels and generate large amounts of wasted prefetch. We take an innovative approach to this problem: rather than designing a new, multi-level prefetching algorithm, we developed PreFetching-Coordinator (PFC), a hierarchy-aware optimization applicable to any existing prefetching algorithms. PFC does not require any application hints, a priori knowledge on the application access pattern or the native prefetching algorithm, or modification to the I/O interface. Instead, it monitors the upper-level access patterns as well as the lower-level cache status, and dynamically adjusts the aggressiveness of the lower-level prefetching activities. We evaluated PFC with extensive simulation study using a verified multi-level storage simulator, an accurate disk simulator, and access traces with different access patterns. Our results indicate that PFC dynamically controls lowerlevel prefetching in reaction to multiple system and workload parameters, improving the overall system performance in all 96 test cases. Working with four well-known existing prefetching algorithms adopted in real systems, PFC obtains an improvement of up to 35% to the average request response time, with an average improvement of 14.6% over

Original languageEnglish
Title of host publicationProceedings - The 28th International Conference on Distributed Computing Systems, ICDCS 2008
Pages740-751
Number of pages12
DOIs
Publication statusPublished - 22 Sep 2008
Externally publishedYes
Event28th International Conference on Distributed Computing Systems, ICDCS 2008 - Beijing, China
Duration: 17 Jul 200820 Jul 2008

Other

Other28th International Conference on Distributed Computing Systems, ICDCS 2008
CountryChina
CityBeijing
Period17/7/0820/7/08

Fingerprint

Simulators
Servers
Costs

ASJC Scopus subject areas

  • Hardware and Architecture
  • Software

Cite this

Zhang, Z., Lee, K., Ma, X., & Zhou, Y. (2008). PFC: Transparent optimization of existing prefetching strategies for multi-level storage systems. In Proceedings - The 28th International Conference on Distributed Computing Systems, ICDCS 2008 (pp. 740-751). [4595949] https://doi.org/10.1109/ICDCS.2008.89

PFC : Transparent optimization of existing prefetching strategies for multi-level storage systems. / Zhang, Zhe; Lee, Kyuhyung; Ma, Xiaosong; Zhou, Yuanyuan.

Proceedings - The 28th International Conference on Distributed Computing Systems, ICDCS 2008. 2008. p. 740-751 4595949.

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

Zhang, Z, Lee, K, Ma, X & Zhou, Y 2008, PFC: Transparent optimization of existing prefetching strategies for multi-level storage systems. in Proceedings - The 28th International Conference on Distributed Computing Systems, ICDCS 2008., 4595949, pp. 740-751, 28th International Conference on Distributed Computing Systems, ICDCS 2008, Beijing, China, 17/7/08. https://doi.org/10.1109/ICDCS.2008.89
Zhang Z, Lee K, Ma X, Zhou Y. PFC: Transparent optimization of existing prefetching strategies for multi-level storage systems. In Proceedings - The 28th International Conference on Distributed Computing Systems, ICDCS 2008. 2008. p. 740-751. 4595949 https://doi.org/10.1109/ICDCS.2008.89
Zhang, Zhe ; Lee, Kyuhyung ; Ma, Xiaosong ; Zhou, Yuanyuan. / PFC : Transparent optimization of existing prefetching strategies for multi-level storage systems. Proceedings - The 28th International Conference on Distributed Computing Systems, ICDCS 2008. 2008. pp. 740-751
@inproceedings{3f67d76b18df4691ab05d570264b0d92,
title = "PFC: Transparent optimization of existing prefetching strategies for multi-level storage systems",
abstract = "The multi-level storage architecture has been widely adopted in servers and data centers. However, while prefetching has been shown as a crucial technique to exploit the sequentiality in accesses common for such systems and hide the increasing relative cost of disk I/O, existing multi-level storage studies have focused mostly on cache replacement strategies. In this paper, we show that prefetching algorithms designed for single-level systems may have their limitations magnified when applied to multi-level systems. Overly conservative prefetching will not be able to effectively use the lower-level cache space, while overly aggressive prefetching will be compounded across levels and generate large amounts of wasted prefetch. We take an innovative approach to this problem: rather than designing a new, multi-level prefetching algorithm, we developed PreFetching-Coordinator (PFC), a hierarchy-aware optimization applicable to any existing prefetching algorithms. PFC does not require any application hints, a priori knowledge on the application access pattern or the native prefetching algorithm, or modification to the I/O interface. Instead, it monitors the upper-level access patterns as well as the lower-level cache status, and dynamically adjusts the aggressiveness of the lower-level prefetching activities. We evaluated PFC with extensive simulation study using a verified multi-level storage simulator, an accurate disk simulator, and access traces with different access patterns. Our results indicate that PFC dynamically controls lowerlevel prefetching in reaction to multiple system and workload parameters, improving the overall system performance in all 96 test cases. Working with four well-known existing prefetching algorithms adopted in real systems, PFC obtains an improvement of up to 35{\%} to the average request response time, with an average improvement of 14.6{\%} over",
author = "Zhe Zhang and Kyuhyung Lee and Xiaosong Ma and Yuanyuan Zhou",
year = "2008",
month = "9",
day = "22",
doi = "10.1109/ICDCS.2008.89",
language = "English",
isbn = "9780769531724",
pages = "740--751",
booktitle = "Proceedings - The 28th International Conference on Distributed Computing Systems, ICDCS 2008",

