Academic Journals Database
Disseminating quality controlled scientific knowledge

ESRWF: Extreme State-Rank based Workload Factoring for Integrated Cloud Computing Model

ADD TO MY LIST
 
Author(s): Snehil Sharma | Abhishek Mathur | Shailendra Shrivastava

Journal: International Journal of Electronics and Computer Science Engineering
ISSN 2277-1956

Volume: 1;
Issue: 3;
Start page: 1340;
Date: 2012;
VIEW PDF   PDF DOWNLOAD PDF   Download PDF Original page

Keywords: Memory-Prediction-Error (εεε)

ABSTRACT
— Despite all the hypes, the designers of Workload Factoring solutions have consistently debated on the tradeoffs between factoring of workload and their resulting performance. When computational resources are provided to the existing system; one of the main challenges that the computing community is face dazzling workload of application requests and deteriorating performance at runtime of particular server-site. To address those concerns that, we propose an “Extreme State-Rank based Workload Factoring for Integrated Cloud Computing Model”. It consolidates components in Cloud infrastructure at On-Off both type premises of internet based applications. The Intelligence in this approach lies in extreme factoring of performances of system into States (Regular & Critical) during period of experiencing workload by running system and dynamically distribution of State-Rank {, , ∅} to incoming requests at duration of particular system state in running system. We also propose ESRATE as efficient Workload Factoring algorithm with some specified features, which enables factoring of incoming oppressive requests on the basis of ‘State, State-Rank & Request-WeightCoefficient’ explicitly. Through extensive analysis with efficient simulated evaluations, we predict and showed Workload Factoring technology to achieve Request Concurrency with availability of servers, Unique Video Requested with best resource proficiency ( = %), the ratio of the maximum load to the average load ( = . ) with formulation of speedup () and Memory-Prediction-Error (εεε) as well as reduced data cache and replication overhead in critical state especially
Save time & money - Smart Internet Solutions      Why do you need a reservation system?