open source

ZFS Performance Testing: Dell PowerEdge 2950 III

In my previous post on ZFS performance testing I ran through various tests on a particular test system that I had running at the time. That system has come and gone to a better place in the proverbial cloud. This go round, I have a similar server with a different ZFS configuration. Lets dive in to the system and tests.

Test rig:

  • Dell PowerEdge 2950
  • Xeon Quad Core 1.6GHz
  • 8GB RAM
  • PERC5 – Total of 5 logical drives with read ahead and write back enabled.
  • 2x160GB SATAII 7200 RPM – Hardware RAID1
  • 4x2TB SATAII 7200 RPM – Four (4) Hardware RAID0’s (Controller does not support JBOD mode)
  • FreeNAS 0.7.2 Sabanda (revision 5543)-ZFS v13

GNU dd:
Tests performed from the CLI using good ole’ GNU dd. The following command was used to first write, and then read back:

dd if=/dev/zero of=foo bs=2M count=10000 ; dd if=foo of=/dev/null bs=2M

Results:
Results are listed as configuration, write, then read.

  • Reference run: 2x160GB 7200 RPM SATAII RAID1
    • 85.6 MB/s
    • 92.5 MB/s

  • ZFS stripe pool utilizing two (2) SATA disks
    • 221 MB/s
    • 206 MB/s

  • ZFS stripe pool utilizing two (2) SATA disks with dataset compression set to “On”
    • 631 MB/s
    • 1074 MB/s

  • ZFS mirror pool utilizing two (2) SATA disks
    • 116 MB/s
    • 145 MB/s

  • ZFS mirror pool utilizing two (2) SATA disks with dataset compression set to “On”
    • 631 MB/s
    • 1069 MB/s

    • Notes, Thoughts & Mentionables:
    There are a few things worth mentioning about this system:
    On the hard disk side of things, the hardware RAID1 was made up of Western Digital Blue disks while the other four (4) disks are Western Digital 2TB Green drives. If you have done your home work, you already know that the WD EARS disks use 4K sectors and masks this as 512byte sectors so that OS’ don’t complain. If disks are not properly formatted and/or sector aligned with this in mind, performance takes a tremendous hit. The reason for such inexpensive disks for this build is simple: This server is configured as a backup destination and as such, size is more important than the reliability that a SAS solution would provide.

    Compressions test results were, to say the least, quite interesting. It should be noted that the stripe and mirror pools performed quite similarly. Further testing of these results will be required, but it seems that the maximum score of 1074 MB/s was limited only by the CPU. During the read test all four cores of the quad core CPU were maxed. This becomes even more interesting when you compare the results of this two disk stripe pool with my previous findings on the six disk stripe pool running the same test. The earlier test rig scored much lower and it would appear to be the difference in CPUs that made such a strong difference.

    –himuraken

ZFS Performance Testing: Dell PowerEdge 2900

***Update***
This started a simple post to share a few results from various levels of zfs/array testing. Be sure to check back from time to time as I add additional configuration results.

***Begin Original Text***
I have been playing around with ZFS on various operating systems lately and have been trying to compare performance. I figured that sharing some of my results would give others something to compare with. Plus, I am on borrowed time with this unit, it is big, loud, and taking up free space and spare time in the home office.

Test rig:

  • Dell PowerEdge 2900
  • Xeon Dual Core 3.0GHz (HT Enabled-OS showing 4 cores)
  • 14GB RAM
  • PERC5 – Total of 7 logical drives with read ahead and write back enabled.
  • 2x146GB SAS 15K RPM – Hardware RAID1 for OS
  • 6x1TB SATAII 7200 RPM – Six (6) SATAII 7200 RPM Disks for testing. (Controller does not support JBOD mode)
  • FreeNAS 0.7.2 Sabanda (revision 5543)-ZFS v13

GNU dd:
Tests performed from the CLI using good ole’ GNU dd. The following command was used to first write, and then read back:

dd if=/dev/zero of=foo bs=2M count=10000 ; dd if=foo of=/dev/null bs=2M

Results: Each disk configured as a separate RAID0 array on controller.
Results are listed as configuration, write, then read.

  • ZFS raidz1 pool utilizing six (6) SATA disks
    • 133 MB/s
    • 311 MB/s

  • ZFS raidz1 pool utilizing six (6) SATA disks with dataset compression set to “On”
    • 414 MB/s
    • 359 MB/s

  • ZFS raidz2 pool utilizing six (6) SATA disks
    • 180 MB/s
    • 286 MB/s

  • ZFS raidz2 pool utilizing six (6) SATA disks with dataset compression set to “On”
    • 414 MB/s
    • 361 MB/s

  • ZFS stripe pool utilizing six (6) SATA disks
    • 190 MB/s
    • 263 MB/s

  • ZFS stripe pool utilizing six (6) SATA disks with dataset compression set to “On”
    • 429 MB/s
    • 381 MB/s

    • Results: Each disk configured as a member of a single RAID0 array.
    Results are listed as configuration, write, then read.

