Results 1  10
of
40
Benchmarking the task graph scheduling algorithms
 in IPPS/SPDP
, 1998
"... Abstract † The problem of scheduling a weighted directed acyclic graph (DAG) to a set of homogeneous processors to minimize the completion time has been extensively studied. The NPcompleteness of the problem has instigated researchers to propose a myriad of heuristic algorithms. While these algorith ..."
Abstract

Cited by 47 (2 self)
 Add to MetaCart
(Show Context)
Abstract † The problem of scheduling a weighted directed acyclic graph (DAG) to a set of homogeneous processors to minimize the completion time has been extensively studied. The NPcompleteness of the problem has instigated researchers to propose a myriad of heuristic algorithms. While these algorithms are individually reported to be efficient, it is not clear how effective they are and how well they compare against each other. A comprehensive performance evaluation and comparison of these algorithms entails addressing a number of difficult issues. One of the issues is that a large number of scheduling algorithms are based upon radically different assumptions, making their comparison on a unified basis a rather intricate task. Another issue is that there is no standard set of benchmarks that can be used to evaluate and compare these algorithms. Furthermore, most algorithms are evaluated using small problem sizes, and it is not clear how their performance scales with the problem size. In this paper, we first provide a taxonomy for classifying various algorithms into different categories according to their assumptions and functionalities. We then propose a set of benchmarks which are of diverse structures without being biased towards a particular scheduling technique and still allow variations in important parameters. We have evaluated 15 scheduling algorithms, and compared them using the proposed benchmarks. Based upon the design philosophies and principles behind these algorithms, we interpret the results and discuss why some algorithms perform better than the others.
Analysis, Evaluation, and Comparison of Algorithms for Scheduling Task Graphs on Parallel Processors
 In Proceedings of the Second International Symposium on Parallel Architectures, Algorithms, and Networks
, 1996
"... Abstract 1 In this paper, we survey algorithms that allocate a parallel program represented by an edgeweighted directed acyclic graph (DAG), also called a task graph or macrodataflow graph, to a set of homogeneous processors, with the objective of minimizing the completion time. We analyze 21 such ..."
Abstract

Cited by 38 (5 self)
 Add to MetaCart
(Show Context)
Abstract 1 In this paper, we survey algorithms that allocate a parallel program represented by an edgeweighted directed acyclic graph (DAG), also called a task graph or macrodataflow graph, to a set of homogeneous processors, with the objective of minimizing the completion time. We analyze 21 such algorithms and classify them into four groups. The first group includes algorithms that schedule the DAG to a bounded number of processors directly. These algorithms are called the bounded number of processors (BNP) scheduling algorithms. The algorithms in the second group schedule the DAG to an unbounded number of clusters and are called the unbounded number of clusters (UNC) scheduling algorithms. The algorithms in the third group schedule the DAG using task duplication and are called the task duplication based (TDB) scheduling algorithms. The algorithms in the fourth group perform allocation and mapping on arbitrary processor network topologies. These algorithms are called the arbitrary processor network (APN) scheduling algorithms. The design philosophies and principles behind these algorithms are discussed, and the performance of all of the algorithms is evaluated and compared against each other on a unified basis by using various scheduling parameters.
A Task Duplication Based Scheduling Algorithm for Heterogeneous Systems
 in Proc. of 14th International Parallel and Distributed Processing Symposium (IPDPS'00
, 2000
"... Optimal scheduling of tasks of a directed acyclic graph (DAG) onto a set of processors is a strong NPhard problem. In this paper we present a scheduling scheme called TDS to schedule tasks of a DAG onto a heterogeneous system. This models a network of workstations, with processors of varying comput ..."
Abstract

Cited by 31 (0 self)
 Add to MetaCart
Optimal scheduling of tasks of a directed acyclic graph (DAG) onto a set of processors is a strong NPhard problem. In this paper we present a scheduling scheme called TDS to schedule tasks of a DAG onto a heterogeneous system. This models a network of workstations, with processors of varying computing power. The primary objective of this scheme is to minimize schedule length and scheduling time itself. The existing task duplication based scheduling scheme is primarily done for totally homogeneous systems. We compare the performance of this algorithm with an existing scheduling scheme for heterogeneous processors called BIL. In initial simulations TDS has been observed to generate scheduling lengths shorter than that of BIL, for communicationtocomputation cost ratios (CCR) of 0.2 to 1. Moreover TDS is far more superior than BIL as far as scheduling time is concerned. 1.
Scheduling workflow applications on processors with different capabilities
, 2006
"... Efficient scheduling of workflow applications represented by weighted directed acyclic graphs (DAG) on a set of heterogeneous processors is essential for achieving high performance. The optimization problem is NPcomplete in general. A few heuristics for scheduling on heterogeneous systems have been ..."
Abstract

Cited by 26 (1 self)
 Add to MetaCart
Efficient scheduling of workflow applications represented by weighted directed acyclic graphs (DAG) on a set of heterogeneous processors is essential for achieving high performance. The optimization problem is NPcomplete in general. A few heuristics for scheduling on heterogeneous systems have been proposed recently. However, few of them consider the case where processors have different capabilities. In this paper, we present a novel list scheduling based algorithm to deal with this situation. The algorithm (SDC) has two distinctive features. First, the algorithm takes into account the effect of Percentage of Capable Processors (PCP) when assigning the task node weights. For two task nodes with same average computation cost, our weight assignment policy tends to give higher weight to the task with small PCP. Secondly, during the processor selection phase, the algorithm adjusts the effective Earliest Finish Time strategy by incorporating the average communication cost between the current scheduling node and its children. Comparison study shows that our algorithm performs better than related work overall.
Fast and Effective Task Scheduling in Heterogeneous Systems
, 2000
"... Recently, we presented two very lowcost approaches to compiletime list scheduling where the tasks' priorities are computed statically or dynamically, respectively. For homogeneous systems, these two algorithms, called FCP and FLB, have shown to yield a performance equivalent to other much mor ..."
Abstract

