We are listing some great work done by researchers over the years with very brief comments by us to aid anyone looking to venture a bit deeper into the areas that our project spans.
The literature is expected to grow to cover the following categories:
- High performance computing in general, with focus on HPC with FPGAs and GPUs
- FPGA-focused literature
- High-level languages, compilers, translators, and related material with emphasis on FPGAs and GPUs.
- Scientific and Numerical Modelling
Suggestions to the list are always welcome.
|2008||The promise of high performance reconfigurable computing||This is a bit dated paper, but a very useful and friendly introduction to high-performance reconfigurable computing. Some realisitc performance and efficiency figures are quite useful, and gives a helpful classification for HPC systems based on FPGAs.
|2012||GRAPE-8 — An Accelerator for Gravitational N-body Simulation with 20.5Gflops/W Performance||They have developed special purpose pipelined processors for particle-particle interactions, and FPGA is used for PCI interface. It is an interesing case study as it gives a reference point in the form of the performance and power efficiency of a custom chip for a well-known kernel used in scientific models.|
|2002||Using floating point arithmetic on FPGAs to accelerate scientific n-body simulations||A useful paper for a perspective on use of floating-point arithmetic in FPGAs, and the returns possible by use of custom representations rather than aritifcially limiting to IEEE representations.|
|2008||Multiparty Asynchronous Session Types||This paper presents the work that extends binary session types to multiparty asynchoronous session types (MPSTs). Use of MPSTs to formalize and automate the translation of high-level code for FPGA implementation is a key aspect of our project.|