CUDA: WEEK IN REVIEW

Welcome to CUDA: WEEK IN REVIEW, a news summary for the worldwide CUDA, GPGPU and parallel programming community.

CUDA Pro Tip: The new Understand Fat Binaries and JIT Caching Parallel Forall blog post offers an overview of CUDA "fat binaries" and compilation for multiple GPU architectures, as well as just-in-time PTX compilation for forward compatibility.

CUDA SPOTLIGHT
GPU-Accelerated Guidance and Control for Robotic Systems This week’s Spotlight is on Jon Rogers of the Department of Aerospace Engineering at Texas A&M University. Jon is director of the Helicopter and Unmanned Systems Laboratory, where he works on new technologies for autonomous systems. Read our interview with Jon Rogers.
CUDA NEWS
CUDA 5.5 RC Includes ARM Support NVIDIA announced that the CUDA 5.5 Release Candidate is now available to the public. Features include native compilation on ARM-based platforms; enhanced Hyper–Q support; MPI workload prioritization; and new guided performance analysis. Download CUDA 5.5 RC | CUDA 5.5 webinar (June 25) | Become a registered developer Barcelona Taps Tesla, Tegra for Next-Gen Hybrid Supercomputer The Register reports that Barcelona Supercomputing Center will combine "a baby ARM processor aimed at smartphones and tablets with a full-on GPU coprocessor… With the Kayla system board, which comes in a Mini-ITX form factor, the CPU side is again a Tegra 3 chip from NVIDIA, but the board has a PCI Express 2.0 x16 link to hook a full-on Tesla GPU to the ARM processor." Titan Completes Acceptance Testing Oak Ridge National Lab's Titan supercomputer has completed acceptance testing to ensure the functionality, performance and stability of one of the world's most powerful supercomputing systems for open science. It is the first to combine different types of processing units to maximize performance at such a large scale. New CUDA Book Congrats to author Nicholas Wilt on his new book titled "The CUDA Handbook: A Comprehensive Guide to GPU Programming." The book covers hardware and software, and provides more than 25,000 lines of open source code for developers. The book is available for purchase here. The publishers are offering a 35% discount code for CUDA newsletter readers: CUDAHBNVIDIA. NMath Premium CenterSpace Software released NMath Premium, an easy–to–use library that automatically detects the presence of a CUDA-enabled GPU at runtime and seamlessly redirects appropriate computations to it. The library can be configured to specify which problems should be solved by the GPU, and which by the CPU. If a GPU is not present at runtime, the computation automatically falls back to the CPU without error. INCITE Call for Proposals (Ends June 28) Accelerate your research on Titan, the world’s #1 supercomputer for science, by harnessing more than 20 petaflops of parallel processing using GPUs. Find out more. Upcoming Webinars June 20: GPU-Accelerated XenDesktop, Thomas Poppelgaard, Poppelgaard.com June 25: CUDA 5.5 Features and Overview, Ujval Kapasi, NVIDIA June 26: Understanding Dynamic Parallelism with Allinea’s Unified Tools, Mark O’Connor, Allinea July 9: NVIDIA GRID VCA: A Turnkey Appliance Delivering Remote Graphics, Ankit Patel, NVIDIA July 10: Intro to the CUDA Toolkit as an Application Build Tool, Adam DeConinck, NVIDIA
CUDA EDUCATION	back to the top
Need a CUDA trainer? See list of worldwide CUDA trainers and consultants. Want to get smarter? Enroll in the open online course Intro to Parallel Programming. Have CUDA questions? Check out NVIDIA DevTalk forums and Stack Overflow. Require fast access to docs? Visit the CUDA doc library.
GPU MEETUPS	back to the top
Have you been to a Meetup yet? Join one or start your own. Upcoming events: Paris GPU Meetup, June 20 Rome GPU Meetup, June 21 New York GPU Meetup, June 24 Brisbane GPU Meetup, June 27 Toronto GPU Meetup, June 27