Efficient and Robust Database Support for Data-Intensive Applications in Dynamic Environments

Proceedings of the 2011 ACM Symposium on Applied Computing

2011

Self Cite

Semantic caching augments cached data with a semantic description of the data. These semantic descriptions can be used to improve execution time for similar queries by retrieving some data from cache and issuing a remainder query for the rest. This is an improvement over traditional page caching, since caches are no longer limited to only base tables but are extended to contain intermediate results. In large-scale distributed database systems, using a central server with complete knowledge of the system will be a serious bottleneck and single point of failure. In this paper, we propose a distributed semantic caching method where sites make autonomous caching decisions based on locally available information, thereby reducing the need for centralized control. We implement the method in the DASCOSA-DB distributed database system prototype and use this implementation to do experiments that show the applicability and efficiency of our approach. Our evaluation shows that execution times for queries with similar subqueries are significantly reduced and that overhead caused by cache management is marginal.

Section: Introductionmentioning

confidence: 99%

Site-autonomous distributed semantic caching

Proceedings of the 2011 ACM Symposium on Applied Computing

2011

Self Cite

“…Metadata management, including information on fragmentation and where replicas are stored, is performed through a common catalog service. This catalog service can be realized in a number of ways, for example in our prototype system we use a distributed hash table where all sites participate [16].…”

Section: System Modelmentioning

confidence: 99%

“…Comparative results for the two multi-site workloads using w BE = 0.9 and w FS = 0.50, are shown in Fig. 7(a) and Table 6. 7.4 Implementation of DYFRAM in DASCOSA-DB In this experiment, DYFRAM was implemented in the DASCOSA-DB distributed database system [16] in order to verify simulation results. The workloads tested are similar to the grid and general workloads presented in Sect.…”

Section: Workloads Involving Two Sitesmentioning

confidence: 99%

See 1 more Smart Citation

DYFRAM: dynamic fragmentation and replica management in distributed database systems

Distrib Parallel Databases

2010

Self Cite

In distributed database systems, tables are frequently fragmented and replicated over a number of sites in order to reduce network communication costs. How to fragment, when to replicate and how to allocate the fragments to the sites are challenging problems that has previously been solved either by static fragmentation, replication and allocation, or based on a priori query analysis. Many emerging applications of distributed database systems generate very dynamic workloads with frequent changes in access patterns from different sites. In such contexts, continuous refragmentation and reallocation can significantly improve performance. In this paper we present DYFRAM, a decentralized approach for dynamic table fragmentation and allocation in distributed database systems based on observation of the access patterns of sites to tables. The approach performs fragmentation, replication, and reallocation based on recent access history, aiming at maximizing the number of local accesses compared to accesses from remote sites. We show through simulations and experiments on the DASCOSA distributed database system that the approach significantly reduces communication costs for typical access patterns, thus demonstrating the feasibility of our approach.

The DASCOSA-DB Grid Database System

Grid and Cloud Database Management

2011

Computational science applications performing distributed computations using grid networks are now emerging. These applications have new and demanding requirements for efficient query processing. In order to meet these requirements, we have developed the DASCOSA-DB distributed database system. In this chapter, a detailed overview of the architecture and implementation of DASCOSA-DB is given, as well as a description of novel features developed in order to better support typical data-intensive applications running on a grid system: fault-tolerant query processing, dynamic refragmentation, allocation and replication of data fragments, and distributed semantic caching.