}

TY - GEN

T1 - PFC

T2 - Transparent optimization of existing prefetching strategies for multi-level storage systems

AU - Zhang, Zhe

AU - Lee, Kyuhyung

AU - Ma, Xiaosong

AU - Zhou, Yuanyuan

PY - 2008/9/22

Y1 - 2008/9/22

N2 - The multi-level storage architecture has been widely adopted in servers and data centers. However, while prefetching has been shown as a crucial technique to exploit the sequentiality in accesses common for such systems and hide the increasing relative cost of disk I/O, existing multi-level storage studies have focused mostly on cache replacement strategies. In this paper, we show that prefetching algorithms designed for single-level systems may have their limitations magnified when applied to multi-level systems. Overly conservative prefetching will not be able to effectively use the lower-level cache space, while overly aggressive prefetching will be compounded across levels and generate large amounts of wasted prefetch. We take an innovative approach to this problem: rather than designing a new, multi-level prefetching algorithm, we developed PreFetching-Coordinator (PFC), a hierarchy-aware optimization applicable to any existing prefetching algorithms. PFC does not require any application hints, a priori knowledge on the application access pattern or the native prefetching algorithm, or modification to the I/O interface. Instead, it monitors the upper-level access patterns as well as the lower-level cache status, and dynamically adjusts the aggressiveness of the lower-level prefetching activities. We evaluated PFC with extensive simulation study using a verified multi-level storage simulator, an accurate disk simulator, and access traces with different access patterns. Our results indicate that PFC dynamically controls lowerlevel prefetching in reaction to multiple system and workload parameters, improving the overall system performance in all 96 test cases. Working with four well-known existing prefetching algorithms adopted in real systems, PFC obtains an improvement of up to 35% to the average request response time, with an average improvement of 14.6% over

AB - The multi-level storage architecture has been widely adopted in servers and data centers. However, while prefetching has been shown as a crucial technique to exploit the sequentiality in accesses common for such systems and hide the increasing relative cost of disk I/O, existing multi-level storage studies have focused mostly on cache replacement strategies. In this paper, we show that prefetching algorithms designed for single-level systems may have their limitations magnified when applied to multi-level systems. Overly conservative prefetching will not be able to effectively use the lower-level cache space, while overly aggressive prefetching will be compounded across levels and generate large amounts of wasted prefetch. We take an innovative approach to this problem: rather than designing a new, multi-level prefetching algorithm, we developed PreFetching-Coordinator (PFC), a hierarchy-aware optimization applicable to any existing prefetching algorithms. PFC does not require any application hints, a priori knowledge on the application access pattern or the native prefetching algorithm, or modification to the I/O interface. Instead, it monitors the upper-level access patterns as well as the lower-level cache status, and dynamically adjusts the aggressiveness of the lower-level prefetching activities. We evaluated PFC with extensive simulation study using a verified multi-level storage simulator, an accurate disk simulator, and access traces with different access patterns. Our results indicate that PFC dynamically controls lowerlevel prefetching in reaction to multiple system and workload parameters, improving the overall system performance in all 96 test cases. Working with four well-known existing prefetching algorithms adopted in real systems, PFC obtains an improvement of up to 35% to the average request response time, with an average improvement of 14.6% over

UR - http://www.scopus.com/inward/record.url?scp=51849090041&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=51849090041&partnerID=8YFLogxK

U2 - 10.1109/ICDCS.2008.89

DO - 10.1109/ICDCS.2008.89

M3 - Conference contribution

SN - 9780769531724

SP - 740

EP - 751

BT - Proceedings - The 28th International Conference on Distributed Computing Systems, ICDCS 2008

ER -