A Dynamic Power-Aware Partitioner with Task Migration for Multicore Embedded Systems

Abstract

Abstract. Nowadays, a key design issue in embedded systems is how to reduce the power consumption, since batteries have a limited energy budget. For this purpose, several techniques such as Dynamic Voltage Scaling (DVS) or task migration can be used. DVS allows reducing power by selecting the optimal voltage supply, while task migration achieves this effect by balancing the workload among cores. This paper first analyzes the impact on energy due to task migration in multicore embedded systems with DVS capability and using the well-known Worst Fit (WF) partitioning heuristic. To reduce overhead, migrations are only performed at the time that a task arrives to and/or leaves the system and, in such a case, only one migration is allowed. The huge potential on energy saving due to task migration, leads us to propose a new dynamic partitioner, namely DP, that migrates tasks in a more efficient way than typical partitioners. Unlike WF, the proposed algorithm examines which is the optimal target core before allowing a migration. Experimental results show that DP can improve energy consumption in a factor up to 2.74 over the typical WF algorithm.

Load Balancing in Multicore Systems using Heuristics Based Approach

“…The main contribution of their work is that the assignment of tasks to the cores is done without sorting the tasks in decreasing order of their utilization which may reduce the number of migrations and further may reduce the migration over head of the system. The number of cores considered by March et al [21] for experimental setup was very less as compared to present use of large numbers of cores in multicore processors.…”

“…Many researchers have proposed different strategies to schedule the tasks and balance the workload of the system considering different areas of application . Some application areas are load balancing in parallel computing [4], fog computing [5], cloud computing [6,7] and real time environment [8][9][10][19][20][21][22][23][24]. An efficient load balancing real time algorithm is one in which the tasks are schedulable, their deadlines are met and the load is balanced after task assignment such that the resources are effectively used with no overload or under load constraints.…”

“…If the number of cores in sleep mode is more than the required and increase in frequency does not result in effectiveness, this means that the cores' utilization is wasted and it is not beneficial to use large number of cores. It also uses the fact that a task once migrated to a foreign core can be further migrated to [21] proposed two algorithms-Single Option Migration (SOM) and Multiple Option Migration (MOM). They allowed task migration, to reduce energy consumption in multicore embedded systems with realtime constraints, by implementing Dynamic Voltage Frequency Scaling (DVFS) capabilities.…”

“…Consider another task set 1 , 2 , 3 with utilization as 1 = 0.33, 2 = 0.25 and 3 = 0.66. The tasks are assigned to the least utilized cores [19,21] and it can be seen that the cores suffers imbalance amongst Copyright © 2018 MECS I.J. Intelligent Systems and Applications, 2018, 12, 56-68 them as explained in Fig.…”

“…= 0.2, 3 = 0.2, 4 = 0.1 and 5 = 0.1. The tasks are sorted in decreasing order of their utilization and are assigned to the first available core[21].…”