The Performance Impact of Kernel Prefetching on Buffer Cache Replacement Algorithms

Butt, Ali R.; Gniady, Chris; Hu, Y. Charlie

doi:10.1109/tc.2007.1029

Cited by 25 publications

(10 citation statements)

References 30 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The behavior of the different cache replacement algorithms in viewperf (the results for viewperf can be found in [6]) is similar to that observed in cscope. As viewperf accesses large files in small chunks, it is able to see maximum benefit from prefetching.…”

Section: Viewperfsupporting

confidence: 62%

“…In gcc (the results for gcc can be found in [6]), the benefit from prefetching is not as pronounced as in cscope and glimpse. This is because in gcc many accesses are to small files, for which there is little opportunity for prefetching.…”

Section: Gccmentioning

confidence: 99%

“…Multi3 has a large number of random accesses due to tpc-h, and therefore its performance curves look similar to those of tpc-h. The detailed results for these applications can be found in [6].…”

Section: Concurrent Applicationsmentioning

confidence: 99%

See 2 more Smart Citations

The performance impact of kernel prefetching on buffer cache replacement algorithms

Butt

Gniady

2005

Proceedings of the 2005 ACM SIGMETRICS International Conference on Measurement and Modeling of Computer Systems

Self Cite

View full text Add to dashboard Cite

A fundamental challenge in improving the file system performance is to design effective block replacement algorithms to minimize buffer cache misses. Despite the well-known interactions between prefetching and caching, almost all buffer cache replacement algorithms have been proposed and studied comparatively without taking into account file system prefetching which exists in all modern operating systems. This paper shows that such kernel prefetching can have a significant impact on the relative performance in terms of the number of actual disk I/Os of many well-known replacement algorithms; it can not only narrow the performance gap but also change the relative performance benefits of different algorithms. These results demonstrate the importance for buffer caching research to take file system prefetching into consideration.

show abstract

Section: Viewperfsupporting

confidence: 62%

Section: Gccmentioning

confidence: 99%

See 1 more Smart Citation

The performance impact of kernel prefetching on buffer cache replacement algorithms

Butt

Gniady

2005

Proceedings of the 2005 ACM SIGMETRICS International Conference on Measurement and Modeling of Computer Systems

Self Cite

View full text Add to dashboard Cite

show abstract

“…These traces are considered typical and representative of applications in that most of them are routinely used in other caching algorithm studies. For example, the cscope, glimpse and gcc traces are used in [5], [7], [8], [17], [34], the gnuplot in [6], and tpch and tpcr in [35]. Table I summarizes the characteristics of these traces and more detailed description of each trace is presented below.…”

Section: Application Traces Used In the Simulation Studymentioning

confidence: 99%

“…Ref. [35] suggests that disk I/O prefetching should be disabled in both tpch and tpcr to prevent cache pollution. Thus we only use the traces without prefetching.…”

Section: ) Gcc Is a Gnu C Compiler Trace And It Compiles And Buildsmentioning

confidence: 99%

RACE: A Robust Adaptive Caching Strategy for Buffer Cache

Zhu

Jiang

2008

IEEE Trans. Comput.

View full text Add to dashboard Cite

While many block replacement algorithms for buffer caches have been proposed to address the well-known drawbacks of the LRU algorithm, they are not robust and cannot maintain an consistent performance improvement over all workloads. This paper proposes a novel and simple replacement scheme, called RACE (Robust Adaptive buffer Cache management schemE), which differentiates the locality of I/O streams by actively detecting access patterns inherently exhibited in two correlated spaces: the discrete block space of program contexts from which I/O requests are issued and the continuous block space within files to which I/O requests are addressed. This scheme combines global I/O regularities of an application and local I/O regularities of individual files accessed in that application to accurately estimate the locality strength, which is crucial in deciding which blocks are to be replaced upon a cache miss. Through comprehensive simulations on eight realapplication traces, RACE is shown to high hit ratios than LRU and all other state-of-the-art cache management schemes studied in this paper.

show abstract

PARC: A novel OS cache manager

Chang

Chiang

2018

Softw Pract Exp

View full text Add to dashboard Cite

To boost input-output performance, operating systems employ a kernelmanaged caching space called the buffer cache or page cache. Given the limited size of a buffer cache, an effective cache manager is required to decide which blocks should be evicted from the cache. Previous cache managers use historical information to make replacement decisions. However, existing approaches are unable to maximize performance since they rely on limited historical information. Motivated by the limitations of existing solutions, this paper proposes a novel manager called the Pattern-assisted Adaptive Recency Caching (PARC) manager. PARC simultaneously uses the historical information of recency, frequency, and access patterns to estimate the locality strengths of blocks and, upon a cache miss, evicts the block with the least strength. Specifically, PARC exploits the reference regularities exhibited in past input-output behaviors to actively and rapidly adapt the recency and frequency information of blocks so as to precisely distinguish blocks with long-and short-term utility. Through comprehensive simulations on a variety of traces of different access patterns, we show that PARC is robust since, except for random workloads where the performance of each cache manager is similar, PARC always outperforms existing solutions. KEYWORDSbuffer cache, cache manager, operating systems, page cache, replacement algorithms Softw Pract Exper. 2018;48:2193-2222. wileyonlinelibrary.com/journal/spe

show abstract

The Performance Impact of Kernel Prefetching on Buffer Cache Replacement Algorithms

Cited by 25 publications

References 30 publications

The performance impact of kernel prefetching on buffer cache replacement algorithms

The performance impact of kernel prefetching on buffer cache replacement algorithms

RACE: A Robust Adaptive Caching Strategy for Buffer Cache

PARC: A novel OS cache manager

Contact Info

Product

Resources

About