Please use this identifier to cite or link to this item:
https://dspace.iiti.ac.in/handle/123456789/4638
Title: | Comprehensive operation chaining based schedule delay estimation during high level synthesis |
Authors: | Mishra, Vipul Kumar Sengupta, Anirban |
Keywords: | Data flow analysis;Data flow graphs;Digital storage;Graphic methods;Information systems;Information use;Nanoelectronics;Scheduling;Control data flow graph (CDFG);Design space exploration;Functional units;Operation chaining;Resource Constraint;Storage elements;Switching devices;Unsolved problems;High level synthesis |
Issue Date: | 2018 |
Publisher: | Institute of Electrical and Electronics Engineers Inc. |
Citation: | Mishra, V. K., & Sengupta, A. (2018). Comprehensive operation chaining based schedule delay estimation during high level synthesis. Paper presented at the Proceedings - 2017 IEEE International Symposium on Nanoelectronic and Information Systems, iNIS 2017, , 2018-February 66-68. doi:10.1109/iNIS.2017.23 |
Abstract: | Design space exploration (DSE) during high level synthesis (HLS) involves a major step called scheduling which is responsible for estimating the delay of a control data flow graph (CDFG). However, a DSE process which concurrently estimates schedule delay by considering functional unit (FU), switching devices (such as mux, demux) and storage elements (such as latches), much before creation of its controller timing sequence, is an unsolved problem in the literature. Current DSE approaches either consider only FU during scheduling, or generate the complete controller timing sequence for delay evaluation of a CDFG based on provided resource constraint. The prior case, though fast but is not realistic in delay estimation. The latter case, though very slow, but provides realistic delay estimation. This paper solves the aforesaid problem by proposing a balanced DSE methodology that includes comprehensive delay estimation by considering combined delay of FU, switching devices and storage elements directly from scheduling. Results indicate improvement in achieving more realistic delay estimation process than previous approaches. © 2017 IEEE. |
URI: | https://doi.org/10.1109/iNIS.2017.23 https://dspace.iiti.ac.in/handle/123456789/4638 |
ISBN: | 9781538613566 |
Type of Material: | Conference Paper |
Appears in Collections: | Department of Computer Science and Engineering |
Files in This Item:
There are no files associated with this item.
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.
Altmetric Badge: