Discretionary caching for I/O on clusters

Vilayannur, Murali; Sivasubramaniam, Anand; Kandemir, Mahmut; Thakur, Rajeev; Ross, Rob

doi:10.1109/ccgrid.2003.1199357

Cited by 19 publications

(18 citation statements)

References 33 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Nitzberg et al proposed collective buffering [29], and Ma et al proposed active buffering [27] to boost the output performance of many scientific applications. Vilayannur et al proposed discretionary caching for parallel I/O to use compilation and runtime support to bypass caching if the caching hurts the performance [43]. Eshel et al designed a cluster file system cache named Panache that exploits parallelism in many aspects of its design and has been proven effective and scalable as a parallel file system cache [12].…”

Section: B Parallel I/o Caching and Prefetchingmentioning

confidence: 99%

LACIO: A New Collective I/O Strategy for Parallel I/O Systems

Chen

Sun

Thakur

et al. 2011

2011 IEEE International Parallel &Amp; Distributed Processing Symposium

View full text Add to dashboard Cite

Abstract-Parallel applications benefit considerably from the rapid advance of processor architectures and the available massive computational capability, but their performance suffers from large latency of I/O accesses. The poor I/O performance has been attributed as a critical cause of the low sustained performance of parallel systems. Collective I/O is widely considered a critical solution that exploits the correlation among I/O accesses from multiple processes of a parallel application and optimizes the I/O performance. However, the conventional collective I/O strategy makes the optimization decision based on the logical file layout to avoid multiple file system calls and does not take the physical data layout into consideration. On the other hand, the physical data layout in fact decides the actual I/O access locality and concurrency. In this study, we propose a new collective I/O strategy that is aware of the underlying physical data layout. We confirm that the new Layout-Aware Collective I/O (LACIO) improves the performance of current parallel I/O systems effectively with the help of noncontiguous file system calls. It holds promise in improving the I/O performance for parallel systems.

show abstract

Section: B Parallel I/o Caching and Prefetchingmentioning

confidence: 99%

LACIO: A New Collective I/O Strategy for Parallel I/O Systems

Chen

Sun

Thakur

et al. 2011

2011 IEEE International Parallel &Amp; Distributed Processing Symposium

View full text Add to dashboard Cite

show abstract

“…Since multiple clients share the same memory cache, its efficient utilization is clearly very critical. Since global caches have already been studied in the context of PVFS and it is not one of the contributions of this paper, we do not elaborate on our global cache implementation any further in this paper, except for saying that it closely follows the implementation presented in [33]. Our global cache management method employs a LRU (least-recently-used) policy with aging method to determine a best candidate for replacement as a result of a cache miss.…”

Section: Experimental Platform and Benchmarksmentioning

confidence: 99%

Prefetch throttling and data pinning for improving performance of shared caches

Öztürk

Son

Kandemir

et al. 2008

2008 SC - International Conference for High Performance Computing, Networking, Storage and Analysis

View full text Add to dashboard Cite

Abstract-In this paper, we (i) quantify the impact of compilerdirected I/O prefetching on shared caches at I/O nodes. The experimental data collected shows that while I/O prefetching brings some benefits, its effectiveness reduces significantly as the number of clients (compute nodes) is increased; (ii) identify interclient misses due to harmful I/O prefetches as one of the main sources for this reduction in performance with increased number of clients; and (iii) propose and experimentally evaluate prefetch throttling and data pinning schemes to improve performance of I/O prefetching. Prefetch throttling prevents one or more clients from issuing further prefetches if such prefetches are predicted to be harmful, i.e., replace from the memory cache the useful data accessed by other clients. Data pinning on the other hand makes selected data blocks immune to harmful prefetches by pinning them in the memory cache. We show that these two schemes can be applied in isolation or combined together, and they can be applied at a coarse or fine granularity. Our experiments with these two optimizations using four disk-intensive applications reveal that they can improve performance by 9.7% and 15.1% on average, over standard compiler-directed I/O prefetching and no-prefetch case, respectively, when 8 clients are used.

show abstract

“…Since multiple CPUs (computation nodes) can share the same memory cache, its efficient utilization is clearly critical. Since global caches have already been studied in the context of PVFS and it is not one of the contributions of this paper, we do not elaborate on our PVFS-based global cache implementation any further in this paper, except for saying that it closely follows the implementation presented in [23]. Our global cache management method employs an LRU (least-recently-used) policy with aging method to determine the best candidate for replacement as a result of a cache miss.…”

Section: Methodsmentioning

confidence: 99%

“…Targeting multi-level caches, several multi-level buffer cache management policies have been proposed [43,40,23,14]. [40] introduced a DEMOTE operation where an evicted cache block is migrated to lower level of buffer cache.…”

Section: Related Workmentioning

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

Discretionary caching for I/O on clusters

Cited by 19 publications

References 33 publications

LACIO: A New Collective I/O Strategy for Parallel I/O Systems

LACIO: A New Collective I/O Strategy for Parallel I/O Systems

Prefetch throttling and data pinning for improving performance of shared caches

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

Contact Info

Product

Resources

About