Dynamic Data Allocation and Task Scheduling on Multiprocessor Systems With NVM-Based SPM

Abstract

Low-power and short-latency memory access is critical to the performance of chip multiprocessor (CMP) system devices, especially to bridge the performance gap between memory and CPU. Together with increased demand for low-energy consumption and high-speed memory, scratch-pad memory (SPM) has been widely adopted in multiprocessor systems. In this paper, we employ a hybrid SPM, composed of a static random-access memory and a nonvolatile memory (NVM), to replace the cache in CMP. However, there are several challenges related to the CMP that need to be addressed, including how to dynamically assign processors to application tasks and dynamically allocate data to memories. To solve these problems based on this architecture, we propose a novel dynamic data allocation and task scheduling algorithm, i.e., dynamic greedy data allocation and task scheduling (DGDATS). Experiments on DSP benchmarks demonstrate the effectiveness and efficiency of our proposed algorithms; namely, our proposed algorithm can generate a highly efficient dynamic data allocation and task scheduling approach to minimize the total execution cost and produce the least amount of write operations on NVMs. Our extensive simulation study demonstrates that our proposed algorithm exhibits an excellent performance compared with the heuristic allocation (HA) and adaptive genetic algorithm for data allocation (AGADA) algorithms. Based on the CMP systems with hybrid SPMs, DGDATS reduces the total execution cost by 22.18% and 51.37% compared with those of the HA and AGADA algorithms, respectively. Additionally, the average number of write operations on NVM is 19.82% lower than that of HA.

Task scheduling on heterogeneous multiprocessor systems through coherent data allocation

“…In recent years, the research of joint consideration of TSDA has received lots of attention (see Table 2). For homogeneous architecture situations, there have been many studies on TSDA 9,10,39,40,42,45 . Wang et al 9 investigated the energy optimization for TSDA problem with hybrid SRAM+NVM SPM.…”

“…In a heterogeneous distributed shared‐memory multiprocessor system (HDSMS), the processing time and energy consumption of the same task on different processors are different, and access time and energy consumption of transmitting the same data between processors and memories are also different 4,5 . References 4,5 studied TSDA on a HDSMS, and they have been received extensive attention 8‐14 . In this work, we also investigate the TSDA problem based on HDSMS.…”

“…First, the strategy of assigning data or scheduling tasks based solely on the energy‐time ratio as an indicator, ignoring the data allocation or task scheduling that minimizes energy consumption, may not be able to obtain an approximate optimal solution. Second, some studies consider TSDA separately, 15,16 and some studies consider TSDA on other architecture models; 5,9,10 however, their performance still needs to be improved on our architecture. Therefore, our research on data allocation and task scheduling on this architecture is reasonable and necessary.…”

A Balancing Method of Mixed-model Disassembly Line in Random Working Environment

“…Although GA has strong global optimization ability, it has insufficient local optimization ability when solving the NP problem. Therefore, in the basic genetic algorithm, the mutation operator and crossover operator are changed from the original fixed value to evolve with the change of algebra, which enables the algorithm to expand the search scope and enhance the global search ability in the early stage of optimization, and avoid the destruction of the optimal sequence and miss the optimal solution due to the high crossover and mutation probability in the later stage of the algorithm [29]. Moreover, the results obtained by GA are substituted into SA as the initial solution of SA, making full use of the advantages of GS and SA to make ASAGA have both the ability of rapid convergence, global search and getting rid of local optimal, so as to enhance the solving performance of the algorithm.…”

Energy-Aware Task Scheduling on Heterogeneous Computing Systems With Time Constraint

“…It is difficult to find an effective way to solve the above problems, because a processor has several voltage settings, and the same task has different processing times and energy consumption levels when executing on different processors. Traditional scheduling studies focus on heuristic-based algorithms, which are often based on greedy local optimal selection for some heuristic strategies [3], [12], [13]. However, due to the greedy nature, heuristic-based methods can not always produce consistent results for different problem instances [3], [12].…”