GPU Accelerated Database for Image Processing; Gaming and Supercomputing
Details
Agenda:
General Discussion:
6:15-7:00pm: What’s new and first-time attendee intros
Main Program:
7:00-7:45pm: Using a GPU Accelerated Database for Image Processing (Tim Child, 3DMashUp)
7:55-8:30 pm: An Unlikely Symbiosis: How the Gaming and Supercomputing Industries are Learning From and Influencing Each Other (Sarah Tariq, NVIDIA)
**********************************
Refreshments sponsored by Supermicro
**********************************
Title: Using a GPU Accelerated Database for Image Processing
Abstract:
The talk will show how open source database can be extended with GPU programming languages (OpenCL and CUDA) to provide a shared, reliable, secure, multi-user, scalable, high performance image processing platform. The talk will describe the use of databases for storage and processing images. It will cover various aspects of images processing; File Format, CODEC, GPU support. For GPUs, it will explain how Image processing techniques can speeded up and optimized for GPUs.
Speaker Bio:
Mr Tim Child is the CEO of a Silicon Valley start-up that is focusing on using GPU accelerated databases for rich content management systems. He previously held Vice President of Engineering positions at Oracle Corporation, Informix Inc. and BEA Systems Inc.; where he led teams that developed advanced database, 3D graphics and web applications. This includes large database systems for the NASA EOSDIS project, Navteqs vehicle navigation systems and the Government of New Zealand LINZ (Land Information New Zealand) system. The author obtained a BSc in Materials Science from Imperial College, London. His interests include advanced database systems, parallel programing, 3D computer graphics, CAD systems, image processing and Web applications.
***
Title: Unlikely Symbiosis: How the Gaming and Supercomputing Industries are Learning From and Influencing Each Other
Abstract:
Over the last couple of decades video games have evolved from simple 2D sprite based animations to nearly realistic cinematic experiences. Today's games are able to do cloth, rigid body and fluid simulations, compute realistic shading and lighting, and apply complex post processing effects including motion blur and depth of field, all in less than a 60th of a second. The hardware powering these effects, the Graphics Processing Unit, has evolved over the last 15 years from simple fixed function triangle rasterizer to a highly programmable general purpose massively parallel processor with high memory bandwidth and high performance per watt. These characteristics make the GPU ideally suited for typical supercomputing tasks. GPUs have become widely adopted accelerators for high performance computing. the game industry has continued to push the increase in visual fidelity, many algorithms that were once only useful in the HPC world are becoming interesting for real-time use.
Speaker Bio:
Sarah Tariq is an engineer at NVIDIA Corporation.