Investigating the Methodology of papers at the 2020 International Symposium on Computer Architecture (ISCA2020)



We are gradually going through some of the papers published at ISCA2020 and reviewing the methodology used, and see if the papers themselves are repeatable scientific research. Some details on the methodology challenges found in architecture papers can be found here.

More papers will be added as we have time, as it can be a lot of work to properly look at the papers, and often the papers themselves are a bit vague about the methodology used.

Commutative Data Reordering: A New Technique to Reduce Data Movement Energy on Sparse Inference Workloads


Tailored Page Sizes


BabelFish: Fusing Address Translations for Containers


Efficiently Supporting Dynamic Task Parallelism on Heterogeneous Cache-Coherent Systems


Back to the ISCA methodology overview