Simulation of dynamic data replication strategies in Data Grids

Houda Lamehamedi, Bolesław K. Szymański, Ewa Deelman

2004 · 166 citations · 14 references

Concepts

TL;DR

Data Grids provide geographically distributed resources for large‑scale data‑intensive applications, but high Internet latencies hinder efficient access to the massive, widely dispersed datasets they generate. The paper proposes dynamic replication and caching strategies, including adaptive replica distribution topologies, to support many users with continuously changing data and processing needs through a dynamic memory middleware and replication algorithm. Replication decisions are guided by a cost‑estimation model that incorporates runtime read/write statistics, network latency, bandwidth, and replica size, and the approach is implemented in a dynamic memory middleware evaluated with the GridNet simulator. Simulation results show that replication reduces data access time in Data Grids, with larger performance gains for larger datasets.

Abstract

Data Grids provide geographically distributed resources for large-scale data-intensive applications that generate large data sets. However, ensuring efficient access to such huge and widely distributed data is hindered by the high latencies of the Internet. We address these challenges by employing intelligent replication and caching of objects at strategic locations. In our approach, replication decisions are based on a cost-estimation model and driven by the estimation of the data access gains and the replica's creation and maintenance costs. These costs are in turn based on factors such as runtime accumulated read/write statistics, network latency, bandwidth, and replica size. To support large numbers of users who continuously change their data and processing needs, we introduce scalable replica distribution topologies that adapt replica placement to meet these needs. In this paper we present the design of our dynamic memory middleware and replication algorithm. To evaluate the performance of our approach, we developed a Data Grid simulator, called the GridNet. Simulation results demonstrate that replication improves the data access time in Data Grids, and that the gain increases with the size of the datasets involved.

References

14