Search

Cache-partitioned architectures allow subsections of the shared last-level cache (LLC) to be exclusively reserved for some applications. This technique dramatically limits interactions between applications that are concurrently executing on a multicore machine. Consider n applications that execute concurrently, with the objective to minimize the makespan, defined as the maximum completion time of the n applications. Key scheduling questions are as follows: (i) which proportion of cache and (ii) how many processors should be given to each application? In this article, we provide answers to (i) and (ii) for Amdahl applications. Even though the problem is shown to be NP-complete, we give key elements to determine the subset of applications that should share the LLC (while remaining ones only use their smaller private cache). Building upon these results, we design efficient heuristics for Amdahl applications. Extensive simulations demonstrate the usefulness of co-scheduling when our efficient cache partitioning strategies are deployed. [ABSTRACT FROM AUTHOR]

Recently, the benefits of co-scheduling several applications have been demonstrated in a fault-free context, both in terms of performance and energy savings. However, large-scale computer systems are confronted by frequent failures, and resilience techniques must be employed for large applications to execute efficiently. Indeed, failures may create severe imbalance between applications and significantly degrade performance. In this article, we aim at minimizing the expected completion time of a set of co-scheduled applications. We propose to redistribute the resources assigned to each application upon the occurrence of failures, and upon the completion of some applications, in order to achieve this goal. First, we introduce a formal model and establish complexity results. The problem is NP-complete for malleable applications, even in a fault-free context. Therefore, we design polynomial-time heuristics that perform redistributions and account for processor failures. A fault simulator is used to perform extensive simulations that demonstrate the usefulness of redistribution and the performance of the proposed heuristics. [ABSTRACT FROM AUTHOR]

This paper investigates co-scheduling algorithms for processing a set of parallel applications. Instead of executing each application one by one, using a maximum degree of parallelism for each of them, we aim at scheduling several applications concurrently. We partition the original application set into a series of packs, which are executed one by one. A pack comprises several applications, each of them with an assigned number of processors, with the constraint that the total number of processors assigned within a pack does not exceed the maximum number of available processors. The objective is to determine a partition into packs, and an assignment of processors to applications, that minimize the sum of the execution times of the packs. We thoroughly study the complexity of this optimization problem, and propose several heuristics that exhibit very good performance on a variety of workloads, whose application execution times model profiles of parallel scientific codes. We show that co-scheduling leads to faster workload completion time (40 % improvement on average over traditional scheduling) and to faster response times (50 % improvement). Hence, co-scheduling increases system throughput and saves energy, leading to significant benefits from both the user and system perspectives. [ABSTRACT FROM AUTHOR]

A large class of applications need to execute the same workflow on different datasets of identical size. Efficient execution of such applications necessitates intelligent distribution of the application components and tasks on a parallel machine, and the execution can be orchestrated by utilizing task, data, pipelined, and/or replicated parallelism. The scheduling problem that encompasses all of these techniques is called pipelined workflow scheduling, and it has been widely studied in the last decade. Multiple models and algorithms have flourished to tackle various programming paradigms, constraints, machine behaviors, or optimization goals. This article surveys the field by summing up and structuring known results and approaches. [ABSTRACT FROM AUTHOR]

Multiple applications that execute concurrently on heterogeneous platforms compete for CPU and network resources. In this paper, we consider the problem of scheduling applications to ensure fair and efficient execution on a distributed network of processors. We limit our study to the case where communication is restricted to a tree embedded in the network, and the applications consist of a large number of independent tasks (Bags of Tasks) that originate at the tree's root. The tasks of a given application all have the same computation and communication requirements, but these requirements can vary for different applications. The goal of scheduling is to maximize the throughput of each application while ensuring a fair sharing of resources between applications. We can find the optimal asymptotic rates by solving a linear programming problem that expresses all necessary problem constraints, and we show how to construct a periodic schedule from any linear program solution. For single-level trees, the solution is characterized by processing tasks with larger communication-to-computation ratios at children with larger bandwidths. For multilevel trees, this approach requires global knowledge of all application and platform parameters. For large-scale platforms, such global coordination by a centralized scheduler may be unrealistic. Thus, we also investigate decentralized schedulers that use only local information at each participating resource. We assess their performance via simulation and compare to an optimal centralized solution obtained via linear programming. The best of our decentralized heuristics achieves the same performance on about 2/3 of our test cases but is far worse in a few cases. Although our results are based on simple assumptions and do not explore all parameters (such as the maximum number of tasks that can be held on a node), they provide insight into the important question of fairly and optimally scheduling heterogeneous applications on heterogeneous grids. [ABSTRACT FROM AUTHOR]

