| Author | Lalinthip Tangjittaweechai |
| Note | A thesis submitted in partial fulfillment of the requirements for the degree of Master of Engineering in Microelectronics and Embedded Systems, School of Engineering and Technology |
| Publisher | Asian Institute of Technology |
| Abstract | Inside IC chips, several millions of nets have to be routed to complete the circuit. There are many routing algorithms but most of algorithms are either the sequential approach or the concurrent approach running on CPU. On the other hand, GPU has many cores and can run many processes in parallel. As a result, this thesis proposed the routing algorithms that are performed on GPU in parallel to reduce the execution time and improve the routing quality of the circuit. This thesis proposed two algorithms which run on GPU. The first algorithm gets short total wire length and less execution time but there are higher overflows comparing with the CPU algorithm. For the second GPU algorithm, the algorithm gets more total wire length and slower comparing with the first GPU algorithm but it gets less total wire length and total overflows than the CPU algorithm. In the case of the execution time, the second GPU Algorithm is faster than the CPU algorithm in large benchmarks. For the small benchmarks in case of the execution time is quite the same. This thesis proposed GPU algorithm that can reduce the execution time and improve the routing quality |
| Year | 2011 |
| Type | Thesis |
| School | School of Engineering and Technology (SET) |
| Department | Department of Industrial Systems Engineering (DISE) |
| Academic Program/FoS | Microelectronics (ME) |
| Chairperson(s) | Mongkol Ekpanyapong ; |
| Examination Committee(s) | Dailey, Matthew N.;Luong, Huynh Trung; |
| Scholarship Donor(s) | Royal Thai Government Fellowship ; |