Cited by 24 (0 self)
 Add to MetaCart
Recently, we presented two very lowcost approaches to compiletime list scheduling where the tasks' priorities are computed statically or dynamically, respectively. For homogeneous systems, these two algorithms, called FCP and FLB, have shown to yield a performance equivalent to other much more costly algorithms such as MCP and ETF. In this paper we present modified versions of FCP and FLB targeted to heterogeneous systems. We show that the modified versions yield a good overall performance, which is generally comparable to algorithms specifically designed for heterogeneous systems, such as HEFT or ERT. There are a few cases, mainly for irregular problems and large processor speed variance, where FCP and FLB's performance drops down to ### and ###, respectively. Considering the good overall performance and their very low cost however, FCP and FLB are interesting options for scheduling very large problems on heterogeneous systems.
On the Complexity of List Scheduling Algorithms for DistributedMemory Systems
 ICS
, 1999
"... This paper presents a novel heuristic, called Fast Critical Path (FCP), intended as a compiletime scheduling algorithm on a distributedmemory system. While similar to existing list scheduling algorithms, FCP has two important differences: (a) it does not sort all the tasks at the beginning but mai ..."
Abstract

Cited by 19 (5 self)
 Add to MetaCart
This paper presents a novel heuristic, called Fast Critical Path (FCP), intended as a compiletime scheduling algorithm on a distributedmemory system. While similar to existing list scheduling algorithms, FCP has two important differences: (a) it does not sort all the tasks at the beginning but maintains only a limited number of tasks sorted at any given time, and (b) instead of considering all processors as possible target for a given task, the choice is restricted to either the processor from which the last messages to the given task arrives or the processor which becomes idle the earliest. As a result, the time complexity is drastically reduced to O(V log (P ) +E), where V and E are the number of tasks and edges in the task graph, respectively, and P is the number of processors. We demonstrate through theory and experiments that FCP performs equally to existing list scheduling algorithms of much higher complexity.
DFRN: A New Approach for Duplication Based Scheduling for Distributed Memory Multiprocessor Systems
 In Proc. of the IPPS
, 1997
"... Duplication Based Scheduling (DBS) is a relatively new approach for solving multiprocessor scheduling problems. The problem is defined as finding an optimal schedule which minimizes the parallel execution time of an application on a target system. In this paper, we classify DBS algorithms into two c ..."
Abstract

Cited by 17 (0 self)
 Add to MetaCart
(Show Context)
Duplication Based Scheduling (DBS) is a relatively new approach for solving multiprocessor scheduling problems. The problem is defined as finding an optimal schedule which minimizes the parallel execution time of an application on a target system. In this paper, we classify DBS algorithms into two categories according to the task duplication method used. We then present our new DBS algorithm that extracts the strong features of the two categories of DBS algorithms. Our simulation study shows that the proposed algorithm achieves considerable performance improvement over existing DBS algorithms with equal or less time complexity. We analytically obtain the boundary condition for the worst case behavior of the proposed algorithm and also prove that the algorithm generates an optimal schedule for a tree structured input directed acyclic graph. 1.
Robust task scheduling in nondeterministic heterogeneous computing systems
 in Proc. IEEE Intl. Conf. Cluster Computing
, 2006
"... The paper addresses the problem of matching and scheduling of DAGstructured application to both minimize the makespan and maximize the robustness in a heterogeneous computing system. Due to the conflict of the two objectives, it is usually impossible to achieve both goals at the same time. We give ..."
Abstract

Cited by 14 (4 self)
 Add to MetaCart
(Show Context)
The paper addresses the problem of matching and scheduling of DAGstructured application to both minimize the makespan and maximize the robustness in a heterogeneous computing system. Due to the conflict of the two objectives, it is usually impossible to achieve both goals at the same time. We give two definitions of robustness of a schedule based on tardiness and miss rate. Slack is proved to be an effective metric to be used to adjust the robustness. We employ ǫconstraint method to solve the biobjective optimization problem where minimizing the makespan and maximizing the slack are the two objectives. Overall performance of a schedule considering both makespan and robustness is defined such that user have the flexibility to put emphasis on either objective. Experiment results are presented to validate the performance of the proposed algorithm.
GAST: A Flexible and Extensible Tool for Evaluating Multiprocessor Assignment and Scheduling Techniques
 Proc. of the Int'l Conf. on Parallel Processing
, 1998
"... Automatic tool support for scheduling applications on multiprocessor platforms is of paramount importance both to guarantee critical application demands and to keep development costs down. In this paper, we present GAST, an objectoriented evaluation environment for multiprocessor assignment and sche ..."
Abstract

Cited by 11 (7 self)
 Add to MetaCart
Automatic tool support for scheduling applications on multiprocessor platforms is of paramount importance both to guarantee critical application demands and to keep development costs down. In this paper, we present GAST, an objectoriented evaluation environment for multiprocessor assignment and scheduling techniques. GAST is based on a decomposition approach where existing assignment and scheduling techniques have been broken down into a set of common operations. By combining these operations into a complete scheduling process, GAST offers a powerful toolbox of existing assignment and scheduling techniques. GAST has been implemented on several popular operating systems and is accompanied by a graphic visualization tool that allows for flexible interaction with the application scheduling process. 1 Introduction For economical reasons, singleprocessor computers have traditionally been the only viable option for many computer system designers. This has, in turn, limited the sphere of f...