Replication Strategy Based on Data Relationship in Grid Computing

Yusof, Yuhanis

doi:10.5121/csit.2013.3831

Cited by 2 publications

(3 citation statements)

References 15 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…If the replication strategy unwisely copies all requested files over cloud nodes, a huge amount of storage has been consumed. In determining the file importance, we use the exponential growth/decay model for the file's number of access in the history of accesses every 30 seconds [48]. In general, the trend of accessing files in distributed environments such as grid and cloud follows an exponential model.…”

Section: Popular File Determinationmentioning

confidence: 99%

Energy-Aware Data Replication in Green Cloud Computing using Improved Bat Algorithm and Paralleling Technique

Zade

Mansouri

Javidi

2022

Preprint

View full text Add to dashboard Cite

In green cloud computing, resources are allocated in a reasonable way so that cloud computing systems are energy efficient and consume little power. In this paper, Parallel-based data replication with an Improved Fuzzy-Bat technique called PIF-Bat is proposed. At first, we introduce an Improved Fuzzy-Bat (IF-Bat) strategy that tunes some parameters of the standard Bat algorithm and controls the tradeoff between exploration and exploitation. By balancing the trade-offs among the five optimization objectives (i.e., availability, service time, load, latency, centrality, energy consumption), the PIF-Bat algorithm determines the optimal locations for replicas using a multi-objective optimization strategy based on the Improved Fuzzy-Bat technique. A successful attack against each site will reveal no useful information even if only a single part of the file is exposed. A fuzzy inference system is used in the PIF-Bat algorithm to determine whether to parallelize files based on data such as file size, free space, and average bandwidth at each node. In order to reduce retrieval time, a parallel download technique, which allows users to download portions of a file simultaneously from different sites, is applied. The experimental results and statistical tests with a set of well-known test functions demonstrate the superior exploitation and exploration ability of IF-Bat. Furthermore, PIF-Bat obtains lower access latency around 15%∼20% and better performance than other similar replication algorithms under high load conditions.

show abstract

Section: Popular File Determinationmentioning

confidence: 99%

Energy-Aware Data Replication in Green Cloud Computing using Improved Bat Algorithm and Paralleling Technique

Zade

Mansouri

Javidi

2022

Preprint

View full text Add to dashboard Cite

show abstract

“…Get the first f candidate from the sorted list ; 12 AccumulativeFreeSpace += Size(f candidate ) ; 13 AccumulativeCandWeight += F ileW eight(f candidate ) ; …”

Section: Replication Processmentioning

confidence: 99%

“…4) Dynamic replication strategy based on exponential model and dependency relationships in data grid [13]: Periodically, the strategy selects the most important files which must be replicated by combining information from users' behavior and files behavior. The former is measured by estimating the future access number to files and the latter by calculating the dependency level of each file to other files in the grid.…”

Section: Background and Related Workmentioning

confidence: 99%

New Replication Strategy Based on Maximal Frequent Correlated Pattern Mining for Data Grids

Slimani

Hamrouni

Charrada

2014

2014 15th International Conference on Parallel and Distributed Computing, Applications and Technologies

View full text Add to dashboard Cite

Data replication in data grids is an efficient technique that aims to improve response time, reduce the bandwidth consumption and maintain reliability. In this context, a lot of work is done and many strategies have been proposed. Unfortunately, most of existing replication techniques are based on single file granularity and neglect correlation among different data files. Indeed, file correlations become an increasingly important consideration for performance enhancement in data grids. In fact, the analysis of real data intensive grid applications reveals that job requests for groups of correlated files and suggests that these correlations can be exploited for improving the effectiveness of replication strategies. In this paper, we propose a new dynamic periodic decentralized data replication strategy, called RSBMFCP ( 1 ) , which consider a set of correlated files as granularity. Our strategy gathers files according to a relationship of simultaneous accesses between files by jobs and stores correlated files at the same site. In order to find out these correlations, a maximal frequent correlated pattern mining algorithm of the data mining field is introduced. We choose the all-confidence as correlation measure. The proposed strategy consists of four steps: storing file access history, converting the file access history into a logical history file, applying maximal frequent correlated pattern mining algorithm and performing replication and replacement. Experiments using the well-known data grid simulator OptorSim show that our proposed strategy has better performance in comparison with other strategies in terms of job execution time and effective network usage.

show abstract

Replication Strategy Based on Data Relationship in Grid Computing

Abstract: ABSTRACT

Cited by 2 publications

References 15 publications

Energy-Aware Data Replication in Green Cloud Computing using Improved Bat Algorithm and Paralleling Technique

Energy-Aware Data Replication in Green Cloud Computing using Improved Bat Algorithm and Paralleling Technique

New Replication Strategy Based on Maximal Frequent Correlated Pattern Mining for Data Grids

Contact Info

Product

Resources

About