An Improved Greedy Based Global Optimized Placement Algorithm

An improved Greedy based global optimized Placement algorithm was proposed in this paper. Probability controlling was applied to avoid local minimization of traditional Greedy algorithm. Some new technique such as Marker bit method, dynamical selection and weight matrix were introduced to improve efficiency of this algorithm. From the experiments and comparison with Simulated Annealing algorithm, this novel algorithm shows better stability and accuracy.

[1]  Ranga Vemuri,et al.  An efficient algorithm for finding empty space for online FPGA placement , 2004, Proceedings. 41st Design Automation Conference, 2004..

[2]  Christian Müller-Schloer,et al.  Organic and Pervasive Computing – ARCS 2004 , 2004, Lecture Notes in Computer Science.

[3]  Majid Sarrafzadeh,et al.  Fast Template Placement for Reconfigurable Computing Systems , 2000, IEEE Des. Test Comput..

[4]  Jürgen Teich,et al.  A Dynamic Scheduling and Placement Algorithm for Reconfigurable Hardware , 2004, ARCS.

[5]  Ranga Vemuri,et al.  An Integrated Online Scheduling and Placement Methodology , 2004, FPL.

[6]  Zonghua Gu,et al.  An Efficient Algorithm for Online Management of 2D Area of Partially Reconfigurable FPGAs , 2007, 2007 Design, Automation & Test in Europe Conference & Exhibition.

[7]  Weichen Liu,et al.  An Efficient Algorithm for Online Soft Real-Time Task Placement on Reconfigurable Hardware Devices , 2007, 10th IEEE International Symposium on Object and Component-Oriented Real-Time Distributed Computing (ISORC'07).

[8]  Marco Platzner,et al.  Online scheduling and placement of real-time tasks to partially reconfigurable devices , 2003, RTSS 2003. 24th IEEE Real-Time Systems Symposium, 2003.

[9]  Marco Platzner,et al.  Fast online task placement on FPGAs: free space partitioning and 2D-hashing , 2003, Proceedings International Parallel and Distributed Processing Symposium.

[10]  Marco Platzner,et al.  Field Programmable Logic and Application , 2004, Lecture Notes in Computer Science.