As AI workloads become increasingly demanding, data centers are facing a critical memory-wall challenge – the growing gap between processor performance and the speed at which data can be accessed from memory. Credo Technology Group Holding Ltd. CRDO is addressing this bottleneck by developing a new open interconnect standard within the Open Compute Project (OCP). It has established the OCP Open Chiplet Economy (OCE) Lightweight Serial Interconnect (LSI) Workstream, aimed at developing efficient interconnect solutions for AI infrastructure. As part of the initiative, Credo plans to contribute its OmniConnect lightweight AXI framer specification to OCP.

HBM has become a key component of modern AI systems, but its high cost, limited availability and density constraints are creating challenges as AI infrastructure scales. Credo believes a lightweight serial interconnect could offer a more flexible way to connect compute and memory resources. By enabling memory disaggregation and chip-to-chip communication, the technology could allow AI system designers to build more modular architectures. According to Credo, these composable designs could deliver up to 25 times greater memory density and 5% higher bandwidth than HBM4 in certain configurations.

Originally published on zacks.com, part of the BLOX Digital Content Exchange.

(0) comments

Welcome to the discussion.

Keep it Clean. Please avoid obscene, vulgar, lewd, racist or sexually-oriented language.
PLEASE TURN OFF YOUR CAPS LOCK.
Don't Threaten. Threats of harming another person will not be tolerated.
Be Truthful. Don't knowingly lie about anyone or anything.
Be Nice. No racism, sexism or any sort of -ism that is degrading to another person.
Be Proactive. Use the 'Report' link on each comment to let us know of abusive posts.
Share with Us. We'd love to hear eyewitness accounts, the history behind an article.