A Resource Scheduling Model Supporting Differentiated Service for Grid Computing

Resource Scheduling Techniques in Utility Computing

“…It makes use of the time-shared and spaceshared tactics. Both tactics assign jobs on a processing element and exclusively use the "bandwidth" in scheduling to save time (Weiyi, Junwei, Minglu & Chuliang, 2008).…”

“…T1 (Buyya & Murshed, 2002;Buyya, et al, 2005) • on the cost-optimization and time-optimization scheduling algorithms • Groups resources with the same cost and then applies time-optimisation scheduling strategy • Time complexity of O(NM) where N is the number of jobs and M is number of resources T2 (Islam et al, 2003;Islam et al, 2004;Yeo & Buyya, 2005) • Employs a kill-and-restart mechanism • Time complexity of O(K log N) where N is the number of jobs T3 (Milligan, 2006) • Used in a contract-based workflow management system relating to demand use of a Grid enabled data warehouse • Flexible scheduling algorithms T4 (Liu et al, 2005) • A Utility-driven heuristic scheduling strategy • More suitable for handling dynamic demand of users T5 (Yeo & Buyya, 2007) • Admits jobs after examining their required QoS if it can be satisfied or not • Allocates jobs to nodes with desired resources T6 (Netto & Buyya, 2007) • Handles online scheduling of adaptive requests with hard deadlines • The scheduler schedules jobs after analyzing profit values along with reduction of average job completion time T7 (Li et al, 2007) • Provides QoS constrained service management system based on economic model • Handles QoS attributes such as task execution time, service request-to-response time and execution cost T8 (Chunlin & Layuan, 2007) • Uses an iterative algorithm to solve the problems • Composed of three dimensions: cost, deadline, and reliability T9 (Buyya & Sulistio, 2008;Ramakrishnan, 2007) • It performs removal of data files when they are no longer needed • Provides efficient mapping and execution of workflows T10 (Dong & Luo, 2007) • Introduces QoS matching offset between task and resource • Resources in the scheduling process are allocated on demand T11 (Vengerov et al, 2010) • Deals with the problem of dynamic scheduling of data-intensive multiprocessor jobs • Uses Reinforcement learning method T12 (Lorpunmanee et al, 2007) • Can efficiently and effectively allocate jobs to appropriate resources • Solves discrete optimization problems(static and dynamic combinational optimization problems) T13 (Weiyi et al, 2008) • A non-centralized resource scheduling • Supports service level subscription mechanism T14 (Lin & Shih, 2008) • An effective strategy for reinforcing resources to or reclaiming resources from the tasks • Scheduler periodically reallocates computing nodes to the ongoing tasks T15 (Lee et al, 2010) • Focuses on multiple service requests from a number of consumers • Executes two profit driven service request scheduling algorithms MaxProfit and MaxUtil. T16 …”