OpenSUSE Install :: Multithreading And Dual Channel Memory - Finding The Differences
May 27, 2011
Until now I have been running a single threaded cpu at single channel speed but yesterday I bought a 3.2 gig p4 for 20 bucks. It has hyperthreading and an 800 mhz fsb which supports dual channel speeds on the chipset. It hasn't arrived yet so I haven't put it in the system but what I would like to know is whether 11.4 work properly with these changes, and will it detect and implement them properly or will I need to reinstall it?
PC seems to run fine in single channel mode but monitor reports no input in dual channel mode and memtest86+ lists from 1-3 errors per pass in single channel mode. Motherboard (Gigabyte P43T-ES3G) can't be set for the 1.65 volts required by the RAM (2x2 of OCZ3P1333LV4GK PC3 10666). The mobo voltage choices are 1.6 or 1.7 volts. I have tried both and increased the MCH voltage from 1.1 v to 1.2 v, but so far can't get it to run in dual channel mode at all and single channel mode seems to run fine but I still get errors in memtest86+. Also tried optimized Bios settings and loosened memory timings (from 7-7-7-16 to 8-8-8-27 and tRFC 60 to 70) and reset Command Rate from 0 to 2. CPU is Core 2 Quad @ 2.4 Ghz, PSU is rated at 400 watts, Video card is Gigabyte GT220.
Thinking about building a new system, with 4GB of DDR3 RAM. I am just wondering, which would you prefer: Dual-channel DDR3 or Quad-channel DDR3? Does one have better speed performance over another? Will one way keep your system cooler?
I am wondering just how the triple channel memory architecture, like found in various socket 1366 CPUs, including i7, works. For maximum speed in close locality accesses, in theory it would need to interleave at the memory access width (64 bit?).But interleaving at a multiple of 3 is what looks odd. A multiple of 2 or 4 would be trivial, by just using the lowest 1 or 2 bits (of the access word address), to select with part of the interleave stride. But a multiple of 3 would mean doing an arithmetic division by 3 and using the modulus, for each memory cycle. Can it really do that, that fast?
Or is something else going on where it approaches having a perfect interleave, but not quite, in order to keep the operation speed at maximum (e.g. some simpler kind of arithmetic that can keep up with a slightly slower memory speed that results from a less than perfect interleave)?
Also, I was reading somewhere where someone suggested that if you put 2 DIMMs in one channel, and 1 DIMM each in the other two channels (where all DIMMS are the same size), it will operate in triple channel mode for 3/4 of the memory address space, and in single channel mode for the remaining 1/4 of memory. Is it designed to do that? Is there an advantage doing that instead of putting 2 DIMMs in the 1st channel and 2 DIMMs in the 2nd channel, and leaving the 3rd channel empty (e.g. getting only double-channel speed, but doing so for all of memory)?
IWSTM that if I'm running multiple threads or hyperthreads, these may be accessing memory at sufficiently widely arranged locations (depending on virtual memory page placement) that even if there is no interleave at all, there can be some improvement.
Is this the latest version. i don't know how old 11.4 is, but there weren't a great deal of updates. have added a few repos like packman etc.
Kept /home as the same and what a difference that made. i used to do a fresh install every time, it saved just about all my settings!!! just needed a few more programs and stuff but nothing major.
Really quick, clean install. just need to delete my old root partition for the old 11.3 system.
Would also just like to know are there many differences between 11.3 and 11.4?
We recently bought an i7 X58 server with 6 memory slots. Since it was cheaper, we opted to get the min memory configuration with the machine (3x1Gb) and then buy 3x4Gb separately. It seems a shame to waste the 3Gb, so I'm wondering how to "mix" the memory without sending everything into single channel mode. The information I've read is rather confusing, and often contradictory.
we found that if we use 'top' to show the memory usage of a server (SuSe Linux 10), we can get virtual memory usage as well as 'Resident memory' usage. For virtual mem or a particular process, it is around 1.1GB, which is large but for resident memory, it only consumes 300MB. Are there anyone who knows what the differences are? I would also like to know whether the difference (1.1GB - 300MB) = 800MB are actually available for use by other applications in the system.
I would like to try another distro and OpenSuse seems interesting. What are the differences between Ubuntu and OpenSuse? Also, I know that my wireless card will not work as soon as Suse is installed. I also know that I will have to use ndiswrapper to install it, my question on this is, I have the drivers (have Ubuntu in file name) and ndiswrapper I had to use for Ubuntu to get it to work, but will I be able to use all that or do I have to get a driver that is for Suse?
