Today the Gen-Z Consortium announced the world’s first Gen-Z multi-vendor technology demonstration, connecting compute, memory, and I/O devices at the Flash Memory Summit in Santa Clara. Gen-Z technology enables increased performance and scalability for existing enterprise applications and future memory-centric computing applications.
The demonstration utilizes FPGA-based Gen-Z adapters connecting compute nodes to memory pools through a Gen-Z switch, creating a fabric connecting multiple server vendors and a variety of memory vendors. The multi-vendor participation reflects strong industry support for Gen-Z and showcases how future data centers can leverage this technology to attain a unified, high-performance and scalable fabric/interconnect. Additionally, a separate demonstration will show the scalable prototype connector defined by the Gen-Z Consortium, running at 112 giga-transfers/sec.
The Gen-Z Consortium has doubled its member base since inception and now includes more than 40 companies. It has released four draft specifications to date, all of which are available online.
At Bloomberg, we provide information that powers the global capital markets and our customers depend on us to quickly deliver accurate data, despite exponentially increasing volumes,” said Justin Erenkrantz, head of compute architecture for the finance, media and tech company based in New York City. “Gen-Z’s memory-centric standards-based approach will allow us to bring even more powerful analytics to our customers through a distributed memory and compute fabric.”
Sign up for our insideHPC Newsletter