Samsung Begins Mass Producing of 4GB Stacks of HBM2 – World’s Fastest DRAM
Today, Samsung Electronics announced that it has officially commenced mass production of the industry’s first 4-gigabyte (GB) DRAMs package based on the second-generation High Bandwidth Memory (HBM2) interface. The newest HBM2 DRAM chips, which are designed with the latest 20nm technology, will target the high performance computing (HPC), advanced graphics and network systems, as well as enterprise servers.
“By mass producing next-generation HBM2 DRAM, we can contribute much more to the rapid adoption of next-generation HPC systems by global IT companies,” said Sewon Chun, senior vice president, Memory Marketing, Samsung Electronics. “Also, in using our 3D memory technology here, we can more proactively cope with the multifaceted needs of global IT, while at the same time strengthening the foundation for future growth of the DRAM market.”
Samsung claims that its new HBM solution is more than seven times faster than the current generation DRAM chips. That allows HPC systems and graphics cards to respond faster when running tasks including parallel computing, graphics rendering and machine learning. Moreover, the HBM2 DRAM marks a new milestone for the South Korean electronics giant in TSV (Through Silicon Via) technology that is a core component of the HBM DRAM design.
As for the internal configuration of the latest 4GB HBM2 DRAM chips, Samsung states in a press release:
The 4GB HBM2 package is created by stacking a buffer die at the bottom and four 8-gigabit (Gb) core dies on top. These are then vertically interconnected by TSV holes and microbumps. A single 8Gb HBM2 die contains over 5,000 TSV holes, which is more than 36 times that of a 8Gb TSV DDR4 die, offering a big improvement in data transmission performance compared to typical wire-bonding based packages.
Additionally, Samsung will begin production of an 8GB HBM2 DRAM package within this year.
HBM2 is going to be the next generation DRAM standard and will power both the Polaris as well as Pascal, the upcoming GPU architectures from AMD and Nvidia respectively. The flagship GPUs from these families will be using over 1 TB/s HBM2 chips with VRAM of up to 32 GB, as recently confirmed by JEDEC.