There are 3 channel DDR-3 RAM kits available on market.Would its performance be better than dual channel RAM kit, apart from RAM capacity? I'm planning to build a new box testing server consolidation on virtualizaion running Xen/Vserver/OpenVZ as software. All servers running 64bit Linux OS. My new planning is to get 2 sets DDR-3 1333/1600 4Gx2 kit. I may consider 3 channel kit dropping the total RAM capacity to 6G if it has better performance. Besides can mobo support dual channel RAM also allowing 3 channel RAM to work on it without sacrifying performance?
differences between Kernel Default and Kernel Desktop? I've found some past threads like this link and this other link, and some other google info, which suggest the only difference would be the io scheduler. Also, I see the default grub choice is "Desktop" and not "Default", so I take this as a suggestion to prefer one over the other.
However, my broadcom 4312 wireless only works on the "default" and not on the "desktop" kernel, so I guess there must be other differences. I just want to evaluate which one is the less long-term risk option to go.
I'm trying to crack my wep password just for kicks, but I keep running into the same error. When I try to use airplay-ng for the packet injection, it always gives me the error " wlan0 is on channel -1, but the AP uses channel 6". I've tried setting wlan0 to channel 6 manually, but that doesn't help. I spent a while searching for a solution, but it seemed like each page had a different, usually highly complex solution, that still doesn't work. I have the most recent drivers for my wireless card, and it meets all the requirements. So, is there any solution to this that will actually work?
If i have a shell script to run on a Linux unit, which has a command to Reset(or say upgrade) the unit in between the script.Is it possible to find out the memory address of the next waiting command so that i can store the address to any environment variable in flash , then after Reset will continue Run from the stored address.
About 2 years ago I installed SUSE10.3 over the internet. Now the older repositories seem to have disappeared completely and I can't install any new software or little tools that I have missed. Is there any way to find 10.3 repositories somewhere or do I need to make a completely new installation every year?
I was looking around on Wikipedia, and ran across the UltraSPARC T2 processor. This is a processor with 8 cores (8 physical processors) and 8 threads per core (64 logical processors!)I'm aware of Intel's Hyper-Threading and other technologies, and I know collisions between the threads have the possibility of decreasing the processor's throughput.From what I know, the O(1) scheduler (and surely the current CFS scheduler) is aware of Hyper-Threading, and is careful with processor affinity for threads (mainly, it tries to fill every physical processor before giving any multiple threads, and it tries to keep processes that much move between logical processors on the same physical processor.)
However, although that certainly helps somewhat, I believe the largest gain can come if the compiler is allowed to re-arrange instructions, so that more than two threads can be run at once without collisions in one physical processor. For this to happen, I believe we'd need to come up with a new executable format that can store multiple threads in one file, so the scheduler can schedule them as one block.For example, with the current scheduler and executable formats, the UltraSPARC would appear as 64 logical processors. However, if we consider a system that consists entirely of thread-aware systems, it might look at it as 8 logical processors, and schedule eight processes at a time. Since only one process would exist on each physical processor, the compiler would be able to optimize and significantly reduce resource contention (which I mention below.)
We can look at one physical processor with 8 threads of execution (or any number >2.) If you run 8 different processes, as Linux would currently do, there would likely be a lot of resource contention. My guess is that most of the time, several threads would be waiting for the unit they need to become available. I feel that this is supported because I saw a number from Sun (I forgot where) that the correct optimization settings on their compiler can yield a 300% speedup, on a practical workload.Is there any infrastructure available for Linux that supports processors like the UltraSPARC T2, that have a high number of hardware threads per core? Also, I know this would be a large task to implement, and there would be hurdles (how do you mix multithreaded processes with single-threaded processes on one machine?), and I am wondering if others think there would be enough benefit to justify the effort.
I have developed a proprietary multithreaded application using Qt library version 4.2.1 on RHEL 4 platform. The application communicates with different external interfaces using serial communication using well defined protocols. For each system I have a thread that handles data acquisition and processing, it sets an update flag which is read by the GUI thread to update GUI after specified number of seconds. The incoming data from external systems is processed and updated on GUI at certain intervals. Each system update rate on GUI is different. Generally it is 1 second. Their rate of sending data is different and is generally 1Hz, 5 Hz and 10 Hz. For systems with string data I have no issues. But some systems send raw data and I have to process every byte of the entire packet, for which I have to use for loops. Now my problem is if I keep the update rate of GUI for such systems less than a fixed number of seconds (which I have to calculate by hit and trial) the application hangs after sometime. I have tested this system exlusively by switching of all other systems and it hangs standalone.
