CentOS 5 Hardware :: Disk Write Performance Sustained (200 MBs)
Oct 27, 2010
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.
View 2 Replies
ADVERTISEMENT
Nov 18, 2010
After about an hour of using my Eeepc 1000 there is random sustained SSD access. How can i check what is accessing the SSD so much?
View 3 Replies
View Related
Sep 12, 2010
I've been having some problems copying files to USBs. If I'm copying a large (100MB+) amount of data, at random points the transfer will just stop for 30+ seconds before continuing. Sometimes it doesn't start up again at all. Consequently, the write speed drops to less than 1MB/sec, sometimes as low as 100 KB/sec. I do not have these problems on Windows 7, where I achieve speeds of ~16 MB/sec easily. I have had the same results with several USBs (2-32 GB) on several file systems (fat32, ext2) with several different computers running fully patched versions of Ubuntu 10.04, which suggests the problem is related to the way the OS accesses the hardware.
View 6 Replies
View Related
Nov 16, 2010
I wasn't sure where to post this question so administrators, feel free to move it.I have a media server I set up running Ubuntu 10.4 Server, and I set up a software raid 5 using 5 Western Digital Caviar Green 2TB 7200RPM 64MB drives. Individually they benchmark (using the Ubuntu's mdadm GUI (pali?somthing...) at about 100-120mb/s read write.I set the raid 5 up with a stripe size of 256kb, and then I waited the 20 hours it took to synchronize. My read speeds in raid are up to 480mb/s, but my write max is just under 60mb/s. I knew my write performance would be quite a bit lower than my read, but I was also expecting at least single drive performance. I have seen other people online with better results in software, but have been unable to achieve the results they have gotten.
My bonnie++ results are more or less identical (I used mkfs.ext4 and set the stride and stripe-width).The PC has 2048mb of RAM and a 2.93Ghz Dual Core Pentium (Core 2 Architecture), so I doubt think that's the bottle neck. These drives are on the P55 (P45*) South Bridge SATA controller.
View 7 Replies
View Related
Sep 7, 2010
I'm currently experiencing some serious issues with WRITE performance on a RAID-1 array. I'm running Ubuntu 10.04 64 bit server with the latest updates. To evaluate the performance ran the following test: [URL]... (great article btw!) Using dd to measure, write performance is only at 8.7 MB/s. Read is great though at 74.5 MB/s. The tests were ran straight after rebooting and I have not (YET!) done any kernel tuning or customization, running the default server package of the Ubuntu kernel. Here's the motherboard in the server: [URL]... with a beta bios to support drives over 300GB.
[code]...
As you can see from the bo column there is definitely something stalling. As per top output, the %wa (waiting for i/o) is always around %75 however as per above, writes are stalling. CPU is basically idle all the time. Hard drives are quite new and smartctl (smartmontools) does not detect any faults.
View 4 Replies
View Related
Jan 5, 2011
I have recently migrated my file server over to a HP Microserver. The server has two 1TB disks, in a software RAID-1 array, using MDADM. When I migrated simply moved the mirrored disks over, from the old server Ubuntu 9.10 (server) to the new one 10.04.1 (server).I Have recently noticed that write speed to the RAID array is *VERY* slow. In the order of 1-2MB/s order of magnitude (more info below). Now obviously this is not optimal performance to say the least. I have checked a few things, CPU utilisation is not abnormal (<5%) nor is memory / swap. When I took a disk out and rebuilt the array, with only one disk (tried both) performance was as to be expected (write speed >~70MB/s) The read speed seems to be unaffected however!
I'm tempted to think that there is something funny going on with the storage subsystem, as copying from the single disk to the array is slower than creating a file from /dev/zero to the array using DD..Either way I can't try the array in another computer right now, so I though I was ask to see if people have seen anything like this!At the moment I'm not sure if it is something strange to do with having simply chucked the mirrored array into the new server, perhaps a different version of MDADM? I'm wondering if it's worth backing up and starting from scratch! Anyhow this has really got me scratching my head, and its a bit of a pain! Any help here would be awesome, e-cookies at the ready! Cheers
View 6 Replies
View Related
Oct 27, 2010
So I have been doing some RAID 5 performance testing and am getting some bad write performance when configuring the RAID with an even number of drives. I'm running kernel 2.6.30 with software based RAID 5. This seems rather odd and doesn't make much since to me. For RAID 0 my performance consistently increases as I add more drives, but this is not the case for RAID 5. Does anyone know why I might be seeing lower performance when constructing my RAID 5 with 4 or 6 drives rather than 3 or 5?
View 2 Replies
View Related
May 20, 2010
Compared to my laptop notebook with a HD of 5400rpm, the write performance of raid1 on an ubuntu lucid server is unacceptable. In the begining, I installed ubuntu 9.04 server(alternate) using raid1 with two WD 1TB HDs of 7200rpm(Green Power) and then performed dist upgrade to 9.10 and then to 10.04.
I guess the write performance initially was reasonable since the installation and data migration(copy from another computer over LAN) didn't take too much time. However, after upgrading the server to 9.10 or so, I found large file upload through samba or ftp tends to block and time out. It is of no use whether to change the daemon or the client program so that I tried to test the read/write performance on the server to figure out the situation.