    • ZFS stripe pool utilizing six (6) SATA disks
      • 353.4 MB/s
      • 473.0 MB/s

    • ZFS stripe pool utilizing six (6) SATA disks with dataset compression set to “On”
      • 420.8 MB/s
      • 340.9 MB/s

      • Results: Each disk configured as a member of a single RAID5 array.
      Results are listed as configuration, write, then read.

      • ZFS stripe pool utilizing six (6) SATA disks
        • 322.0 MB/s
        • 325.9 MB/s

      • ZFS stripe pool utilizing six (6) SATA disks with dataset compression set to “On”
        • 438.8x MB/s
        • 371.8 MB/s

        • Results: Each disk configured as a member of a single RAID10 array.
        Results are listed as configuration, write, then read.

        • ZFS stripe pool utilizing six (6) SATA disks
          • 251.2 MB/s
          • 304.3 MB/s

        • ZFS stripe pool utilizing six (6) SATA disks with dataset compression set to “On”
          • 430.7 MB/s
          • 360.9 MB/s

          • Notes, Thoughts & Mentionables:
          It is worth noting that the results of the datasets with compression on can be a bit misleading. This is due to the source we are using with dd; /dev/zero. Feeding a string of zeroes into a compression algorithm is probably the best case scenario when it comes to compression. In real world conditions, data being read or written that is compressible would experience an increase in performance, while non-compressible data would likely suffer a penalty.

          I am hoping to conduct the same tests on the exact hardware in the near future. I will be switching the six (6) SATA disks over to varying hardware RAID levels and comparing them again.

          ***Update***
          In a follow-up post to this one, I concluded that compression read and write performance on this particular test rig was being limited by the CPU. I am hoping to swap out the current Intel Xeon 3.0GHz dual core for a quad core for additional comparison.

          –himuraken

Open Source for Schools

Why Linux and open source matters for small businesses and schools Hans Knobloch, Philadelphia Information technology Examiner

This blog was shared with me last night and I think the points made for using open source (linux) are right on target. The general public may or may not choose to accept this but business use of open source products continues to rise. Especially in this economic climate. (the reason this blog was started!) A quick Google search will reveal that there is an open source tool for just about any business need. This means freedom from initial software purchase, recurring license fees, etc.

Rockin’ on without Microsoft David Becker, Staff Writer, CNET News

This post is from 2003 and shows that even six years ago it was possible to convert to open source and continue as a successful business. It is easier today.
Linux is no longer relegated to the dark recesses of IT Geekdom. Major hardware vendors sell desktops and laptops pre-loaded with various distributions of Linux. A non-technical co-worker recently bought a mini-notebook loaded with Linux “because it was cheaper” and learned how to use it.
As virtualization projects grow in the SMB and education markets, there is often a natural progression towards open source related tools. Advanced configuration and troubleshooting of VMware ESX or Citrix XenServer hosts requires understanding and use of SSH and linux-based file systems.

Quoted from the blog post:

” Here are some facts about free and open source software and Linux:

  • A typical Linux and free open source software equipped PC will cost up to 60 – 70% less to operate over a typical three year write off time.
  • A typical PC hardware suited for Linux does not need to be as expensive as a typical Windows or Mac OS X PC. Linux requires much less resources. As a result, Linux PCs have an extended average lifetime.
  • Linux with typically installed quality programs, used for graphics, office, music, and Internet, are freely available as downloads from the Internet.
  • Linux versions, like Ubuntu, come with a host of no charge business software.
  • Pre-configured Linux versions, like the Ubuntu derivate Edubuntu, come with a host of no charge educational software, including teacher and student administration programs.
  • Linux installations can be upgraded and maintained without additional costs – ever.
  • Linux is much more stable and usually safer than its proprietary alternatives.
  • Linux is much less virus endangered than Windows or Apple OS X.
  • Linux is easy to learn and behaves in many ways just like Windows or Mac OS X.
  • Company specific software solutions are available where needed from commercial vendors.
  • Vendor supplied or vendor independent support programs are available if the need for professional support arises.
  • Hundreds of Linux end user and administrator forums on the Internet provide cost free support, configuration and installation how-to information, and help desk services.
  • Linux, with its open source character, is ideally suited to teach students the working of a modern computing system, better then any of its proprietary competitors. “

Openfiler iSCSI Share For CPS – Update

Original Post: Open Source Technology As Business Solutions: Openfiler – Symantec CPS Destination

We quickly noticed an issue with the iSCSI connector not automatically re-connecting the drive mapping when either the Openfiler server or the CPS server was rebooted. After some troubleshooting and most testing failed to resolve this issue, this project was scrapped in favor of using an Ubuntu NFS share as a backup2disk target.

Next step will be to test an NFS share as a CPS target.