This leads me to conclude that timing of GUI update is what is causing the application to hang. Also I read somewhere on the net that in multithreading application timers play an important role and slight variation in timing can lead to hung applications. As of now I have kept 5 seconds as the GUI update time and app runs fine. Can somebody suggest something else which might be the cause?Note that I have used mutexes, wait conditions and all necessary tools which are used in multithreading app. But it is only the timer on which I could pin down the issue.
Im implementing a chat application using Jabber/XMPP and gloox framework which should send and receive messages concurrently in Ubuntu Linux.
The problem faced is both the threads are created and pthread_join( ) is called for both.The iSend thread is scheduled first but gets suspended at cin. Once the recv( ) function is called, which runs in iRecv thread, the recv call back function handleMessage( ) is called. However the control never shifts back to the iSend thread which should call SendMessage( ) function.
Anyone have the links to the upgrade repos for 11.4? Wanting to go ahead and set those up so when I get in to work in the morning I can start the upgrade.
I'm trying to make my USB bootable with the OpenSUSE 11.4 64 bit KDE iso image. When I open up SUSE Studio Image Writer (by following the Windows instructions here), it doesn't find the .iso image on my desktop. When I'm selecting the image, it's looking for a .raw file. Do I have to do some trickery to change the .iso image to a .raw?
When there is a total of 2 GB of physical RAM, how much RAM would you give to openSUSE-11.2 (i586) with KDE as a VirtualBox guest in Vista? There is no intention of running a second VM at the same time.
I am going to install 11.3 on a system with a Phenom X6 and 16 GB memory.I'm now unsure whether to set up a 64 bit system or a 32 bit system with pae.Are they equivalent what concerns use of the complete 16 GB memory? I assume 64 would be more effective?On the other hand there seem to be less problems with the availability and/or installation of applications on a 32 bit system.
I'm thinking of trying out opensuse for a while. However, I'm used to Ubuntu so I'd like to know what the major differences are between the two distros. Hopefully, that will allow me to make a better comparison. I'm not talking about differences like "this is more 'polished' than that" or " this looks cooler than that".. I want to if there are differences in the file system... the software..
On my windows console, i have installed sun virtualbox. I only have 768MB ram and so i downloaded a trial version of eboostr. I thought it would work, but when i tried an install of 11.2, also, i tried a debian install, it did the same thing, it froze up in the middle of the install with the error Code: unable to allocate and lock memory. I guess that eboostr doesn't work, but is there a way to use a flash drive as ram in virtual box?
openSUSE 11.2 installed on machine with 5GB memory but System Information in KDE desktop shows only 3GB total memory. Just added a further 4GB but no change shown in System Information.
Is there something I must do to have sysinfo report true value and does this mean that memory not shown is not being used?
I am getting a "No devices matches MBR identifier 0x8c71ad6e!" message along with a reboot in 120 seconds message. This occurs after kernel load and at the start of openSUSE boot.
I have checksum checked both images after download. I have attempted the install on CD and USB stick with both images and the result is always the same.
I've included a screen image on my site: No Devices Matches MBR Identifier | Badzilla
When the systen is running there is a program/utility/kernel-driver? kthreadd in system monitor all the following programs are a child process of kthreadd. Is kthreadd part of elevator=cfq and can kthreadd be influenced by params. kthreads uses no cpu time memory or shared memory so it must be in kernel. is there any documentation on this subject? System 11.3 on quad processor 8 gb mem nvidia-card and driver.
I would like to add the option mem=4096M to make some software works. but as long as I try with more than 512M, I have the error: error 28 selected item cannot fit into memory.
I've noticed since switching from Kubuntu 10.04 to openSUSE 11.3 on my laptop that the Adobe Flash plugin seems much more prone to crash and fullscreen video playback (e.g. Hulu) performs terribly. I can't figure out what's different between the distros. In both cases, I'm using a 64-bit OS, Firefox and KDE, and the proprietary NVIDIA driver. Anyone know what might be happening, or more importantly, how to improve Flash stability and performance on 64-bit openSUSE?
i am not able to start gnome desktop though most gnome applications work in icewm and xfce4 newly installed. the system config edit button i cannot understand what keys where to edit? how do we reinstall gnome fully? by the way my system is
intel 845 mother board[gigabyte] LG Studioworks 520Si monitor. 250 80 Gb disks 512mb mem LG dvd writer
