Improving network systems performance by clustering distributed database sites

Abstract

Clustering network sites is a vital issue in parallel and distributed database systems DDBS. Grouping distributed database network sites into clusters is considered an efficient way to minimize the communication time required for query processing. However, clustering network sites is still an open research problem since its optimal solution is NP-complete. The main contribution in this field is to find a near optimal solution that groups distributed database network sites into disjoint clusters in order to minimize the communication time required for data allocation. Grouping a large number of network sites into a small number of clusters effectively increases the transaction response time, results in better data distribution, and improves the distributed database system performance. We present a novel algorithm for clustering distributed database network sites based on the communication time as database query processing is time dependent. Extensive experimental tests and simulations are conducted on this clustering algorithm. The experimental and simulation results show that a better network distribution is achieved with significant network servers load balance and network delay, a minor communication time between network sites is realized, and a higher distributed database system performance is recognized.

Scalable Data-Oriented Replication with Flexible Consistency in Real-Time Data Systems

“…Table 1 summarizes the set of parameters and the baseline settings for the experiment. In fact, the settings of the evaluation experiments are very similar to those of the closest related work in (Mathiason, Andler & Son 2007) and (Hababeh 2012).…”

“…However the work by Hababeh (2012) must come to the fore. This work introduces an intelligent clustering technique that segments the distributed database network sites into disjoint clusters according to the lowest average communication cost between network sites.…”

“…To make another comparative evaluation against the clustering method by (Hababeh 2012) in terms of the maximum load (bit/sec) which is generated from all nodes as an indicator of the consumed communication cost, we considered case of 10 nodes compared with (Hababeh 2012) and his most related work (Kumar et al 2007) and (Fronczak et al 2002). Figure 3 depicts the load of the clustering methods under comparison.…”

“…The degree of replication is a result of allocating nodes to the clusters where its data objects are accessed. This requires typically much fewer nodes than used in a fully replicated database (Hababeh 2012). Because time is critical in such systems, the clustering method must take into account the timing properties of both the communication network and data of the access pattern.…”

Dynamic data reallocation and replication over a cloud environment

“…37 As a result, clustering DDBS sites is an important issue in DDBS. 37 As a result, clustering DDBS sites is an important issue in DDBS.…”