Heuristic resource allocation algorithms for maximizing allowable workload in dynamic, distributed real-time systems

Abstract

This paper examines several heuristic algorithms for the maximum allowable workload (MAW) problem for realtime systems with tasks having variable workloads. Briefly, the problem concerns the allocation of n tasks to m processors, where each task t is characterized by a function t.r(w) that gives the running time of the task in terms of its workload (or input size) w. The objective of the maximum allowable workload problem is to find an allocation of tasks to processors so that the allocation is feasible (no task misses its deadline) when each task is given a workload of w or smaller and w is maximized. This optimization problem uses a robustness measure that is closely related to the MAIL (maximum allowable increase in load) metric recently proposed by Gertphol et. al. The main contribution of this paper is the the comparison of several heuristic algorithms for the MAW-RMS problem. Hillclimbing, random search, simulated annealing, and first-fit heuristics are presented and evaluated via simulation. As we show here, the first-fit greedy heuristic produces solutions of a reasonable quality compared to the other algorithms. In addition, we demonstrate the applicability of our model in air defense systems.

Robust Task Allocation for Dynamic Distributed Real-Time Systems Subject to Multiple Environmental Parameters

“…In the previous work on single external workload variable (MAW) [10], a first-fit based allocation algorithm was developed to find the allocation with maximal robustness. By imposing two well-behaved restrictions on execution time profile functions, Juedes further developed a conditional two component approximation bound and an asymptotic approximation ratio for this algorithm [11].…”

“…By imposing two well-behaved restrictions on execution time profile functions, Juedes further developed a conditional two component approximation bound and an asymptotic approximation ratio for this algorithm [11]. In this section, we will extend the definition of well-behaved profile functions and present an improved bound on the one dimensional robustness when this first-fit based algorithm is used.…”

“…Briefly, the first fit based algorithm [10] combined a binary search with a first fit allocation to find the allocation with maximal robustness. It used a robustness metric that was defined as the maximal workload value allowed by an allocation (MAW).…”

“…This understanding will help us design an efficient search for the maximal robustness value under multiple dimensions. Generally, an algorithm's running time does not decrease as its input size grows larger, thus we previously made the assumption that execution time functions are non-decreasing in one workload dimension [10]. When a profile function depends on multiple environmental variables, it is reasonable to assume the function also to be non-decreasing in each dimension while keeping others fixed, or mathematically…”

“…We previously studied a task allocation problem with a single workload variable for dynamic systems [10]. The goal was to find the maximal allowable workload (MAW) for a system of independent real-time tasks allocated to processors scheduled rate monotonically (RM).…”