A cost-benefit scheme for high performance predictive prefetching

Vellanki, Vivekanand; Chervenak, Ann

doi:10.1145/331532.331582

Cited by 21 publications

(16 citation statements)

References 11 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…While many prediction-based prefetching schemes [25][26][27] suffer from low hit ratio of the prefetched data, FAST can achieve near 100% hit ratio. This is because the application launch sequence changes little over repeated launches of an application.…”

Section: What To Prefetchmentioning

confidence: 99%

Improving Application Launch Performance on Solid State Drives

Joo¹,

Ryu

Park³

et al. 2012

J. Comput. Sci. Technol.

View full text Add to dashboard Cite

Application launch performance is of great importance to system platform developers and vendors as it greatly affects the degree of users' satisfaction. The single most effective way to improve application launch performance is to replace a hard disk drive (HDD) with a solid state drive (SSD), which has recently become affordable and popular. A natural question is then whether or not to replace the traditional HDD-aware application launchers with a new SSD-aware optimizer. We address this question by analyzing the inefficiency of the HDD-aware application launchers on SSDs and then proposing a new SSD-aware application prefetching scheme, called the Fast Application STarter (FAST). The key idea of FAST is to overlap the computation (CPU) time with the SSD access (I/O) time during an application launch. FAST is composed of a set of user-level components and system debugging tools provided by Linux OS (operating system). Hence, FAST can be easily deployed in any recent Linux versions without kernel recompilation. We implement FAST on a desktop PC with an SSD running Linux 2.6.32 OS and evaluate it by launching a set of widely-used applications, demonstrating an average of 28% reduction of application launch time as compared to PC without a prefetcher.

show abstract

Section: What To Prefetchmentioning

confidence: 99%

Improving Application Launch Performance on Solid State Drives

Joo¹,

Ryu

Park³

et al. 2012

J. Comput. Sci. Technol.

View full text Add to dashboard Cite

show abstract

Section: Related Workmentioning

confidence: 99%

“…Vellanki et al (1999) [13] used a similar cost-benefit analysis as Patterson et al [9] except modified the algorithm for use with predictive prefetching. This adds the quality of application independence that was lacking in the informative approach [13]. The cost-benefit analysis investigated by Vellanki and Chervenak [13] dynamically partitions an aggregate cache buffer according to weighing the costs of ejecting blocks of memory from the cache partition, versus allocating blocks to the cache partition based on probabilities of the block's future accesses.…”

Section: Related Workmentioning

confidence: 99%

“…This adds the quality of application independence that was lacking in the informative approach [13]. The cost-benefit analysis investigated by Vellanki and Chervenak [13] dynamically partitions an aggregate cache buffer according to weighing the costs of ejecting blocks of memory from the cache partition, versus allocating blocks to the cache partition based on probabilities of the block's future accesses. The APACS system presented here differs from their approach, in that APACS relies on a more general way of using cache and prefetch buffer hit ratios as an indicator for buffer/cache adjustment.…”

Section: Related Workmentioning

confidence: 99%

“…Several models for encoding past access patterns and future probabilities have been developed by researchers [5], [12], [13]. The directed weighted graph, originally proposed by Griffioen et al [5], is employed to estimate access probabilities.…”

Section: Related Workmentioning

confidence: 99%

See 2 more Smart Citations

An automatic prefetching and caching system

Lewis

Alghamdi

Assaf

et al. 2010

International Performance Computing and Communications Conference

View full text Add to dashboard Cite

Abstract-Steady improvements in storage capacities and CPU clock speeds intensify the performance bottleneck at the I/O subsystem of modern computers. Caching data can efficiently short circuit costly delays associated with disk accesses. Recent studies have shown that disk I/O performance gains provided by a cache buffer do not scale with cache size. Therefore, new algorithms have to be investigated to better utilize cache buffer space. Predictive prefetching and caching solutions have been shown to improve I/O performance in an efficient and scalable manner in simulation experiments. However, most predictive prefetching algorithms have not yet been implemented in realworld storage systems due to two main limitations: first, the existing prefetching solutions are unable to self regulate based on changing I/O workload; second, excessive number of unneeded blocks are prefetched. Combined, these drawbacks make predictive prefetching and caching a less attractive solution than the simple LRU management. To address these problems, in this paper we propose an automatic prefetching and caching system (or APACS for short), which mitigates all of these shortcomings through three unique techniques, namely: (1) dynamic cache partitioning, (2) prefetch pipelining, and (3) prefetch buffer management. APACS dynamically partitions the buffer cache memory, used for prefetched and cached blocks, by automatically changing buffer/cache sizes in accordance to global I/O performance. The adaptive partitioning scheme implemented in APACS optimizes cache hit ratios, which subsequently accelerates application execution speeds. Experimental results obtained from trace-driven simulations show that APACS outperforms the LRU cache management and existing prefetching algorithms by an average of over 50%.

show abstract

Improving reading performance by file prefetching mechanism in distributed cache systems

Gui,

Wang,

Shuai

2024

Concurrency and Computation

View full text Add to dashboard Cite

SummaryDistributed cache systems are utilized to enhance I/O performance between computing applications and storage systems. However, the traditional file access predictors employed in these cache systems are only suitable for workloads with simple file access patterns, rendering them inadequate for the complex access patterns found in big data computing scenarios. In this article, we propose a file access predictor (DFAP) based on WaveNet, which has exhibited promising results in file access tasks when compared to other baseline models. Cache systems are often constrained by limited cache space due to cost, cluster size, and other factors. In big data scenarios, cached data and prefetched data often compete for limited space. To address this issue, we introduce a cache prefetching algorithm (CBAP) for cache systems, which is based on cost‐benefit analysis to improve cache utilization. Furthermore, we implement a novel file prefetching framework on Alluxio, which accelerates computing jobs by up to 18%.

show abstract

A cost-benefit scheme for high performance predictive prefetching

Cited by 21 publications

References 11 publications

Improving Application Launch Performance on Solid State Drives

Improving Application Launch Performance on Solid State Drives

An automatic prefetching and caching system

Improving reading performance by file prefetching mechanism in distributed cache systems

Contact Info

Product

Resources

About