General :: Disk Performance Slower After Upgrade To 4GB RAM
Oct 19, 2010
I am running openSuse 11.2 (32-bit), my CPU only supports 32-bits. I have a hardware RAID device. My system has 4GB of RAM. When I configure my system to only use 3GB, 2GB, or even 1GB, using mem=1024M in grub, my RAID performance is much better then when letting my system use the default 4GB available.Can anyone explain to me why this is? Is there anything I can do, i.e. kernel configuration, that will help performance when running with all 4GB enabled?
I upgraded my old Kubuntu installation to 10.10 Maverick Meercat and I am now experiencing a really annoying problem. I boot my computer and everything seems fine for a while, but eventually my disk performance drops to horrible levels. It's not gradual. It's fine one second and then abysmal the next.
If I do "cp file1 file2" and then kill the cp process after 10s, there are only a couple of MB copied.When I run dmesg after the performance degradation, I see this:
taking out the windows master drive while installing ubuntu on the slave drive as i'm not entirely sure about the reasoning -- something to do with a GRUB confliction.
It used to have a lightening speed. But now every application is taking few seconds to start, even on Konsole, the commands take some seconds to get typed ! (I literally have to wait for cursor to move and commands to get typed on Konsole). Folders are taking some seconds to get opened !
I rebooted the computer but that proved to be useless.
Here the ps output:
Code: anisha@linux-uitj:~> pstree -Apuh `id -anisha` id: invalid option -- 'i' Try `id --help' for more information. init(1)-+-acpid(1054)
I've just bought a 6-core Phenom with 16G of RAM. I use it primarily for compiling and video encoding (and occassional web/db). I'm finding all activities get disk-bound and I just can't keep all 6 cores fed. I'm buying an SSD raid to sit between the HDD and tmpfs. I want to setup a "layered" filesystem where reads are cached on tmpfs but writes safely go through to the SSD. I want files (or blocks) that haven't been read lately on the SSD to then be written back to a HDD using a compressed FS or block layer.
So basically reads: - Check tmpfs - Check SSD - Check HD
And writes: - Straight to SSD (for safety), then tmpfs (for speed) And periodically, or when space gets low: - Move least frequently accessed files down one layer. I've seen a few projects of interest. CacheFS, cachefsd, bcache seem pretty close but I'm having trouble determining which are practical. bcache seems a little risky (early adoption), cachefs seems tied to specific network filesystems. There are "union" projects unionfs and aufs that let you mount filesystems over each other (USB device over a DVD usually) but both are distributed as a patch and I get the impression this sort of "transparent" mounting was going to become a kernel feature rather than a FS.
I know the kernel has a built-in disk cache but it doesn't seem to work well with compiling. I see a 20x speed improvement when I move my source files to tmpfs. I think it's because the standard buffers are dedicated to a specific process and compiling creates and destroys thousands of processes during a build (just guessing there). It looks like I really want those files precached.....
I have doubts regarding storage: How to configure the Events of Storage Processor? What are performance issues will come daily in a critical production server? What are first steps for disk performance Check? What are first steps for Storage Processor performance Check? What are first steps for MetaLUN performance Check?
I recently installed the KDE (Kubuntu) desktop via Synaptic Package Manager. I have a few questions about this: Do I have to update Kubuntu when I update Ubuntu in October, or will it update on its own? Does running Kubuntu simultaneously on an Ubuntu machine cause any issues (slower performance, conflict in programs, etc.)? How do I uninstall Kubuntu if I decide that Ubuntu works just fine for me?
I have Etch running in my laptop. I found that if the laptop is only running on battery, the disk seems to performs slower than if I running on a power adapter. How do I possibly check the setting and fix it?
I'm the only user on the system most of the time and I keep it as minimal as possible so it's not other services and programs keeping the disk busy.
I am using OpenSuse 11.2 (64 bits) with KDE 4.3.1. My hard disk partition contains 3 areas : / + /home + swap. Frequently the / area is filling with, I suppose, temporary files until it is full (15 Go) and that in a very short time, typically about 5 mn. Then, these files are cleaned by the system, when disk space is needed I suppose. The point is that during this time, the system goes slower and slower until the "cleaning". That seems to occur only when Firefox (3.5.6) is running so I suppose that these files are temporary internet files.
Not sure if i am correct but I was under the impression from all sorts of posts & videos that 10.04 was supposed to have a fast bootup. I have upgraded from 9.04 jaunty bootup is quite a bit slower than on the old version and is quite tedious is this correct? are there any logs i can upload to see if i have an error some where?
I've been using fedora on my Lenovo R61i since I got it...I believe the first version I used on this laptop was f10. I always try to keep it on the latest distro and I also enjoy a clean install every year or so. One thing I was always impressed with was how quickly I could go from sleep to back up and online. So I decided it was time for a clean install this weekend to fedora 15. Now when I wake it up (open the lid), it takes over a minute for the wifi to connect again. This hasn't been the case for the last 4 distro's so I'm wondering what I can do to speed that process back up, is there a particular package I can downgrade or is this something on their radar already and I should just wait for an upgrade?
kernel upgrade to 2.6.32-23-generic after i did this ... when im using 2 vmwares, it massivesly laggs my main OSS (ubuntu) then i went back to 2.6.32-22-generic <- and all came back to normal..when im running 4 Vmwares, it still doesnt lagg my main oss
On one of my computers I did something I rarely do--used the version upgrade method. I went from 9.10 to 10.04. I normally upgrade by burning a CD, wiping / and clean installing the new version.