Abstract: In this paper, we assess the impact of heterogeneity on scheduling independent tasks on master-slave platforms. We assume a realistic one-port model where the master can communicate with a single slave at any time. We target both on-line and off-line scheduling problems, and we focus on simpler instances where all tasks have the same size. While such on-line problems can be solved in polynomial time on homogeneous platforms, we show that there does not exist any optimal deterministic algorithm for heterogeneous platforms. Whether the source of heterogeneity comes from computation speeds, or from communication bandwidths, or from both, we establish lower bounds on the competitive ratio of any deterministic algorithm. We provide such bounds for the most important objective functions: the minimization of the makespan (or total execution time), the minimization of the maximum response time (difference between completion time and release time), and the minimization of the sum of all response times. Altogether, we obtain nine theorems which nicely assess the impact of heterogeneity on on-line scheduling. For off-line scheduling, we prove several result for problems with release dates, either optimality or NP-hardness. These theoretical contributions are complemented on the practical side by the implementation of several heuristics on a small but fully heterogeneous MPI platform. Our results show the superiority of those heuristics which fully take into account the relative capacity of the communication links. [Copyright &y& Elsevier]

This paper considers steady-state scheduling techniques for heterogeneous systems, such as clusters and grids. The use of steady-state scheduling is advocated to solve a variety of important problems, which would be too difficult to tackle with the objective of makespan minimization. Several examples are given, namely master-slave tasking, mixed task and data parallelism, and pipelined macro-communications (scatter, broadcast, multicast). For each example, both the advantages and the limitations of the approach are discussed. [ABSTRACT FROM AUTHOR]

The article discusses various reports published within the issue, including the data redistribution for parallel applications by E. Jeannot and F. Wagner, the high-throughput computing with donated compute resources by R. Trivedi, A. Chandra and J. Weissman, and the topic of L. Eyraud on the gap between the theory job scheduling and its practice.

The article briefly discusses research papers that are contained within the section, titled "Algorithm Design and Scheduling Techniques (Realistic Platform Models) for Heterogeneous Clusters," of this issue. The purpose of these papers is to explore the impact of platform heterogeneity on the design and evaluation of new algorithms, such as, distributed algorithms and scheduling algorithms. One paper by Ligang He and his colleagues discusses two new workload allocation strategies for large-scale heterogeneous platforms. The paper "On the Distribution of Sequential Jobs in Random Brokering for Heterogeneous Computational Grids," by Vandy Berten and colleagues, focuses on resource brokering for scheduling sequential jobs onto a grid platform. In their article, "Multiple Job Scheduling in a Connection-Limited Data Parallel System," authors Alessandro Amoroso and Keith Marzullo describe a new method for scheduling jobs in a distributed system.

This paper focuses on energy minimization for the mapping and scheduling of real-time workflows under reliability constraints. Workflow instances are input periodically to the system. Each instance is composed of several tasks and must complete execution before the arrival of the next instance, and with a prescribed reliability threshold. While the shape of the dependence graph is identical for each instance, task execution times are stochastic and vary from one instance to the next. The reliability threshold is met by executing several replicas for each task. The target platform consists of identical processors equipped with Dynamic Voltage and Frequency Scaling (DVFS) capabilities. A different frequency can be assigned to each task replica to save energy, but it may have negative effect on the deadline and reliability target. This difficult tri-criteria mapping and scheduling problem (energy, deadline, reliability) has been studied only recently for workflows with arbitrary dependence constraints. We investigate new mapping and scheduling strategies based upon layers in the task graph. These strategies better balance replicas across processors, thereby decreasing the time overlap between different replicas of a given task, and saving energy. We compare these strategies with two state-of-the-art approaches and a reference baseline on a variety of benchmark workflows. Our best heuristics achieve an average energy gain of 60% over the competitors and of 82% over the baseline. • Tri-criteria mapping and scheduling problem for real-time workflows of arbitrary shape. • First general solution without any limitation on the number of replicas used. • Innovative solution based upon layers of the workflow outperforms competitors. [ABSTRACT FROM AUTHOR]

In this paper, we discuss several algorithms for scheduling divisible workloads on heterogeneous systems. Our main contributions are (i) new optimality results for single-round algorithms and (ii) the design of an asymptotically optimal multi-round algorithm. This multi-round algorithm automatically performs resource selection, a difficult task that was previously left to the user. Because it is periodic, it is simpler to implement, and more robust to changes in the speeds of the processors and/or communication links. On the theoretical side, to the best of our knowledge, this is the first published result assessing the absolute performance of a multi-round algorithm. On the practical side, extensive simulations reveal that our multi-round algorithm outperforms existing solutions on a large variety of platforms, especially when the communication-to-computation ratio is not very high (the difficult case). [Copyright &y& Elsevier]