Here’s a quick update on my previous results for striping and RAID-Z when testing ZFS on an old system with multiple SCSI drives (see the previous post for details of the system config).
If you are a business person who is new to the world of Open Source, or you need to introduce business people to Open Source and what it can do for you then the Specialty Equipment Market Association (SEMA, an automobile speciality parts trade association) has a very good introduction to Open Source called Open Source Explained : What it is — What it isn’t — and How SEMA Members Can Benefit From It.
Using open source is like owning a 1960s musclecar. You can drive it just as it is, or you can tweak and modify it to the limits of your (or a friend’s) time and talents. On the other hand, a closed source program is quite the opposite. The work of the people who create the software is not available for tweaking or performance enhancements by the general public. It’s like getting a car with the hood welded shut by the manufacturer. “Here are the keys. Put gas in it, and call us if it breaks.†Closed source has its place, and it might be the only choice in certain instances.
As well as being a gentle introduction to what Open Source is it covers all the sorts of things that businesses care about, servers, oeprating systems, databases, word processing, desktop publishing, CRM, ERP, etc. Well worth a read even if you are involved in Open Source because you’re almost certain to come across something you’d not heard of before.
The author ? Well he’s not a professional geek, he is:
Walter D. Vaughan, Jr. is a vice president at Steele Rubber Products. He is a current member of SEMA’s Business Technology Committee, an ARMO Select Committee member and a recipient of ARMO’s 1998 Industry Person of the Year Award. He also uses or has tested extensively all the software mentioned in this article.
Well I’m happy to say that today I got the Request For Proposals (RFPs) out of the door for the proposed replacement of VPACs ageing Linux cluster Brecca (180 cores of 2.8GHz Intel Pentium 4 Xeon) and a significant amount of additional storage. Needless to say the new cluster must run Linux!
Please see the announcement on the VPAC systems blog for more information and copies of the RFPs.
This has been keeping my very busy recently and so I’m looking forward to a nice quiet break over Easter!
If you thought SCO couldn’t stoop any lower, think again. They have filed a motion in SCO versus IBM saying they wish to depose PJ, the creator of Groklaw.
I can say this: SCO in its wisdom has just guaranteed that the judges in SCO v. IBM and SCO v. Novell will have to read Groklaw. So, welcome Judge Kimball. Welcome, Judge Wells. We’ve enjoyed very much learning about the law by watching you at work. SCO told you something that isn’t true. No one tried to serve me that I knew about. No one informed me of any deposition date. That is true. It doesn’t feel so nice to be smeared like this, I can tell you that, and to have to pay a lawyer to deal with this harassment. I view it as such, as a kind of SLAPP suit, a vendetta to pay me back for blowing the whistle, and to shut Groklaw up. SCO wants to put a pin on a map and point to it and say, “Here’s PJ.” Then someone drops by and shoots me, I suppose. I certainly have nothing to tell them that is relevant to this litigation.
Basically SCO have gotten so fed up with PJ and the various other Groklaw contributors poking huge holes in the farcical SCO law suit that they have convinced themselves that the site is a front for IBM and that PJ doesn’t exist and now want to prove it. Sadly for them their fear-induced paranoia can’t change fact into fiction and so, as usual, they’ll loose eventually but they want to make life as painful as possible for anyone who dares to laugh at the emperors new clothes.
I do hope that this motion doesn’t succeed, but I feel that SCO will find it rather painful for their reality if it does.
RMIT came back online at around 09:30, hopefully it will last!
VPAC systems are unreachable from the outside world as it appears that all RMIT networks failed at around 8am.
This means that the Linux Users of Victoria (LUV) server is also down as it is hosted at VPAC, so no LUV email or website for the moment.
The systems themselves are still functioning normally, just needs the RMIT ITS networks folks to track down the problem and fix it (good luck people!).
Having upgraded to the Feisty beta I thought it would be fun to see what (if any) affect it had on filesystem performance (especially given my previous aide memoir).
For these tests I stuck to my 3 favourites, JFS (from IBM), XFS (from SGI) and ZFS (from Sun, ported to Linux using FUSE by Ricardo Correia due to Sun’s GPL-incompatible license). This is a follow on from a slew of earlier ZFS & XFS benchmarking I did reported on previously (( here, here, here and here )).
Summary: for Bonnie++ JFS is fastest, XFS next fastest and ZFS slowest and Feisty made XFS and ZFS go faster (didn’t record my previous JFS results sadly).
The fact that ZFS is slowest of the three is not surprising as the Linux FUSE port hasn’t yet been optimised (Ricardo is concentrating on just getting it running) and is also hampered by running in user space. That said it still manages a respectable speed on this hardware and does have useful functionality that makes it useful to me.
Now that Ubuntu Feisty (which will be version 7.04) has just entered beta I decided to take the plunge and upgrade my home machine to it – having done a full backup of all my data onto two different media first!
I’m a bit behind at the moment, but this is something worth a mention.
Ricardo Correia’s port of Sun’s ZFS (which I’ve been playing with for a while) has finally reached its first beta release!
He’s made some useful performance improvements recently as well as tidying up some of the memory handling and fixing various bugs, including that bug from the upstream that yours truly got bitten by.
Here’s an updated Bonnie++ run for comparison.
Version 1.03 ------Sequential Output------ --Sequential Input- --Random-
-Per Chr- --Block-- -Rewrite- -Per Chr- --Block-- --Seeks--
Machine Size K/sec %CP K/sec %CP K/sec %CP K/sec %CP K/sec %CP /sec %CP
inside 2G 18838 5 6995 2 18277 2 144.4 0
------Sequential Create------ --------Random Create--------
-Create-- --Read--- -Delete-- -Create-- --Read--- -Delete--
files /sec %CP /sec %CP /sec %CP /sec %CP /sec %CP /sec %CP
16 2795 5 9658 9 3462 5 2739 5 13736 11 4015 6
inside,2G,,,18838,5,6995,2,,,18277,2,144.4,0,16,2795,5,9658,9,3462,5,2739,5,13736,11,4015,6
real 10m12.258s
user 0m0.840s
sys 0m18.605s
Compared to previous results the write speed has improved, but the read speed seems to have dropped off a bit. Still, it’s early days yet.
Dear Lazyweb,
Just been playing with SmartMonTools under Linux to monitor the attributes of my SATA drives (something I’d not bothered to do since upgrading from my old Adaptec SCSI RAID system) and found that it was alerting me about:
Device: /dev/sda, Failed SMART usage Attribute: 190 Unknown_Attribute.
A lot of digging around found nothing (except other people wondering what it was), until I stumbled upon a page about the Seagate Barracuda 7200.9 ST3300622AS at Cropel S.M.A.R.T. hard drive monitoring which says:
190 Airflow temperature
Aha! Now smartctl -a /dev/sda -d ata tells me:
190 Unknown_Attribute 0x0022 040 040 045 Old_age Always FAILING_NOW 74024484924
Now given that I’m unwilling to believe that the air temperature of my drive is really 74 billion Celsius (three orders of magnitude hotter than the corona of the Sun) there’s got to be something else going on here..
So – does anyone out there have any good ideas ?