A Realistic Network/Application Model for Scheduling Divisible Loads on Large-Scale Platforms

Abstract

Divisible load applications consist of an amount of data and associated computation that can be divided arbitrarily into any number of independent pieces. This model is a good approximation of many real-world scientific applications, lends itself to a natural master-worker implementation, and has thus received a lot of attention. The critical issue of divisible load scheduling has been studied extensively in previous work. However, only a few authors have explored the simultaneous scheduling of multiple such applications on a distributed computing platform. We focus on this increasingly relevant scenario and make the following contributions. We use a novel and more realistic platform model that captures some of the fundamental network properties of Grid platforms. We formulate a steady-state multi-application scheduling problem as a linear program that expresses some notion of fairness between applications. This scheduling problem is NP-complete and we propose several heuristics that we evaluate and compare via extensive simulation experiments conducted over 250,000 platform configurations. Our main finding is that some of our heuristics can achieve performance close to the optimal and we quantify the trade-offs between achieved performance and heuristic complexity.

Load Balancing in Optical Grids

“…This is acceptable in applications where data transfer time is negligible compared to jobs executions time. However, many e-science applications, such as linear algebra, image processing, and data mining [9], are becoming more data intensive. In [10], two adaptive and distributed load balancing algorithms considering transfer cost and network heterogeneity were introduced.…”

PPDD: scheduling multi-site divisible loads in single-level tree networks

“…Since the early days of DLT research, the research has spanned from addressing general optimal scheduling problems on different network topologies to various scenarios with practical constraints, such as time-varying channels [9], minimizing cost factors [10], resource management in Grid environments [11,12], and distributed image processing [13]. Thorough surveys of DLT can be found in [5,6,14,15].…”

“…Recently, [35] investigates the case of two load origination sources in a linear daisy chain architecture. The divisible load scheduling problems with multiple load sources in Grid environments have been studied in [11,36]. In this paper, we consider a general load scheduling and balancing problem with multiple loads originating from multiple processors in a single-level tree network.…”

“…After the communication of L m , the remaining load on processor p m−1 will be shared among all processors. At iteration i, p m−i+1 schedules its available load among all processors and let them finish processing of this load at the same time using (9) to (11). Note that at each iteration, we distribute the load in the order of link speed, faster link first.…”

Promoting cooperation in selfish computational grids

“…In the context of Grid computing, multi-criteria approaches may be used. Different criteria either express performance of different jobs [27] (centralized scheduling), or different factors, such as completion time and cost, of one job [22] (broker-based scheduling [6]). A scheduling algorithm is expected to deliver Pareto-optimal solutions.…”