Article 26VBP Memory Bandwidth and System Balance in HPC Systems

Memory Bandwidth and System Balance in HPC Systems

by
Rich Brueckner
from High-Performance Computing News Analysis | insideHPC on (#26VBP)
john-mccalpin-150x150.jpg

"This talk reviews the history of the changing balances between computation, memory latency, and memory bandwidth in deployed HPC systems, then discusses how the underlying technology changes led to these market shifts. Key metrics are the exponentially increasing relative performance cost of memory accesses and the massive increases in concurrency that are required to obtain increased memory throughput. New technologies (such as stacked DRAM) allow more pin bandwidth per package, but do not address the architectural issues that make high memory bandwidth expensive to support."

The post Memory Bandwidth and System Balance in HPC Systems appeared first on insideHPC.

External Content
Source RSS or Atom Feed
Feed Location http://insidehpc.com/feed/
Feed Title High-Performance Computing News Analysis | insideHPC
Feed Link https://insidehpc.com/
Reply 0 comments