Hybrid Memory Cube Consortium Releases HMCC 2.0 Specification 

The Hybrid Memory Cube Consortium (HMCC), dedicated to the development and establishment of an industry-standard interface specification for Hybrid Memory Cube (HMC) technology, has announced the finalization and public availability of its HMCC 2.0 specification (HMCC 2.0).

The new HMCC 2.0 specification advances data rate speeds from 15 Gb/s up to 30 Gb/s, establishing a new generation of memory performance. HMCC 2.0 also migrates the associated channel model from short reach (SR) to very short reach (VSR) to align with existing industry nomenclature.

“With 150 members, the Hybrid Memory Cube Consortium has gained considerable momentum since its inception and, as a result, has more and better inputs on how the interface can best fit tomorrow’s applications,” said Jim Handy, Director of Objective Analysis. “This latest release of the HMCC 2.0 specification shows a commitment to evolving a family of specifications targeting all high-performance computing applications.”

The HMCC was founded in October 2011 by co-developers Altera, Micron, Open-Silicon, Samsung and Xilinx. The HMCC finalized and released their first specification in May 2013, demonstrating consensus among leading semiconductor developers to drive adoption of HMC into next-generation systems. Since its launch, the HMCC has grown to include over 150 OEMs, enablers and integrators who regularly participate in the development and discussion of HMC standards. The finalization of the second generation of HMCC specifications is a key milestone in the development of this leading memory technology and an indication of its continued adoption.

“HMCC 2.0 gives designers a mature solution for breaking through memory bottlenecks and delivering a new generation of systems with unprecedented memory performance,” said Hans Boumeester, Open-Silicon’s vice president of IP and engineering operations. “The ratification of the new standard means that these designers will have access to standards-compliant IP for immediate integration into chips and systems that meet the growing bandwidth demands of next-generation data center and high-performance computing applications.”