January/February 2018 issue of acmqueue

The January/February issue of acmqueue is out now


  Download PDF version of this article PDF

ITEM not available


Originally published in Queue vol. 11, no. 4
see this item in the ACM Digital Library



Christoph Lameter - NUMA (Non-Uniform Memory Access): An Overview
NUMA becomes more common because memory controllers get close to execution units on microprocessors.

David Bacon, Rodric Rabbah, Sunil Shukla - FPGA Programming for the Masses
The programmability of FPGAs must improve if they are to be part of mainstream computing.

Andrew Danowitz, Kyle Kelley, James Mao, John P. Stevenson, Mark Horowitz - CPU DB: Recording Microprocessor History
With this open database, you can mine microprocessor trends over the past 40 years.

Alexandra Fedorova, Sergey Blagodurov, Sergey Zhuravlev - Managing Contention for Shared Resources on Multicore Processors
Contention for caches, memory controllers, and interconnects can be alleviated by contention-aware scheduling algorithms.


(newest first)

Antonis Karalis | Mon, 29 Jul 2013 22:32:02 UTC

Congratulation for your research.

Antonis Karalis Systems Architect www.hpcmusic.com

Daniel Courville | Sat, 11 May 2013 16:31:32 UTC

@ Jens M Andreasen

On MacBook Pros with dual video card (Intel & Nvidia), it's possible to force the OS to stick to the Intel video card, thus freeing up the Nvidia card from screen-related tasks. Maybe that could be sufficient to obtain low latencies?

Jens M Andreasen | Sat, 11 May 2013 05:40:24 UTC

You are having unreasonably high latencies in OSX, way above 10 ms. The reason for this is that the GPU occasionally will, by the system, also be used for refreshing the screen, outside of the RT-threads control. This happens even if nothing in particular needs to be refreshed. To get past that you'll need to disable the graphic shell - to my knowledge only possible in Linux? - after which latencies below 1 ms becomes doable if not to say trivial.

Leave this field empty

Post a Comment:

© 2018 ACM, Inc. All Rights Reserved.