Dynamic Task Scheduling and Load Balancing on Cell Processors

Abstract

The shift to multicore processors demands efficient parallel programming on a diversity of architectures, including homogeneous and heterogeneous chip multiprocessors (CMPs). Task parallel programming is one approach that maps well to CMPs. In this model, the programmer focuses on identifying parallel tasks within an application, while a runtime system takes care of managing, scheduling, and balancing the tasks among a number of processors or cores. Heterogeneous CMPs, such as the Cell Broadband Engine, present new challenges to task parallel programming and corresponding runtime systems. In this paper, we present a library based on task pools for dynamic task scheduling and load balancing on Cell processors. In contrast to other approaches, our task pools include support for creating tasks using the Synergistic Processing Elements (SPEs), which enables the implementation of a wide range of task parallel applications. Our experiments show that task pools provide flexible and efficient support for task parallel programming on Cell processors. In addition, we show that offloading the process of task creation from the PPE to the SPEs provides much potential for exploiting fine-grained parallelism.

A novel task scheduling algorithm with improved make span based on prediction of tasks computation time algorithm for cloud computing

“…As talked about in Section 2, there exists an assortment of heuristic planning calculations, which can work in both group and online modes. A portion of these plans are fitting in heterogeneous booking situations, anyway they can't generally accomplish great make span, speedup, decreased expenses and expanded productivity [6,7,8,1,13] . Henceforth, QoS-based procedures are basic in getting the most extreme targets in order to hold QoS qualities for the two errands and assets.…”

A Task Scheduling Algorithm With Improved Makespan Based on Prediction of Tasks Computation Time algorithm for Cloud Computing

“…The former provides good approximate solutions with polynomial time complexity [14]. Dynamic scheduling is faster in execution than static scheduling, since it's basically not aware of any thread dependencies [7].…”

Enhancing the performance of process level redundancy with coprocessors in symmetric multiprocessors

“…Based on the fact that cores have different performance, threads will be dynamically scheduled to the core which will best fit the current demand. Automatically offloading computations tasks to coprocessors has been achieved on cell processors (Hoffmann et al, 2010), which automigrates programme procedure between main processors and coprocessors. The performance of applications is enhanced by the proposed method (Hoffmann et al, 2010).…”

“…Automatically offloading computations tasks to coprocessors has been achieved on cell processors (Hoffmann et al, 2010), which automigrates programme procedure between main processors and coprocessors. The performance of applications is enhanced by the proposed method (Hoffmann et al, 2010). Multiplexed redundant execution (MRE) (Subramanyan et al, 2010) is a hardware approach which partitions into three different pools of cores.…”