Sponsored Content
Special Forums UNIX and Linux Applications High Performance Computing Memory Barriers for (Ubuntu) Linux (i686) Post 302431048 by Corona688 on Sunday 20th of June 2010 09:14:36 PM
Old 06-20-2010
Quote:
Originally Posted by gorga
Let me put it this way, I have a list of jobs to do, as I progress through the list I find some jobs aren't ready. I can either block/sleep etc and wait for the job to become ready, or I can move onto the next task and execute that one instead.
...and when you don't assume best-case, you could be going through the next few thousand not-ready tasks. Meanwhile every idle worker's doing the same thing. I begin to understand why you're concerned about contention for memory.
Quote:
If I build my system with locks, I'll adhere to the first approach, whereas what I'm trying to achieve (aided by atomic flags and memory barriers) is the second.
But because it doesn't block, you'll be wasting time scanning the list anyway, time that could have been spent doing actual work. And since your system's as busy idle as it is when actually busy you'll have a difficult time guessing how much. If I read you correctly, the jobs are all tiny. How tiny? How much more work is it to do a job than to scan the list? (And don't assume it could never, ever become mostly empty, that's the goal, not the proof.) If they're even in the same ballpark, you're going to be wasting a worrying proportion of CPU time scanning your list.

Anyway, the queue needn't block like you're describing. Put jobs in the queue when they become ready, don't just stick them there in advance, that way threads won't block when picking up jobs unless you're actually out of jobs -- in which case you want them to block. If the queue's big enough and jobs can be added fast enough, things can run smoothly. You can also do other things to streamline the queue -- hand out jobs 16 at a time instead of one at a time, switch between multiple queues, etc.

Last edited by Corona688; 06-20-2010 at 10:33 PM..
 

4 More Discussions You Might Find Interesting

1. UNIX for Dummies Questions & Answers

Memory-waste in Ubuntu/Debian?

I have 512 mem on this laptop, though 'top' tells me I only have 380. However, Ubuntu is using 288 mb of memory, when I only have 3 terminals, running lynx, vim(for this file) and (of course) top. Considering it I have lynx running a 600 page txt file, which of course would eat some memory but 300?... (0 Replies)
Discussion started by: riwa
0 Replies

2. Linux

i686, x86 64, ppc

Hi, i am quite new to linux. I am interested in fedora linux distro. Fedora Project I dont know which one to choose, either i686, x86 64 or ppc. I prefer a live cd, coz its easy to use. And what is the difference between "Fedora Desktop Live Media" and "Fedora KDE Live Media". (3 Replies)
Discussion started by: superblacksmith
3 Replies

3. Programming

Getting the total virtual memory for ubuntu in c++

Hi guys , i need to get the total virtual memory in ubuntu but i need to write a C++ code for that, any idea on how to go about doing it? any references? or website that i can refer to ? (6 Replies)
Discussion started by: xiaojesus
6 Replies

4. Ubuntu

XP and Linux (Ubuntu) on same disk, Can I install Ubuntu on not-yet partitioned portion of disk?

My PC (Esprimo, 3 yeas old) has one hard drive having 2 partitions C: (80 GB NTFS, XP) and D: (120 GB NTFS, empty) and and a 200 MB area that yet is not-partitioned. I would like to try Ubuntu and to install Ubuntu on the not-partitioned area . The idea is to have the possibility to run... (7 Replies)
Discussion started by: C.Weidemann
7 Replies
queuedefs(4)                                                       File Formats                                                       queuedefs(4)

NAME
queuedefs - queue description file for at, batch, and cron SYNOPSIS
/etc/cron.d/queuedefs DESCRIPTION
The queuedefs file describes the characteristics of the queues managed by cron(1M). Each non-comment line in this file describes one queue. The format of the lines are as follows: q.[njobj][nicen][nwaitw] The fields in this line are: q The name of the queue. a is the default queue for jobs started by at(1); b is the default queue for jobs started by batch (see at(1)); c is the default queue for jobs run from a crontab(1) file. njob The maximum number of jobs that can be run simultaneously in that queue; if more than njob jobs are ready to run, only the first njob jobs will be run, and the others will be run as jobs that are currently running terminate. The default value is 100. nice The nice(1) value to give to all jobs in that queue that are not run with a user ID of super-user. The default value is 2. nwait The number of seconds to wait before rescheduling a job that was deferred because more than njob jobs were running in that job's queue, or because the system-wide limit of jobs executing has been reached. The default value is 60. Lines beginning with # are comments, and are ignored. EXAMPLES
Example 1: A sample file. # # a.4j1n b.2j2n90w This file specifies that the a queue, for at jobs, can have up to 4 jobs running simultaneously; those jobs will be run with a nice value of 1. As no nwait value was given, if a job cannot be run because too many other jobs are running cron will wait 60 seconds before trying again to run it. The b queue, for batch(1) jobs, can have up to 2 jobs running simultaneously; those jobs will be run with a nice(1) value of 2. If a job cannot be run because too many other jobs are running, cron(1M) will wait 90 seconds before trying again to run it. All other queues can have up to 100 jobs running simultaneously; they will be run with a nice value of 2, and if a job cannot be run because too many other jobs are running cron will wait 60 seconds before trying again to run it. FILES
/etc/cron.d/queuedefs queue description file for at, batch, and cron. SEE ALSO
at(1), crontab(1), nice(1), cron(1M) SunOS 5.10 1 Mar 1994 queuedefs(4)
All times are GMT -4. The time now is 03:38 AM.
Unix & Linux Forums Content Copyright 1993-2022. All Rights Reserved.
Privacy Policy