Prof. Manuel Ujaldón Martínez
NVIDIA – CUDA Fellow (Spain)
Bio (NVIDIA)
Prof. Manuel Ujaldón’s presentation at Multicore World will focus in the new software and hardware benefits of CUDA, like dynamic parallelism, Hyper-Q, unified memory and 3D-RAM. His talk at the HPC / SKA workshop will analyse specific cases’ performance.
—————-
Presentation at Multicore World 2015: “Emerging trends in GPU computing”
Abstract:
After a decade being used as hardware accelerators, GPUs constitute nowadays a solid alternative for high performance computing at an affordable cost.
Increasing volumes of data managed by large-scale applications make GPUs very attractive for scientific computing, deploying SIMD parallelism in an unprecedented way.
This talk will review current achievements of many-core GPUs, recent and future hardware enhancements, and emerging challenges to leverage GPUs as accelerators within general-purpose exascale computing. Examples and case studies will be given of software features (dynamic parallelism, unified memory), hardware issues (Hyper-Q, 3D-DRAM) and disruptive low-power devices (Tegra, Jetson and Denver from Nvidia).
———————————————————
Presentation at HPC in the Cloud / SKA Workshop: “Optimizing CUDA kernels for irregular computing”
Abstract:
Real-time processes and HPC computing are ubiquitous within the SKA project.
These goals converge nowadays into GPU computing as the mainstream accelerators.
This talk is aimed to researchers already familiarized with the basic mechanisms of CUDA who want to learn about optimizations and advanced features.
Starting with a simple irregular kernel devoted to sparse matrices, we will illustrate how to maximize performance of a given GPU model.
A list of top 10 optimizations will be deployed to analyze performance, scalability and trade-offs on each step forward towards peak performance.
—–
Manuel’s interests are GPGPU Computing on Image Processing (features extraction, segmentation, classifiers), and Biomedical Applications (large-scale image analysis, biomarkers, bio-inspired algorithms and genomic wide association studies).
Publications at Universidad de Málaga – Spain
Open Parallel 