This HP desktop was BLAZINGLY fast prior to this; now it's slower than frozen molasses.
Seriously, it's awful. It's like everything freezes, including the clock--say it's 1:11:23, it stops there and when it starts again it'll be 1:14:05 (or whatever!). And it does that continually. Meanwhile the trackball works as far as moving the pointer, but it's impossible to actually *do* anything.
I'm really, really glad I left my other four computers running 9.10 for the time being.
I'm going to do a fresh install with a CD, but I'm just curious if anyone else has had this experience after doing a version upgrade? I did it this time because I'm lazy and it was so easy!
how much of a performance impact full disk encryption (say, AES 256-bit) has on disk-related activities? On one particular project I'm involved in I am trying to weigh out security vs performance issues.
Ubuntu 10.10 is dual booted but it is my primary OS.
Unfortunately it's on the outer edges of the disk in an extended partition.
This has always bugged me, with regards to read/write performance.
Do my concerns of reduced performance have any foundation? Should i bite the bullet and format the drive installing ubuntu first?
I ran the disk read benchmark and my read speeds were 100MB/Sec at the beginning of the test to just 55MB/Sec at the end. I have no idea if the position of the test has any bearing on the position of the disk or whether the speed recorded is affected by other factors such as the tests function or simulation.
Im experiencing a strange problem with 10.04. When it first boots up its reasonably quick (as snappy as Gnome ever has been) but after its been running for a day or two it gets ununsably slow and has to be rebooted.
I dont see how it can be anything to do with my hardware as it runs OK when first started. Ive monitored RAM usage and it doesnt seem to differ much. There's always the same amount unused.
I am experiencing disk write performance issues and I cannot find the cause. I have LSI-9211-8i SAS 2 controller (latest firmware), Centos 5.5 latest x86_64 kernel (2.6.18-194.17.4.el5 #1 SMP with latest LSI driver v. 7.00 datet Jul 27) and Seagate Cheetah ST3600057SS drives. These drives have a std write performance (sustained) of > 200MB/s (and read as well); with Fedora core 13 (same machine), issuing a dd if=/dev/zero of=/dev/sdo bs=1024k count=16384 (16 GB direct device write), gets normally to 213 MB/s (repeated retries). On Centos 5.5 I am getting speeds around 110/113 MB/s. iostat does not show anything specific (just 1.3 % wait, CPU 99.7 idle). There are 14 drives: tried with several of them, same figures. Reads go around 200 MB/s.
I am having a lot of performance problems since my upgrade my ATI board does not get properly supported (tried a lot but nothing really helps)my keyboard typing is horrable : i have to see at each click of it got accepted ; cannot type text properly screen performance is bad - often the screen greys out and back to color (probably because of driver)
Since I upgraded to Slackware64 13.1 my system feels slower than before (Slackware64 13.0).I paid attention to the documentation (changes_and_hints.txt, readmes etc.) regarding the update. The only thing I changed is, that I used the KDE-4.4.4 packages of AlienBob instead of the KDE-4.4.3 packages of Slackware.
But since AlienBob at his blog wrote, "Although it seems (by looking at the changelog between 4.4.3 and 4.4.4) that there were no spectacular updates in the latest stable KDE Software Compilation, I got some feedback that 4.4.4 does feel �snappier� than the 4.4.3 which is part of Slackware 13.1.", I thought it would be a good idea to use these packages.
I also recognized a few crashes in KDE, which I didn't have before so often. Therefore I moved the .kde to .kde_bak and configured kde new. This improved the behaviour regarding the crashes, but it still seems not that stable to me like the KDE 4.2 in Slackware64 13.0.Did anyone recognize similar problems after the upgrade, or can give some hints how to solve it?
I would like to re-raise this topic that I have seen on other discussion boards but seems to be going no where.I have been using Ubuntu for many years now having switched from other distributions. All my machines ran some version of Ubuntu, until just recently. I upgraded my personal HP TX2500 from 9.04 to 10.04. No problems. After 10.10 had been out for a while I upgraded again. The wireless would drop when busy/(signal slightly weakened) etc. It would not function again unless the wifi module was reloaded.
Then I installed 10.10 on my HP NC6400 the same thing occured. I installed Ubuntu 10.04 instead. It seemed no better. After some Googling, it seemed I was not alone. I tried all the suggestions I could find and none of them worked. One suggestion was to install Fedora 14. This works perfectly on the nc6400. The HP TX2500 has been reinstalled back with 10.04 where the wireless seems ok and I assume I am stuck there for a while.
