Datapath allocation algorithm for deep submicron VLSI
Lei Wang, Shaojun Wei
Abstract
Lei Wang, Shaojun Wei
Abstract
With the feature size of VLSI scaling down, interconnect delays begin to dominate the circuit performance. Typical Top-Down High level synthesis methodology is not suitable to the deep submicron VLSI design any more. We present a novel design methodology of datapath allocation for deep submicron VLSI. After allocation, we obtain not only the datapath netlists, but also the floorplan information. Floorplan information is used to direct allocation, thus interconnect delays are efficiently optimized. Design examples are presented to help concluding that our algorithm is very efficient.
A significance statement is not available in the OpenAlex record.
A contribution statement is not available in the OpenAlex record.
Method details are not available in the OpenAlex metadata.
Findings are not separately available in the OpenAlex metadata.
Limitations are not available in the OpenAlex metadata.
Application details are not available in the OpenAlex metadata.
With the feature size of VLSI scaling down, interconnect delays begin to dominate the circuit performance. Typical Top-Down High level synthesis methodology is not suitable to the deep submicron VLSI design any more. We present a novel design methodology of datapath allocation for deep submicron VLSI. After allocation, we obtain not only the datapath netlists, but also the floorplan information. Floorplan information is used to direct allocation, thus interconnect delays are efficiently optimized. Design examples are presented to help concluding that our algorithm is very efficient.
Key concepts: Datapath, Very-large-scale integration, Floorplan, Computer science, Interconnection, Routing (electronic design automation), Integrated circuit layout, Parallel computing