Gradient Boosted Regression Trees (GBRT) are the current state-of-the-art learning paradigm for machine learned websearch ranking — a domain notorious for very large data sets. ...
Stephen Tyree, Kilian Q. Weinberger, Kunal Agrawal...
Method speculation of object-oriented programs attempts to exploit method-level parallelism (MLP) by executing sequential method invocations in parallel, while still maintaining c...
This paper describes Tempest, a collection of mechanisms for communication and synchronization in parallel programs. With these mechanisms, authors of compilers, libraries, and ap...
At our university, we have a number of small-tomedium-size compute clusters and some technical simulations which could benefit from using several of these clusters simulaneously....
By federating resources from multiple domains, a shared infrastructure provides aggregated computation resources to a large number of users. With rapid advances in virtualization ...
Dongyan Xu, Paul Ruth, Junghwan Rhee, Rick Kennell...