The HBM-PIM chip is now being tested inside a range of datacentre AI accelerators, expected to be Nvidia and AMD who both support HBM2 interfaces, with all validations expected to be completed within the first half of this year. One of the first users of the chip is in the supercomputer used by the Argonne lab in the US.
“I’m delighted to see that Samsung is addressing the memory bandwidth/power challenges for HPC and AI computing. HBM-PIM design has demonstrated impressive performance and power gains on important classes of AI applications, so we look forward to working together to evaluate its performance on additional problems of interest to Argonne National Laboratory,” said Rick Stevens, Argonne’s Associate Laboratory Director for Computing, Environment and Life Sciences.
Related AI HBM articles
- CAN ADVANCED FAB AND PACKAGING FINALLY REALIZE IN-MEMORY PROCESSING?
- BAIDU AND SAMSUNG ELECTRONICS PARTNER ON 14nm AI CHIP
- RAMBUS, AICHIP TEAM ON MEMORY INTERFACE FOR MACHINE LEARNING
Other articles on eeNews Europe