Michael LeBeane

Optimizing GPU Cache Policies for MI Workloads

In recent years, machine intelligence (MI) applications have emerged as a major driver for the computing industry. Optimizing these workloads is important but complicated. As memory demands grow...

<u>G</u>PU <u>i</u>nitiated <u>O</u>penSHMEM: correct and efficient intra-kernel networking for dGPUs

Genesys: Automatically generating representative training sets for predictive benchmarking

Case Study of Process Variation-Based Domain Partitioning of GPGPUs

Increasing GPU Translation Reach by Leveraging Under-Utilized On-Chip Resources

Neighborhood-Aware Address Translation for Irregular GPU Applications

Watt Watcher: Fine-Grained Power Estimation for Emerging Workloads

Performance Characterization of Modern Databases on Out-of-Order CPUs

GPGPU Benchmark Suites: How Well Do They Sample the Performance Spectrum?

Proxy-Guided Load Balancing of Graph Processing Workloads on Heterogeneous Clusters

ComP-net: command processor networking for efficient intra-kernel communications on GPUs

Data partitioning strategies for graph workloads on heterogeneous clusters

GPU triggered networking for intra-kernel communications

Extended task queuing: active messages for heterogeneous systems

Lost in Abstraction: Pitfalls of Analyzing GPUs at the Intermediate Language Level

Optimizing GPU Cache Policies for MI Workloads

Control flow behavior of cloud workloads