Rob FarberHeterogeneous Code Performance and Portability Using CUDA to SYCL Migration ToolsThe need for faster time to solution along with meeting customer needs for more power- and cost-efficient computing has made heterogenous…Jul 7, 2023Jul 7, 2023
Rob FarberBalancing High Bandwidth Memory and Faster Time-To-Solution for ManufacturingBalancing High Bandwidth Memory and Faster Time-To-Solution for ManufacturingApr 25, 2023Apr 25, 2023
Rob FarberUnderstanding Why AI and Analytics Have to Run on HPC ClustersAI and HPC are converging to become part and parcel of each other.Jan 3, 2020Jan 3, 2020
Rob FarberHurray for DAOS! Exit the Bottlenecks Imposed by POSIX IOBy Rob FarberOct 15, 2019Oct 15, 2019
Rob FarberDemocratizing the Challenges of Visualization in the Age of Data with SoftwareChris Johnson wrote about the top vis research problems. Today software is the democratizing force in real-time viz over all data scales.Jun 21, 2019Jun 21, 2019
Rob FarberLIBXSMM Brings Deep-learning “Lessons Learned” to Many HPC Applications“Once we realized the performance limitation was memory bandwidth, we really had to get down to work”, Alexander Heinecke (Research…Jun 13, 2018Jun 13, 2018