Now my question is why would the networking behave this way for Ubuntu 10.10 (10.04 as well on the NC6400) but wireless on Fedora 14 is perfect (I mean really good! Better than either Ubuntu 10.04 or 10.10 on any of my laptops) on the NC6400. I am hoping that we can find a way to fix this, otherwise I will likely have to drop Ubuntu completely because I do not have an upgrade path for laptops. There must be many other users in the same position who are not discussing it and just rebooting each time this occurs. This must give a negative perception of otherwise great software.
I have bought one of the new Logitech wireless mice for my laptop, but it seems too "fast" - a small movement results in massive movement on the screen, its much faster than the old corded mouse I am using with my desktop.
I just wanted to know if having my laptop set to ondemand, will this affect performance in any way? I realize it increases the clock speed to performance when the CPU is under load, but does the time it take to go from ondemand to performance affect speed? Will there be any noticeable difference between the two setups? I have a dual core intel at 2.2GHz when in performance. When ondemand is set with no load it downclocks to 800Mhz.
While converting some images using ImageMagick, I noticed a somewhat strange effect. Using xargs was significantly slower than a standard for loop. Since xargs limited to a single process should act like a for loop, I tested that, and found it to be about the same.
Thus, we have this demonstration.
Quad core (AMD Athalon X4, 2.6GHz) Working entirely on a tempfs (16g ram total; no swap) No other major loads
I have a big archive with about 10000 documents in a usb stick. What I have noticed is that browsing of that archive with gnome is much slower with ubuntu than winXP ( dual boot , same PC ) where it is almost instant.I have disabled assistive technologies and installed Thunar file browser. It improved things but again the lag is important. Linux is in general much faster than windows, so I wonder why is it happening ?
Recently installed debian lenny.Running very well.WM: ratpoison.Suddenly, today, the internet runs much slower.ONLY ON THIS MACHINE - the parallel windows machine, off of which I was successfully weaning myself, is not having this problem, so it's neither the router nor the actual internet connection.Specifically: I can't attach a document in yahoo mail.I tried it in iceweasel, same problem, so it's not the browser.I piped the output of "top" for a minute or so to a text file, which I'm attaching.
I've found myself using the -v flag for lots of applications less and less (especially for trivial stuff like tar and cp). However, when I did and I was, say, unzipping a large file, it would take longer than when I didn't use the -v flag.I assume this is because the terminal has to process the text and I'm filling up whatever buffer it might have. But my question is, does this make the application actually run slower or does it complete in the same amount of time and what I'm seeing is the terminal trying to catch up?
Debian lenny, old install (I've upgraded to lenny when it was just about to become the stable release), versus windows 7, fresh install.Comparing browsers speeds with numion.com/Stopwatch.html, I had results such as:Iceweasel (firefox) on linux: from 9.154 seconds to 21:860 seconds (the same webpage, reloaded)Firefox on windows: 4.32 seconds - and never much slower than thatThe fastest browser on linux was Opera, ranging from 8.562 to 5.503 secs to load the same page, but even internet explorer beat/match it with its timing of about five seconds.
I have not other browsers on windows; on linux there is aroraonqueror (KDE3), kazehakase, chrome, and dillo, besides text browsers. I didn't test on dillo; Kazehakase and chrome were the only ones which had nearly decent results, but still very bad, 11 to 13 seconds for chrome, and 21 for kazehakase. Konqueror just seemed to never finish to load the page, I gave up when it was still loading somethingfter nutes and 5 seconds.'ve emptied the cache every time I would test, and I was running almost only the browsers and not much else. Whatever comes by default on windows, and on linux, I was on openbox, with nothing much going on I guess, I think the most memory consuming processthe time, besides xorg and the browsers themselves, was dictd.
I've researched a little bit about, but not enough to make a list of possible things to change in order to improve the speed on linux. Most of the time there are people just agreeing that on windows the rendering is faster, and other people saying that with them is the opposite, with some minor variations like people saying that linux is faster for plain downloads while windows is faster for web browsing due to better graphics.
(by the way; I haven't installed any graphic card driver on windows, which is still running on 1024x768, while linux runs on 1280x1024, with the "nv" generic driver, without fancy options, not supported by my old card) The closest to a suggestion of possible solutions was someone saying tha compilation may affect performance, I guess it was both about kernel compilation, and the web browsers themselves.I'll google a bit more about how to "compile for speed", both kernels and programs (maybe the x server
I am new to linux but will try and be as precise as I can. I have a Dell Insp. 1545 with 3GB ram with 10.04_64
Problem: Rather out of the blue upon login, from the time of entering password to the desktop, is MUCH slower. I have no splash enabled and there is some reference to AppArmour profiles (it sits there at this apparmour reference , then sits trying to load desktop). However, I have never configured apparmor. Finally, on last login, there was some sort of screen corruption prior to getting to desktop.
Prior to all this i noticed that firefox "blinked" when I was using it, and then said that it needed to shutdown/restart for updates. I did notice at that time two of my previous tabs sortof acted oddly (moved around then disappeared).
I have no SSH/Server software etc activated.
I did change visudo so that sudo timeout shorter than it was. However that is the only modification to the system recently.
