Joint Computing and Network Resource Scheduling in a Lambda Grid Network

Abstract

Abstract-Data-intensive Grid applications require huge data transfers between grid computing nodes. These computing nodes, where computing jobs are executed, are usually geographically separated. A grid network that employs optical wavelength division multiplexing (WDM) technology and optical switches to interconnect computing resources with dynamically provisioned multigigabit rate bandwidth lightpath is called a Lambda Grid network. A computing task may be executed on any one of several computing nodes which possesses the necessary resources. In order to reflect the reality in job scheduling, allocation of network resources for data transfer should be taken into consideration. However, few scheduling methods consider the communication contention on Lambda Grids. In this paper, we investigate the joint scheduling problem while considering both optical network and computing resources in a Lambda Grid network. The objective of our work is to maximize the total number of jobs that can be scheduled in a Lambda Grid network. An adaptive routing algorithm is proposed and implemented for accomplishing the communication tasks for every job submitted in the network. Four heuristics (FIFO, ESTF, LJF, RS) are implemented for job scheduling of the computational tasks. Simulation results prove the feasibility and efficiency of the proposed solution.

A Tabu Search Approach for Joint Scheduling of Resources in a Lambda Grid Network

“…An efficient greedy algorithm that extends work done in Lakshmiraman et al [4] is proposed. We have modified the objective function of the algorithm to schedule resources such that the profit is maximized and completion time is minimized.…”

“…Since we expect this to be NP-Hard, we propose a Tabu search heuristic for joint optimization in a lambda grid network. We compare our results to a greedy approach that was implemented by extending the classic list-scheduling algorithm in Lakshmiraman et al [4]. We assume that the grid resources are connected by a wavelength-routed optical network with wavelength converters, so there is no wavelength continuity constraint.…”

“…In [5], the authors devised an efficient algorithm for minimizing the resource usage subject to job's deadline constraint assuming that both edge node and computing node have unlimited buffer to store data. Lakshmiraman et al [4] defined a joint scheduling problem in a homogeneous network to maximize job acceptance rate and minimize total scheduling time assuming wavelength continuity constraint. In [6], the authors also worked on a problem of joint scheduling of computing and network resources for a Directed Acyclic Graph (DAG) assuming a sub-wavelength granularity.…”