Adam Wierman, Assistant Professor of Computer Science, and graduate student Jayakrishnan Nair have received the best paper award at the IFIP International Symposium on Computer Performance, Modeling, Measurements and Evaluation (PERFORMANCE) for their paper "Tail-robust scheduling via Limited Processor Sharing." The paper provides a new policy that can optimally schedule, under both heavy-tailed and light-tailed workload, by learning only the first-order workload statistics. Prior to this work, all known policies required learning the entire workload distribution to achieve such guarantees. [Read the paper]
Written by
Trity Pourbahrami
Image Gallery Lightbox