Online Task Scheduling for Heterogeneous Reconfigurable Systems

Authors:
Xuegong Zhou;Liang Liang;Ying Wang;Chenglian Peng
Affiliations:
State Key Lab of ASIC & System, Fudan University, Shanghai, China 200433;Department of Computing and Information Technology, Fudan University, Shanghai, China 200433;Department of Computing and Information Technology, Fudan University, Shanghai, China 200433;Department of Computing and Information Technology, Fudan University, Shanghai, China 200433
Venue:
Computer Supported Cooperative Work in Design IV
Year:
2008

Citing 7
Cited 0

Fast Template Placement for Reconfigurable Computing Systems

IEEE Design & Test
Fast Online Task Placement on FPGAs: Free Space Partitioning and 2D-Hashing

IPDPS '03 Proceedings of the 17th International Symposium on Parallel and Distributed Processing
Task scheduling for heterogeneous reconfigurable computers

SBCCI '04 Proceedings of the 17th symposium on Integrated circuits and system design
Programming Models for Hybrid FPGA-CPU Computational Components: A Missing Link

IEEE Micro
Online Scheduling for Block-Partitioned Reconfigurable Devices

DATE '03 Proceedings of the conference on Design, Automation and Test in Europe - Volume 1
Operating Systems for Reconfigurable Embedded Platforms: Online Scheduling of Real-Time Tasks

IEEE Transactions on Computers
Adaptive Allocation of Software and Hardware Real-Time Tasks for FPGA-based Embedded Systems

RTAS '06 Proceedings of the 12th IEEE Real-Time and Embedded Technology and Applications Symposium

Quantified Score

Hi-index	0.00

Visualization

Abstract

In heterogeneous reconfigurable systems including FPGA and CPU, besides software tasks and hardware tasks, there exist hybrid tasks which can be either executed on CPU as software tasks or implemented on FPGA as hardware tasks. This paper mainly focuses on the online scheduling problem within heterogeneous reconfigurable systems. According to a reconfigurable system prototype, a task scheduler model and its corresponding scheduling algorithm are presented. The model combines task allocation and task placement with task migration, making use of the great flexibility of hybrid tasks. Simulation results have demonstrated that the proposed algorithm can provide preferable scheduling performance and reduce the rejection rate.