Improving Read Performance with Online Access Pattern Analysis and Prefetching

Tang, Houjun; Zou, Xiaocheng; Jenkins, John; Boyuka, David A.; Ranshous, Stephen; Kimpe, Dries; Klasky, Scott; Samatova, Nagiza F.

doi:10.1007/978-3-319-09873-9_21

Cited by 15 publications

(9 citation statements)

References 14 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This information is used to guide prefetching to a local buffer. Similar approaches -access pattern detection to guide prefetching -are proposed by Patrick et al [81], He, Sun, and Thakur [32], Lu et al [65], and Tang et al [97].…”

Section: Caching and Prefetchingmentioning

confidence: 92%

See 1 more Smart Citation

A Checkpoint of Research on Parallel I/O for High-Performance Computing

et al. 2018

View full text Add to dashboard Cite

We present a comprehensive survey on parallel I/O in the high performance computing (HPC) context. This is an important field for HPC because of the historic gap between processing power and storage latencies, which causes applications performance to be impaired when accessing or generating large amounts of data. As the available processing power and amount of data increase, I/O remains a central issue for the scientific community. In this survey, we focus on a traditional I/O stack, with a POSIX parallel file system. We present background concepts everyone could benefit from. Moreover, through the comprehensive study of publications from the most important conferences and journals in a five-year time window, we discuss the state of the art of I/O optimization approaches, access pattern extraction techniques, and performance modeling, in addition to general aspects of parallel I/O research. Through this approach, we aim at identifying the general characteristics of the field and the main current and future research topics.

show abstract

Section: Caching and Prefetchingmentioning

confidence: 92%

“…The approach proposed by Tang et al [97] periodically analyzes past accesses and applies a rules library to predict future accesses (for prefetching). They collect spatiality information about read requests from the MPI-IO library.…”

Section: Runtime Detectionmentioning

confidence: 99%

A Checkpoint of Research on Parallel I/O for High-Performance Computing

et al. 2018

View full text Add to dashboard Cite

show abstract

“…To predict future accesses, the technique proposed by Dorier et al [8] named Omnisc'IO, intercepts I/O operations and builds a grammar. The approach employed by Tang et al [9] periodically analyzes previous accesses and applies a rule library to predict future accesses (for prefetching). They collect metrics regarding spatiality of read requests from the MPI-IO library.…”

Section: Related Workmentioning

confidence: 99%

Detecting I/O Access Patterns of HPC Workloads at Runtime

Bez

Boito

Nou

et al. 2019

2019 31st International Symposium on Computer Architecture and High Performance Computing (SBAC-PAD)

View full text Add to dashboard Cite

In this paper, we seek to guide optimization and tuning strategies by identifying the application's I/O access pattern. We evaluate three machine learning techniques to automatically detect the I/O access pattern of HPC applications at runtime: decision trees, random forests, and neural networks. We focus on the detection using metrics from file-level accesses as seen by the clients, I/O nodes, and parallel file system servers. We evaluated these detection strategies in a case study in which the accurate detection of the current access pattern is fundamental to adjust a parameter of an I/O scheduling algorithm. We demonstrate that such approaches correctly classify the access pattern, regarding file layout and spatiality of accesses-into the most common ones used by the community and by I/O benchmarking tools to test new I/O optimization-with up to 99% precision. Furthermore, when applied to our study case, it guides a tuning mechanism to achieve 99% of the performance of an Oracle solution.

show abstract

“…Existing work on cache management can be classified into three categories: single-node cache management, distributed cache management, and cache management on the hierarchical storage. Early efforts, such as LRU, LFU, FIFO, CLOCK [24], were primarily dedicated to single-node cache management. These algorithms, as well as their ramifications [18], [20] have been used for cache management on various devices, such as disk cache on DRAM and on SSDs.…”

Section: Related Workmentioning

confidence: 99%

“…Eley [33] and [34] prefetches the input data of applications into SSD to reduce the latency of reading data inputs. Byna [2] and Tang [24], [23] propose prefetching algorithms that fetch data based on the historical access patterns of flat files. In contrast, ARCHIE works efficiently on multi-dimensional arrays to accelerate read operations of analysis applications using hierarchical storage systems.…”

Section: Related Workmentioning

confidence: 99%

ARCHIE: Data Analysis Acceleration with Array Caching in Hierarchical Storage

Dong

Wang

Tang

et al. 2018

2018 IEEE International Conference on Big Data (Big Data)

Self Cite

View full text Add to dashboard Cite

Scientific data analysis typically involves reading massive amounts of data generated by simulations, experiments, and observations. A significant bottleneck in this process is reading such data because the data files are stored on the rotating disks. Recent supercomputing systems are adding non-volatile storage layers to fill the performance gap between fast main memory and the slow disk-based storage. Software libraries for managing this hierarchy not only need to read data efficiently, but also reduce user-involvement for cross-layer data movement. As the scientific data is usually organized as arrays, these libraries also need to support array data access patterns over hierarchical storage systems. Existing software tools manage individual storage layers separately, and require significant manual work to move data among the layers. In this paper, we introduce a new array caching in hierarchical storage (ARCHIE) to accelerate array data analyses in a seamless fashion. ARCHIE evaluates array access patterns and prefetches data with array semantics between storage layers. On a production supercomputing system, our evaluation shows that ARCHIE outperforms state-of-the-art file systems, i.e., Lustre and DataWarp, by up to 5.8× in accessing data by scientific analysis applications.

show abstract

Improving Read Performance with Online Access Pattern Analysis and Prefetching

Cited by 15 publications

References 14 publications

A Checkpoint of Research on Parallel I/O for High-Performance Computing

A Checkpoint of Research on Parallel I/O for High-Performance Computing

Detecting I/O Access Patterns of HPC Workloads at Runtime

ARCHIE: Data Analysis Acceleration with Array Caching in Hierarchical Storage

Contact Info

Product

Resources

About