On Wednesday, Feb. 28, acadianacathedral.com will be upgraded between 6 p.m. - 12 a.m. PT. During this upgrade, the site may not behave as expected and pages may not load correctly. Thank you in advance for your patience.
Solving the toughest challenges requires a compute foundation with fast memory. Our HBM2E delivers the necessary performance, massive parallelism and power efficiency for next-generation data centers and high-performance compute systems.
Read our new tech brief for best practices from Micron experts on how you and your ecosystem of integrators can achieve maximum performance from the high-performance memory in Micron’s ultra-bandwidth solutions portfolio.
When nothing else matters but speed. Rebecca Lewington talks with Micron VP Akshay Singh, who leads the R&D team that created Micron’s amazing HBM2E memory in-package memory. TSV interconnect is just one thing that makes our stacked memory so fast.
This infographic tells what HBM2E is, why it’s necessary for applications like AI and high-performance computing and why being closer to the CPU matters for high-bandwidth performance.
BLOG
Girish Cherussery on the technology and business of high-performance memoryBuilding one of the fastest, highest-performance memories on the planet demands a whole new level of collaboration.
BLOG
HBM2E: Our fast memory for the modern data centerHigh-performance applications like AI training and inference are driving the need for high-throughput memory. New memory approaches like HBM2E are often required for heterogenous data centers, which off-load some functions of the CPU to dedicated hardware (GPUs, ASICs and FPGAs) to gain speed and efficiency.
TECHNICAL BRIEF
The Demand for Ultra-Bandwidth SolutionsDriving the need for affordable Ultra-Bandwidth memory are data-hungry applications like AI, gaming, machine learning, autonomous driving, high-performance computing, virtual reality… GDDR6X raises Micron’s memory offerings to new heights, along with HBM2e and more.
Video
Addressing big science problems with GPU and high-performance memoryAn expert panel at GTC 2021 discusses how compute and memory in the form of NVIDIA GPUs (A100), using high-performance memory (HBM2E), can enable high-capacity compute systems to process more than 100 petabytes per second.