Name of the project
Performance analysis of hardware data structures
Background
Over the years many hardware designs have been given for data structures
commonly found in algorithms, such as stacks, queues, and priority queues. The
main reason for hardware implementations of these structures is performance.
However, for a hardware design there are many performance metrics that can be
optimized most notably area, throughput and latency, but more recently also energy
consumption. Solutions that are optimal with
respect to one metric are usually suboptimal with respect to another metric.
Goals and task
The goal of this project is to describe and analyze a number of hardware
designs for data structures using the
Haste toolset from Handshake Solutions
in search for Pareto optimal
solutions.
Cooperation
None.
References
At the start of the project a number of references to articles in the
literature will be provided.
Pre-requisites
Successful completion of course 2IN30
(VLSI programming).
Planning (time frame & supervisor)
The project will be supervised by Rudolf
Mak from the TU/e. Duration three months (20 ects).
Status (who is doing the project;
maybe a pointer to recent info)
Open for
application.