Slides of the DK CIM Talk
ISC Links
Reproducibility
Classics
- Bailey: Twelve Ways to Fool the Masses When Giving Performance Results on Parallel Computers
- Amaral: How did this get published? Pitfalls in experimental evaluation of computer systems
- Wellein, Hager: Fooling the masses with performance results: Old classics and some new ideas
- Nature: 1,500 scientists lift the lid on reproducibility
Background Papers for ISC 19 Talks
- Hoefler: Performance Reproducibility in HPC and Deep Learning
- Hoefler: Scientific Benchmarking of Parallel Computing Systems Twelve ways to tell the masses when reporting performance results
- Ludwig, Geyer: Reproduzierbarkeit
Floating Point
Classics
Floating Point Alternatives
- Lindstrom et al., Universal coding of the reals: alternatives to IEEE floating point
- Lindstrom: Talk at Next Generation Arithmetic (CoNGA) Conference