Over at ACM QUEUE, Christoph Lameter has posted an overview of NUMA (Non-Uniform Memory Access) in today’s computing architectures.
As the trend toward improving system performance by bringing memory nearer to processor cores continues, NUMA will play an increasingly important role in system performance. Modern processors have multiple memory ports, and the latency of access to memory varies depending even on the position of the core on the die relative to the controller. Future generations of processors will have increasing differences in performance as more cores on chip necessitate more sophisticated caching. As the access properties of these different kinds of memory continue to diverge, operating systems may need new functionality to provide good performance.
Read the Full Story.
Related posts:
- OpenSolaris and NUMA
- DailyTech: excellent overview of Barcelona
- How can we overcome bus saturation in multi-core systems?
The post An Overview of NUMA appeared first on insideHPC.
