Haluk Rahmi Topcuoglu

Performance-effective and low-complexity task scheduling for heterogeneous computing

Efficient application scheduling is critical for achieving high performance in heterogeneous computing environments. The application scheduling problem has been shown to be NP-complete in general cases as well as in several restricted cases. Because of its key importance, this problem has been extensively studied and various algorithms have been proposed in the literature which are mainly for systems with homogeneous processors. Although there are a few algorithms in the literature for heterogeneous processors, they usually require significantly high scheduling costs and they may not deliver good quality schedules with lower costs. In this paper, we present two novel scheduling algorithms for a bounded number of heterogeneous processors with an objective to simultaneously meet high performance and fast scheduling time, which are called the Heterogeneous Earliest-Finish-Time (HEFT) algorithm and the Critical-Path-on-a-Processor (CPOP) algorithm. The HEFT algorithm selects the task with the highest upward rank value at each step and assigns the selected task to the processor, which minimizes its earliest finish time with an insertion-based approach. On the other hand, the CPOP algorithm uses the summation of upward and downward rank values for prioritizing tasks. Another difference is in the processor selection phase, which schedules the critical tasks onto the processor that minimizes the total execution time of the critical tasks. In order to provide a robust and unbiased comparison with the related work, a parametric graph generator was designed to generate weighted directed acyclic graphs with various characteristics. The comparison study, based on both randomly generated graphs and the graphs of some real applications, shows that our scheduling algorithms significantly surpass previous approaches in terms of both quality and cost of schedules, which are mainly presented with schedule length ratio, speedup, frequency of best results, and average scheduling time metrics.

Task scheduling algorithms for heterogeneous processors

Scheduling computation tasks on processors is the key issue for high-performance computing. Although a large number of scheduling heuristics have been presented in the literature, most of them target only homogeneous resources. The existing algorithms for heterogeneous domains are not generally efficient because of their high complexity and/or the quality of the results. We present two low-complexity efficient heuristics, the Heterogeneous Earliest-Finish-Time (HEFT) algorithm and the Critical-Path-on-a-Processor (CPOP) algorithm for scheduling directed acyclic weighted task graphs (DAGs) on a bounded number of heterogeneous processors. We compared the performances of these algorithms against three previously proposed heuristics. The comparison study showed that our algorithms outperform previous approaches in terms of performance (schedule length ratio and speedup) and cost (time-complexity).

Solving the uncapacitated hub location problem using genetic algorithms

Hub location problems are widely studied in the area of location theory, where they involve locating the hub facilities and designing the hub networks. In this paper, we present a new and robust solution based on a genetic search framework for the uncapacitated single allocation hub location problem (USAHLP). To present its effectiveness, we compare the solutions of our GA-based method with the best solutions presented in the literature by considering various problem sizes of the CAB data set and the AP data set. The experimental work demonstrates that even for larger problems the results of our method significantly surpass those of the related work with respect to both solution quality and the CPU time to obtain a solution. Specifically, the results from our method match the optimal solutions found in the literature for all test cases generated from the CAB data set with significantly less running time than the related work. For the AP data set, our solutions match the best solutions of the reference study with an average of 8 times less running time than the reference study. Its performance, robustness and substantially low computational effort justify the potential of our method for solving larger problem sizes.

3-D path planning for the navigation of unmanned aerial vehicles by using evolutionary algorithms

Military missions are turning to more complicated and advanced automation technology for maximum endurance and efficiency as well as the minimum vital risks. The path planners which generate collision-free and optimized paths are needed to give autonomous operation capability to the Unmanned Aerial Vehicles (UAVs). This paper presents an off-line path planner for UAVs. The path planner is based on Evolutionary Algorithms (EA), in order to calculate a curved path line with desired attributes in a 3-D terrain. The flight path is represented by parameterized B-Spline curves by considering four objectives: the shortest path to the destination, the feasible path without terrain collision, the path with the desired minimum and maximum distance to the terrain, and the path which provides UAV to maneuver with an angle greater than the minimum radius of curvature. The generated path is represented with the coordinates of its control points being the genes of the chromosome of the EA. The proposed method was tested in several 3-D terrains, which are generated with various terrain generator methods that differ with respect to levels of smoothness of the terrain.

The software architecture of a virtual distributed computing environment

The requirements of grand challenge problems and the deployment of gigabit networks makes the network computing framework an attractive and cost effective computing environment with which to interconnect geographically distributed processing and storage resources. Our project, Virtual Distributed Computing Environment (VDCE), provides a problem-solving environment for high-performance distributed computing over wide area networks. VDCE delivers well-defined library functions that relieve end-users of tedious task implementations and also support reusability. In this paper we present the conceptual design of VDCE software architecture, which is defined in three modules: (a) the Application Editor, a user-friendly application development environment that generates the Application Flow Graph (AFG) of an application; (b) the Application Scheduler, which provides an efficient task-to-resource mapping of AFG; and (c) the VDCE Runtime System, which is responsible for running and managing application execution and monitoring the VDCE resources.