Linux Administration Tool – webmin

For anyone that is not a Linux command-line expert (myself included!) webmin is an invaluable tool. A very quick install of this allows advanced administration from any web browser that can access the IP address of the server. Version 1.480 is the current release. Check it out and you will be pleasently surprised…

Instructions provided here are specifically for Ubuntu but can be adapted for any distro. (thanks to Himuraken for the proper syntax)

1.sudo apt-get install perl libnet-ssleay-perl openssl libauthen-pam-perl libpam-runtime libio-pty-perl libmd5-perl
2.wget http://prdownloads.sourceforge.net/webadmin/webmin_1.480_all.deb
3.sudo dpkg -i webmin_1.480_all.deb
4: open web browser, point to IP address of server on port 10000

For more info: Webmin

Ubuntu NFS Target for Symantec BE 12.5

Clearly Symantec is NOT open source. The solution to a common data backup problem can be found with open source tools however.

Issue: Backup2Tape provided long backup windows and an unreliable media. Soloution must use open source tools and existing hardware.

Solution: Ubuntu 9.04 (64-bit) NFS share running on Dell 2950 with six Seagate 1.5TB SATA drives in RAID5 configuration. Windows Services for Unix installed on Windows 2003 Server with Symantec Backup Exec 12.5.

NOTE: This will not work without this Symantec command: (verified in BE 11d and 12.5)
bemcmd -o116 -v -f”Name of your backup to disk folder” -e0 -m1 -d\\NFS-Server-Name\export-name

While the backup2disk is running fine, backups of Exchange Storage Groups are currently failing. All other data writes to the NFS share with no errors. Research continues…

Visit Righteous Hack Blog for a complete recap of the project: http://www.righteoushack.net/?p=384

Open Source Everyday – Firefox

One open source tool I use everyday is Firefox web browser from Mozilla. Currently on version 3.5, Firefox is perhaps the main alternative to Microsoft Internet Explorer. Now with the release of IE8 and the significant resources impact on many machines, there is an even stronger demand for a stable, easy-to-use web browser.
The open source security process that Mozilla implements for Firefox results in faster bug-fix and a quicker response to network wide threats. This is real security for the user. According to the Brian Krebs, Washington Post, 1/4/2007 blog post “Internet Explorer users Unsafe for 284 Days in 2006”, Firefox users only had nine days of risk. This is a huge difference! Security is the top priority for Firefox coders.
Comparisons such as installation file size show that Firefox i smuch more compact as well. Firefox 3.5 is a 7.7MB download, while IE8 for Windows XP weighs in at 16.1MB, more than twice the size. Firefox will also run in a portable mode from a thumb drive or other removable storage. This is excellent for travelers who like to carry favorites lists with them. All data is stored on the thumb drive when browsing, which also increases user privacy and security.

As a Network Manager I use Firefox for managing Cisco networking devices, as I have often experienced errors using IE in the past. Firefox connects to the device every time.

Firefox Resources:
Firefox Tweak Guides
Download Firefox
8 Easy Firefox Tweaks for Super Fast Web Browsing

TrueCrypt – Drive Encryption

I recently replaced my laptop and decided to encrypt the hard drive. I use Absolute Computrace for system recovery if lost or stolen, but this does not protect the data. Security experts recommend using encryption with recovery tools. The point of this is that if the hard drive is removed from the system with the recovery tool installed, the data can be accessed without activation the recovery features when reported missing.
After checking out several proposals for enterprise encryption packages, I remembered I had used TrueCrypt to encrypt a thumb drive. It worked great for that purpose so I gave it a shot on my new laptop.
Version 6.2a download is very quick (3.04MB) for Windows Vista/XP/2000. There are also options for Mac OS X and OpenSUSE and Ubuntu Linux distros. As with most open source tools, source code is easily obtained from the website as well.
Installation was painless and I was quickly ready for encryption. There are a couple of options at this point: encrypt entire drive including system partions, etc. or create and encrypted folder on the drive for critical data. This option will show a file on the directory listing that cannot be accessed without the password key that you generate. (NOTE: without the key (or Rescue Disk) the company website claims that there is no way to access data.) For key generation, I used a random tool I found on a Google search. Link is at the bottom. I created a recommended twenty character key. I chose to encrypt the entire hard drive. I followed all of the instructions, including backing up my few data files and proceeded with encryption. This took approximately five hours for a 160GB drive that is 30% full. Make sure you have plenty of power available and don’t need to use the system. Power failure during encryption will lead to data loss. The encryption completed, I rebooted, entered my key and XP loaded right up.
So far I have noticed two performance hits; system hibernate, XP completely locked up on recovery and I had to power cycle. I had the same issue copying 30GB of data to the laptop from a network share. Twice XP completely locked up. Other than that, normal operations seems to be fine.

So once again, an open source tool fits the need!

http://www.truecrypt.org
http://www.thebitmill.com/tools/password.html(Random Password Generator)