Today’s large-scale distributed systems consist of collections of nodes that have highly variable availability — a phenomenon sometimes called churn. This availability variati...
Concurrent garbage collectors require write barriers to preserve consistency, but these barriers impose significant direct and indirect costs. While there has been a lot of work ...
We present MRNet, a software-based multicast/reduction network for building scalable performance and system administration tools. MRNet supports multiple simultaneous, asynchronou...
Philip C. Roth, Dorian C. Arnold, Barton P. Miller
Enormous amounts of information are produced every day, all over the world. but very little of it is true. In this paper, we describe the modeling component of a current events an...
Achieving peak performance from library subroutines usually requires extensive, machine-dependent tuning by hand. Automatic tuning systems have emerged in response, and they typic...