Optimal prefetching and caching for parallel I/O sytems

Kallahalla, Mahesh; Varman, Peter

doi:10.1145/378580.378648

Cited by 37 publications

(34 citation statements)

References 19 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Prefetch buffers for disk load balancing and overlapping of I/O and computation has been intensively studied [21,7,3,14,13,12]. But we have not seen results that guarantee overlapping of I/O and computation during parallel disks merging of arbitrary runs.…”

Section: Related Workmentioning

confidence: 89%

“…Innovations with respect to this work are: A different allocation strategy that enables better theoretical I/O bounds [13,12]; a prefetching algorithm that optimizes I/O steps and never evicts data previously fetched; overlapping of I/O and computation; a completely asynchronous imple- 1 The cost of the hardware is spread over three years. Then it is measured how much data can be sorted in an interval of time that costs one US-cent.…”

Section: Related Workmentioning

confidence: 99%

See 1 more Smart Citation

Asynchronous parallel disk sorting

Dementiev

Sanders

2003

Proceedings of the Fifteenth Annual ACM Symposium on Parallel Algorithms and Architectures

View full text Add to dashboard Cite

We develop an algorithm for parallel disk sorting, whose I/O cost approaches the lower bound and that guarantees almost perfect overlap between I/O and computation. Previous algorithms have either suboptimal I/O volume or cannot guarantee that I/O and computations can always be overlapped. We give an efficient implementation that can (at least) compete with the best practical implementations but gives additional performance guarantees. For the experiments we have configured a state of the art machine that can sustain full bandwidth I/O with eight disks and is very cost effective.

show abstract

Section: Related Workmentioning

confidence: 89%

Section: Related Workmentioning

confidence: 99%

Asynchronous parallel disk sorting

Dementiev

Sanders

2003

Proceedings of the Fifteenth Annual ACM Symposium on Parallel Algorithms and Architectures

View full text Add to dashboard Cite

show abstract

“…While there exist several I/O prefetching algorithms published in literature [27,2,33,36,19,11,15], the one used in this work is inspired by the work done by Mowry et al [33]. The original algo- rithm has actually been proposed for improving hardware cache behavior for memory-resident data sets [32], and has later been extended to implement I/O prefetching targeting virtual memory based execution environments.…”

Section: Compiler-directed I/o Prefetch-ingmentioning

confidence: 99%

“…I/O prefetching is an important optimization for improving performance [27,2,33,1,36,19,11,15,41,7,34]. In I/O prefetching, data is brought from the disk to the memory cache (shared storage buffer) ahead of time to hide the latency of disk accesses.…”

Section: Introductionmentioning

confidence: 99%

Profiler and compiler assisted adaptive I/O prefetching for shared storage caches

Son

Muralidhara

Öztürk

et al. 2008

Proceedings of the 17th International Conference on Parallel Architectures and Compilation Techniques

View full text Add to dashboard Cite

I/O prefetching has been employed in the past as one of the mechanisms to hide large disk latencies. However, I/O prefetching in parallel applications is problematic when multiple CPUs share the same set of disks due to the possibility that prefetches from different CPUs can interact on shared memory caches in the I/O nodes in complex and unpredictable ways. In this paper, we (i) quantify the impact of compiler-directed I/O prefetching -developed originally in the context of sequential execution -on shared caches at I/O nodes. The experimental data collected shows that while I/O prefetching brings benefits, its effectiveness reduces significantly as the number of CPUs is increased; (ii) identify inter-CPU misses due to harmful prefetches as one of the main sources for this reduction in performance with the increased number of CPUs; and (iii) propose and experimentally evaluate a profiler and compiler assisted adaptive I/O prefetching scheme targeting shared storage caches. The proposed scheme obtains inter-thread data sharing information using profiling and, based on the captured data sharing patterns, divides the threads into clusters and assigns a separate (customized) I/O prefetcher thread for each cluster. In our approach, the compiler generates the I/O prefetching threads automatically. We implemented this new I/O prefetching scheme using a compiler and the PVFS file system running on Linux, and the empirical data collected clearly underline the importance of adapting I/O prefetching based on program phases. Specifically, our proposed scheme improves performance, on average, by 19.9%, 11.9% and 10.3% over the cases without I/O prefetching, with independent I/O prefetching (each CPU is performing compiler-directed I/O prefetching independently), and with one CPU prefetching (one CPU is reserved for prefetching on behalf of others), respectively, when 8 CPUs are used.

show abstract

“…There has been a considerable amount of prior work on optimizing I/O and I/O caches [30,10,27,29,19,20,23,33,17,4,25,13], some of which has been on clusters as well. Recently, [6,40] have focused on buffer cache management policies in a multi-level buffer cache system.…”

Section: Related Workmentioning

confidence: 99%

Discretionary caching for I/O on clusters

Vilayannur

Sivasubramaniam

Kandemir

et al. 2003

CCGrid 2003. 3rd IEEE/ACM International Symposium on Cluster Computing and the Grid, 2003. Proceedings.

View full text Add to dashboard Cite

I/O bottlenecks are already a problem in many large-scale applications that manipulate huge datasets. This problem is expected to get worse as applications get larger, and the I/O subsystem performance lags behind processor and memory speed improvements. At the same time, off-the-shelf clusters of workstations are becoming a popular platform for demanding applications due to their cost-effectiveness and widespread deployment. Caching I/O blocks is one effective way of alleviating disk latencies, and there can be multiple levels of caching on a cluster of workstations.Previous studies have shown the benefits of caching -whether it be local to a particular node, or a shared global cache across the cluster -for certain applications. However, we show that while caching is useful in some situations, it can hurt performance if we are not careful about what to cache and when to bypass the cache. This paper presents compilation techniques and runtime support to address this problem. These techniques are implemented and evaluated on an experimental Linux/Pentium cluster running a parallel file system. Our results using a diverse set of applications (scientific and commercial) demonstrate the benefits of a discretionary approach to caching for I/O subsystems on clusters, providing as much as 48% savings in overall execution time over indiscriminately caching everything in some applications.

show abstract

Optimal prefetching and caching for parallel I/O sytems

Cited by 37 publications

References 19 publications

Asynchronous parallel disk sorting

Asynchronous parallel disk sorting

Profiler and compiler assisted adaptive I/O prefetching for shared storage caches

Discretionary caching for I/O on clusters

Contact Info

Product

Resources

About