To my surprise, using strace I found even a simple program like cp would easily get blocked eventually in a write() system call for decades of seconds. Hence, I perform another disk writing test using dd for data size ranging from 50MB to 1GB. Performance test commands are listed as follows:
Quote: dd if=/dev/zero of=test.img count=[5|10|15|20|100] bs=10M
if the data to write is equal or fewer than 150MB, the command returns immediately at very hight speed but the raid disks starts to sync and busy so that the terminal prompt seems to freeze. I think this behavior is normal under the raid1 configuration, isn't it?
But when the data size is equal to 200MB, the test command blocks for seconds and the write speed is measured at about 16.6MB/s. Of course, the raid disk still starts to sync and busy afterwards. Next, I test writing with data of size 1GB. The command blocks so long for about 770 seconds(<2MB/s) while the same test runs for only 17.49 seconds(60MB/s) on my laptop.
I also burn a Lucid LiveCD to boot the server and mount the raid device to run the test again but the results remain similar. Does that means even I re-install the system on the raid, the problem never disappears?
PS: the disks run under the mode of UDMA6 without change.
View 2 Replies
View Related
Mar 21, 2011
I have a netbook I'm not using and which I transformed into a server with Apache, Tomcat6, Netatalk, Webmin, BIND9 and Tor.
Problem is, the disks never stop spinning because all of the programs write a few kb at least every few seconds to disk, even when nobody is connected to it.
My question is: Is there a way to have the computer boot from disk like normal (maybe even a squashfs), keep ALL CHANGES to ram and then save to disk when either the ram is full (unlikely because the server is rebooted every few days) or at shutdown?
I thought about a mixture of ramfs and unionfs but I'm not good enough yet...
View 4 Replies
View Related
Jun 7, 2011
I'm seeing sustained disk writes of about 2 MB/s in the indicator-multiload indicator in Unity. I determined that it is writes on my 500GB HDD on /dev/sdb. This behaviour started after I used Gparted to create a single 500GB ext4 partition and also selected that it should be formatted to ext4 in Gparted.
Is this usual? It also survived a reboot.. I assume that it is the full formatting taking place in the background?
I saw no activity using pidstat or iotop. Only using vmstat -d revealed the writes.
View 3 Replies
View Related
Jan 29, 2010
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.
View 1 Replies
View Related
Apr 6, 2011
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.
View 5 Replies
View Related
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?
View 5 Replies
View Related
Oct 17, 2010
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.....
View 1 Replies
View Related
Aug 12, 2010
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?
View 2 Replies
View Related
Oct 18, 2010
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:
Code:
[12879.434115] irq 22: nobody cared (try booting with the "irqpoll" option)
[12879.434121] Pid: 0, comm: swapper Tainted: P 2.6.35-22-generic #34-Ubuntu
[12879.434124] Call Trace:
[12879.434126] <IRQ> [<ffffffff810cba5b>] __report_bad_irq+0x2b/0xa0
[12879.434137] [<ffffffff810cbc5c>] note_interrupt+0x18c/0x1d0
[Code]...
View 3 Replies
View Related
Jan 4, 2011
df -h [URL] I did the following command to find everything is in /usr or /var, then tracked it down to /usr/lib and /usr/share as the main offenders, but out of all the directories none are more than 1mb or so.
du -sh /* | sort -gr | head -n 5
I tried to uninstall firefox, which is what got me in this mess in the first place, the log claims it will remove ~240 mb but failes on a "E: Write error - write (28 No space left on device)" [URL] If I could juggle something onto an external hard drive so I can uninstall firefox I would be out of the wood. Failing that I believe a new install is in order.
View 1 Replies
View Related
Aug 20, 2011
I'm experiencing slow disk speeds on my server. Since I'm not too experienced in that area, I searched the forums and found a similar thread: Very slow SATA on Centos 5.5 Based on that thread I ran hdparm on all of my disks, with the following results:
[Code]....
View 9 Replies
View Related
Apr 5, 2011
I am using kernel 2.6.32.21, and my hard disk is West digital WD10EARS-00Y, 1TB. This disk is just for data, I made 2 partitions on it, each has half. And I have another small disk for system. I am using ext3.
this is my fdisk /dev/sdc
Device Boot Start End Blocks Id System
/dev/sdc1 1 60800 488375968 83 Linux
/dev/sdc2 60801 121601 488384032+ 83 Linux
[Code]....
I ran some dd to test the write throughput on /dev/sdc2. If I run it in /data2, I got around 70MB/s. If I create some directory, say /data2/dir/, and run dd again, I might get 60MB/s. Sometimes I still get 70MB/s, sometimes I get 60MB/s, differs for directories.
I wonder if this is because the allocation policy of file system, ext3, or this is from my hard drive?
View 7 Replies
View Related
May 19, 2011
In Short: I want something that will show me which files and associated requested URLs are causing the highest load on our web server.
Verbose: We already monitor our Apache web server using Cacti, Nagios, and of course traditional nix commands, but I am looking for something specifically to take an Apache process ID and tie that to a users request for a specific file and URL. The great thing about something like this is that it would be able to show me the worse performing PHP files. Special bonus if it can store this data over time for reports.
