Sumit Gupta is a busy man. Named by HPCwire as a 2013 “Person to Watch“, Sumit does not idly take time to create a blog post unless it conveys a message about the NVIDIA Tesla development and marketing effort. His recent blog, “Fostering an Explosion of Innovation in the Data Center“, posted by Steve Hamm, recognizes how the data-center is going to be supporting mobile applications and the Tesla Groups’s positioning for that new sales paradigm.
The public facing entity of the collaboration is the OpenPower foundation containing some heavy-hitters. From the Wikipedia entry:
- Google, Tyan, Nvidia and Mellanox are all founding members of the OpenPOWER Foundation.
- Nvidia is looking to merge their graphics cores and Mellanox to integrate their high performance interconnects with Power cores.
- Tyan is said to be working on servers using POWER 8 and Google sees using POWER processors in its data centers as a future possibility.
- Altera announced support for OpenPOWER in November 2013 with their FPGA offerings and OpenCL software.
- On January 19, 2014, the Suzhou PowerCore Technology Company and the Research Institute of Jiangsu Industrial Technology announced that they will join the OpenPOWER Foundation and license POWER8 technologies to promote and help build ecosystems around and design custom made processors for use in big data and cloud computing applications.
- On February 12, 2014, Samsung Electronics joined.
- As of March 2014 members also include Altera, Fusion-io, Hynix, Micron, Servergy and Xilinx.
The pending mass production of the Tegra K1 shows that NVIDIA is aggressively moving into the mobile market. Similarly, we see NVIDIA dominating the HPC accelerator market. It looks like Sumit’s blog is laying additional groundwork for a classic snare trap as both of NVIDIA’s IBM Power and ARM projects challenge x86 dominance in the enterprise and cloud. Either an IBM Power or a 64-bit ARM processor will supply the missing piece needed for NVIDIA’s move into the enterprise and cloud data centers. Once that happens, NVIDIA (and partners) can pull the ends of the snare (mobile and HPC) to leave themselves well-poised to compete in the enterprise market on those key balance metrics: performance per watt and performance per dollar.
The software side has not been neglected either with CUDA ARM now available (and running on the Jetson board) plus the potential of OpenACC.
- Create Mobile to HPC Applications using a Single Source Tree: Basic tools are now available to test efficacy — it’s worth taking a look
- Preserving Sanity in the Face of Rampant Technology Change: Portability is the only rational path to follow
- Mobile Tech between a Rock and a Hard Place: All signs indicate a healthy continuing demand for technology that can support ever more demanding eye-candy and apps on very high resolution display devices
- Rob Farber on OpenACC
I also present my analysis of NVlink in my interview with Rich Brueckner of insideHPC, “Video: Rob Farber Looks at NVIDIA’s new NVlink Technology“.
Update 4/24/2014: Well Sumit is not wasting time, that is for certain: “NVIDIA to offer full CUDA support 4Q2014 for OpenPower“.
Leave a Reply