Open Access
Design and Evaluation of Dynamic Replication Strategies for a High-Performance Data Grid
TLDR
A simulation framework that is developed to model a grid scenario, which enables comparative studies of alternative dynamic replication strategies for three different kinds of access patterns, and shows that the best strategy has significant savings in latency and bandwidth consumption if the access patterns contain a moderate amount of geographical locality.Abstract:
Physics experiments that generate large amounts of data need to be able to share it with researchers around the world. High performance grids facilitate the distribution of such data to geographically remote places. Dynamic replication can be used as a technique to reduce bandwidth consumption and access latency in accessing these huge amounts of data. We describe a simulation framework that we have developed to model a grid scenario, which enables comparative studies of alternative dynamic replication strategies. We present preliminary results obtained with this simulator, in which we evaluate the performance of six different replication strategies for three different kinds of access patterns. The simulation results show that the best strategy has significant savings in latency and bandwidth consumption if the access patterns contain a moderate amount of geographical locality.read more
Citations
More filters
Journal ArticleDOI
Tree-based Space Efficient Formats for Storing the Structure of Sparse Matrices
Ivan Simecek,Daniel Langr +1 more
TL;DR: This paper proposes and evaluates new storage formats for sparse matrices that minimize the space complexity of information about matrix structure and proves that the latter are considerably more space efficient.
Proceedings ArticleDOI
FRCS: A File Replication and Consistency Service in Data Grids
TL;DR: The experimental results show that the FRCS is more efficient than other replication strategies and striking a balance between improving data access performance and replica consistency in data grids.
Proceedings ArticleDOI
A Dynamic Replica Placement Mechanism Based on Response Time Measure
Wenfeng Wang,Wenhong Wei +1 more
TL;DR: A model called RPRTM (Replica Placement based on Response Time Measure) is presented, and a genetic algorithm is designed to figure out the model and indicates that it can satisfy every node’s response time requirement of a single request and minimize the number of replicas at the same time.
Journal ArticleDOI
A File Group Data Replication Algorithm for Data Grids
TL;DR: A new dynamic data replication algorithm named Popular File Group Replication, PFGR is introduced which is based on three assumptions: first, users in a grid site (Virtual Organization) have similar interests in files and second, they have the temporal locality of file accesses and third, all files are read-only.
Proceedings ArticleDOI
Cold data eviction using node congestion probability for HDFS based on Hybrid SSD
TL;DR: A new efficient cold data eviction scheme is proposed which is based on the state of node congestion of Hybrid SSD for HDFS and computer simulation reveals that the proposed scheme significantly reduces average recovery and execution time in comparison to the existing replication schemes.
References
More filters
Journal ArticleDOI
The Anatomy of the Grid: Enabling Scalable Virtual Organizations
TL;DR: The authors present an extensible and open Grid architecture, in which protocols, services, application programming interfaces, and software development kits are categorized according to their roles in enabling resource sharing.
Posted Content
The Anatomy of the Grid - Enabling Scalable Virtual Organizations
TL;DR: This article reviews the "Grid problem," and presents an extensible and open Grid architecture, in which protocols, services, application programming interfaces, and software development kits are categorized according to their roles in enabling resource sharing.
Journal ArticleDOI
Summary cache: a scalable wide-area web cache sharing protocol
TL;DR: This paper demonstrates the benefits of cache sharing, measures the overhead of the existing protocols, and proposes a new protocol called "summary cache", which reduces the number of intercache protocol messages, reduces the bandwidth consumption, and eliminates 30% to 95% of the protocol CPU overhead, all while maintaining almost the same cache hit ratios as ICP.
Journal ArticleDOI
The data grid
TL;DR: In this paper, the authors introduce design principles for a data management architecture called the data grid, and describe two basic services that are fundamental to the design of a data grid: storage systems and metadata management.
Proceedings ArticleDOI
Summary cache: a scalable wide-area Web cache sharing protocol
TL;DR: This paper proposes a new protocol called "Summary Cache"; each proxy keeps a summary of the URLs of cached documents of each participating proxy and checks these summaries for potential hits before sending any queries, which enables cache sharing among a large number of proxies.