Positioning and Utilizing Sensors on a 3-D Terrain Part I—Theory and Modeling

Positioning multiple sensors for acquisition of a given environment is one of the fundamental research areas in various fields, such as military scouting, computer vision, and robotics. In this paper, we propose a new model for the problem of sensor deployment. Deploying and configuring a set of given sensors on a synthetically generated 3-D terrain have multiple objectives on conflicting attributes: maximizing the visibility of the given terrain, maximizing the stealth of the sensors, and minimizing the cost of the sensors used. Since they are utility-independent, these complementary and conflicting objectives are modeled by a multiplicative total utility function, based on multiattribute utility theory. The total utility function proposed in this paper can also be adapted for various military scouting missions with different characteristics.

Performance evaluation of evolutionary heuristics in dynamic environments

In recent years, there has been a growing interest in applying genetic algorithms to dynamic optimization problems. In this study, we present an extensive performance evaluation and comparison of 13 leading evolutionary algorithms with different characteristics on a common platform by using the moving peaks benchmark and by varying a set of problem parameters including shift length, change frequency, correlation value and number of peaks in the landscape. In order to compare solution quality or the efficiency of algorithms, the results are reported in terms of both offline error metric and dissimilarity factor, our novel comparison metric presented in this paper, which is based on signal similarity. Computational effort of each algorithm is reported in terms of average number of fitness evaluations and the average execution time. Our experimental evaluation indicates that the hybrid methods outperform the related work with respect to quality of solutions for various parameters of the given benchmark problem. Specifically, hybrid methods provide up to 24% improvement with respect to offline error and up to 30% improvement with respect to dissimilarity factor by requiring more computational effort than other methods.

A hyper-heuristic based framework for dynamic optimization problems

Abstract Most of the real world problems have dynamic characteristics, where one or more elements of the underlying model for a given problem including the objective, constraints or even environmental parameters may change over time. Hyper-heuristics are problem-independent meta-heuristic techniques that are automating the process of selecting and generating multiple low-level heuristics to solve static combinatorial optimization problems. In this paper, we present a novel hybrid strategy for applicability of hyper-heuristic techniques on dynamic environments by integrating them with the memory/search algorithm. The memory/search algorithm is an important evolutionary technique that have applied on various dynamic optimization problems. We validate performance of our method by considering both the dynamic generalized assignment problem and the moving peaks benchmark. The former problem is extended from the generalized assignment problem by changing resource consumptions, capacity constraints and costs of jobs over time; and the latter one is a well-known synthetic problem that generates and updates a multidimensional landscape consisting of several peaks. Experimental evaluation performed on various instances of the given two problems validates that our hyper-heuristic integrated framework significantly outperforms the memory/search algorithm.

Positioning and Utilizing Sensors on a 3-D Terrain Part II—Solving With a Hybrid Evolutionary Algorithm

In this paper, we explore using a hybrid evolutionary algorithm (HEA) for deploying and configuring a set of given sensors on a synthetically generated 3-D terrain. In our evolutionary-algorithm (EA) based solution, various methods are considered in order to incorporate specialized operators for hybridization, including problem-specific heuristics for initial population generation, intelligent variation operators (contribution-based-crossover operator and proximity-based-crossover operator), which comprise problem-specific knowledge, and a local-search phase. The experimental study validates finding the optimal balance among visibility-oriented, stealth-oriented, and cost-oriented objectives. The obtained results also indicate the effectiveness and robustness of our HEA-based solution for various practical scenarios with different objectives.

Solving the Register Allocation Problem for Embedded Systems Using a Hybrid Evolutionary Algorithm

Embedded systems are unique in the challenges they present to application programmers, such as power and memory space constraints. These characteristics make it imperative to design customized compiler passes. One of the important factors that shape runtime performance of a given embedded code is the register allocation phase of compilation. It is crucial to provide aggressive and sophisticated register allocators for embedded devices, where the excessive compilation time can be tolerated due to high demand on code quality. Failing to do a good job on allocating variables to registers (i.e., determining the set of variables to be stored in the limited number of registers) can have serious power, performance, and code size consequences. This paper explores the possibility of employing a hybrid evolutionary algorithm for register allocation problem in embedded systems. The proposed solution combines genetic algorithms with a local search technique. The algorithm exploits a novel, highly specialized crossover operator that takes into account domain-specific information. The results from our implementation based on synthetic benchmarks and routines that are extracted from well-known benchmark suites clearly show that the proposed approach is very successful in allocating registers to variables. In addition, our experimental evaluation also indicates that it outperforms a state-of-the-art register allocation heuristic based on graph coloring for most of the cases experimented.