updated 12:57 pm EDT, Tue March 25, 2014
New effort boosts data sharing speeds from five to 12 times current technology
Nvidia today announced at the 2014 GPU Technology Conference that it plans to integrate a high-speed interconnect, called Nvidia NVlink, into its future GPUs. Nvidia claims the new technology will enable GPUs and CPUs to share data five to 12 times faster than they can today. The improvements purge a long-standing performance bottleneck, and help pave the way for a new generation of supercomputers that are 50-100 times faster than today's most powerful systems.
GPUs are currently connected to x86-based CPUs through the PCI Express (PCIe) interface, which limits the GPU's ability to access the CPU memory system and is four to five times slower than typical CPU memory systems. As the NVLink interface will match the bandwidth of typical CPU memory systems, it will enable GPUs to access CPU memory at its full bandwidth.
This high-bandwidth interconnect should improve accelerated software application performance. Because of memory system differences -- GPUs have fast but small memories, and CPUs have large but slow memories -- accelerated computing applications typically move data from the network or disk storage to CPU memory, and then copy the data to GPU memory before it can be processed by the GPU. With NVLink, the data moves between the CPU memory and GPU memory at much faster speeds, making GPU-accelerated applications run much faster.
Although future Nvidia GPUs will continue to support PCIe, NVLink technology will be used for connecting GPUs to NVLink-enabled CPUs as well as providing high-bandwidth connections directly between multiple GPUs. Also, despite its very high bandwidth, NVLink is reported to be substantially more energy efficient per bit transferred than PCIe.
Nvidia has designed a module to house GPUs based on the Pascal architecture with NVLink. This new GPU module is one-third the size of the standard PCIe boards used for GPUs today. Connectors at the bottom of the Pascal module enable it to be plugged into the motherboard, improving system design and signal integrity.
Nvidia will add NVLink technology into its Pascal GPU architecture following this year's new Nvidia Maxwell computing architecture. The new interconnect was co-developed with IBM, which is incorporating it in future versions of its Power CPUs.