290 likes | 322 Views
A Combined Clustering and Placement Algorithm for FPGAs. Mark Yamashita. Contributions. New algorithm to do clustering and placement Novel approach for trading-off depth for duplication control Timing model/placement incorporated into clustering Delay improves by an average of 11%
E N D
A Combined Clustering and Placement Algorithm for FPGAs Mark Yamashita
Contributions • New algorithm to do clustering and placement • Novel approach for trading-off depth for duplication control • Timing model/placement incorporated into clustering • Delay improves by an average of 11% • Controllable trade-off between area overhead and delay improvements • Plan to submit to FPL ‘08
Motivation • FPGAs need to be faster • 4x slower than ASICs • Limitations of existing clustering approaches: • No depth control during clustering, often greedy • Provide no means for duplication, or • Use duplication in excess • Inaccurate timing models
Motivation • GOAL: • Improve critical-path delay by improving clustering • Approach: • Use placement information to form accurate timing model • Make better clustering decisions • Use duplication to reduce depth • Take advantage of otherwise unused logic in FPGA • Control amount of duplication by relaxing depth
Algorithm Overview T-VP
Phase 1: Additional Duplication Reduction Through Depth Relaxation
Algorithm Overview T-VP
Phase 2: Microcluster Compaction with Orchestrator • Iteratively move microclusters to improve timing • Can fit multiple microclusters to the same CLB position, provided the aggregate of all microclusters meets CLB constraints • If an area constraint is given, remove duplication and fragmentation until constraint is met
Conclusions • Reducing depth contributes to a reduction in critical path delay • Node duplication, when used effectively, reduces critical path delay • Duplication can be used to provide a performance-area tradeoff to the designer
Future Work • Promising Post-Placement Optimizations: • Retiming • Leverage a more significant depth reduction • Logic reintroduction • Create duplication to increase performance
Contributions • New algorithm to do clustering and placement • Novel approach for trading-off depth for duplication control • Timing model/placement incorporated into clustering • Delay improves by an average of 11% • Controllable trade-off between area overhead and delay improvements • Plan to submit to FPL ‘08