An intelligent prefetching strategy for a data grid prototype system
Jianxi Chen, Dan Feng
Abstract
Jianxi Chen, Dan Feng
Abstract
An amount of data-intensive applications pay more attention to the performance of storage system. Prefetching is an efficient approach to improve the I/O performance. The present algorithms of prefetching have some limitations and cannot suit for some kinds of access patterns. Intelligent prefetching strategy can recognize the current access pattern by analyzing the access trace independently. Prefetching based on the pattern recognized is more effective than blindly prefetching. Using this algorithm in a VI-based data grid prototype system (VI-DGPS), the storage nodes is divided into groups by the relation of the data, only one node judges the pattern and all other nodes just receive the pattern info in a group. All nodes in a group prefetch synchronously based on the access pattern to reduce the request response time. Test result shows that the strategy has a good performance in our system.
OpenAlex reports 3 citations for this work. Citation counts describe recorded attention and do not establish research quality.
A contribution statement is not available in the OpenAlex record.
Method details are not available in the OpenAlex metadata.
Findings are not separately available in the OpenAlex metadata.
Limitations are not available in the OpenAlex metadata.
Application details are not available in the OpenAlex metadata.
An amount of data-intensive applications pay more attention to the performance of storage system. Prefetching is an efficient approach to improve the I/O performance. The present algorithms of prefetching have some limitations and cannot suit for some kinds of access patterns. Intelligent prefetching strategy can recognize the current access pattern by analyzing the access trace independently. Prefetching based on the pattern recognized is more effective than blindly prefetching. Using this algorithm in a VI-based data grid prototype system (VI-DGPS), the storage nodes is divided into groups by the relation of the data, only one node judges the pattern and all other nodes just receive the pattern info in a group. All nodes in a group prefetch synchronously based on the access pattern to reduce the request response time. Test result shows that the strategy has a good performance in our system.
Key concepts: Instruction prefetch, Computer science, Data access, Grid, Node (physics), TRACE (psycholinguistics), Computer data storage, Relation (database)