I have been googling around for something like this but can't seem to find it. Bonus if it exists in Cacti or Nagios already and I'm just to blind. I started writing my own PERL script to do this, but have limited time to devote to this at work so if a solution already exists I'm game, else I'll just have to write it myself. I'm also worried about my own scripted solution in that it won't get it write because it will be leveraging the output of a specialzed apache log that records the PID and then doing a look up with ps aux looking for that PID.
View 1 Replies
View Related
Apr 1, 2010
I have formatted to EX3 in gparted all went fine, however I cannot rename the disk or write anything to it as access is locked to root Furthermore a file has been created called "lost and found" its locked and 46GB in size - what is this?
View 14 Replies
View Related
Jun 20, 2010
I have Ubuntu 10.04 installed on my Armada laptop with a Teac CD-W24E internal and a Plextor-DVDR PX504A via USB. I use K3B for writing CD data projects to 700mb CD-RW 1-4x discs. K3B writes ok on both drives if the inserted disk is a blank disk. If I insert an appendable disk with lots of space left, it asks to blank it before writing. I find no way to write on an appendable disk with K3B. Is this a software limitation or am I missing something?
View 9 Replies
View Related
Mar 17, 2010
Well appears to be that i have two partitions..one of 300gb i did on windows time ago and one of 100gb where i place my DEBIAN LENNY 5.0, the problem is that i cannot write into it..only i can copy stuff from it.
View 13 Replies
View Related
Dec 18, 2009
I have a user complaining about performance:-
Quote: At times on our server, during more-or-less idle moments, context switching is around the 2000-3000 mark.When we run the app, we can get this up to the 500,000 CSs per second. Not only this, but at the same time, the CPUs go to >60% system (the kernel is having to do all that context switching after all), leaving only around 30% for user space. We think it is a possibility that the massive context switching is contributing significantly to our performance issues.
The server is a HP DL580 G5 with 4 x Intel Xeon E7340 @ 2.4Ghz.My issue is that I don't know if the contect switching is a problem or not.From what i've read those values are not necesserilly cause for concern.
View 1 Replies
View Related
Mar 26, 2010
I just installed CentOS 5.4 on a brand new DL160 G6. In order to get the data raided I decided to go with mdraid. But I see quite harsh performance hits, I have no GUI or anything and just the basic server installation, and even the console feels sluggish.
The output from getinfo.sh disk:
==================== BEGIN uname -rmi ====================
2.6.18-164.15.1.el5 x86_64 x86_64
==================== END uname -rmi ====================
[code]....
View 3 Replies
View Related
Apr 18, 2010
I am trying to scrape all of the info off of an old Fedora disk so that I can referb the box and put new stuff on it, but for some reason I can't get the permissions (even as root) to write to the disk.
It is currently mounted under /media/backup and I can browse it all I want, but still can't write.
View 5 Replies
View Related
May 18, 2010
Upgraded from 8.04 LTS to 10.04 LTS, but can't get the sound working.Have downloaded the 10.04 Ubuntu CD image to installwhen I try to follow the directions at[URL]BurningIsoHowto for burning the image............it says to Right click on the ISO image file and choose Write to Disc. I have no WRITE TO DISC option when I right click.Will some kind soul please tell me how to burn the downloaded CD image to a CD?
View 6 Replies
View Related
May 18, 2011
I am doing a program that reads data from a gps and some other devices and writes some files with all the information. When I run it normally it works fine, but if I run the program in the background (with the ampersand) files are not created until I bring it to foreground or close it. I am confused, the program should run the same way with and without the ampersand. (Could it be that the main process that creates all the threads does not create them when it is executed in background? It seems like if the program is stopped until I take it to the foreground).
View 4 Replies
View Related
Apr 12, 2011
I'm only somewhat new to linux but I still don't have a real grasp of it's deep innards and I had a fairly outlandish idea that I'm wondering whether is possible/plausible or not.I want to run a game server on CentOS that has a very high dependency on fast writes-to-disk. Disk writes are pretty much the single bottleneck in this server.First I looked at allowing a high queue of writes to pile up before it flushes them to the disk, but I read that this causes fsync, which is still used commonly, to take a very long time.
I've been thinking about the possibility of running the server on a RAM disk, but I still want changes to be saved to non-volatile storage. Not all at once, but have it actively write the changes to disk. The hope is that this would smooth out the peaks and valleys of write activity and improve overall performance, but I have not seen this idea discussed anywhere.
So my question is, is there any plausible way to continuously copy writes to a RAM disk to a physical drive without slowing down the speed of the writes to the RAM drive below the speed of said RAM? Or is there a better way to obtain this sort of performance, short of investing in expensive equipment?
View 1 Replies
View Related
Oct 26, 2010
I have encountered this problem of memory usage is increasing as the during the my program is being run until 1Mb is left then it stays at that.A part of the program is this:
Code:
#define WRITE_BUF_SIZE (1024*1024)
void post_data(const void *data, unsigned long size)
[code]....
View 3 Replies
View Related
