lockup

Member
  • Content Count

    24
  • Joined

  • Last visited

Community Reputation

0 Neutral

About lockup

  • Rank
    Member
  1. lockup

    Storage Review Site Update

    This is great news. I've been checking SR for updates for years, hoping that it would revive itself. Now I'm a (relatively) old fart, I'm now more interested RAID configurations, IOPS and Watts than MB/s! SR Should be the place for this as others have said above. Anandtech have done well with SSD tech recently, would be impressed if SR could match/beat that.
  2. lockup

    RAID 6 vs RAID 10

    Sorry, came to this one a bit late. However I can confirm that you've made the right choice. I run a number of VMWare hosts and RAID5/6 is a crippling bottleneck. You need plenty of write i/o's, disk redundancy and space - ie. RAID10.
  3. lockup

    Maxtor SATA drives unreliable?

    Purchasing a multi-national company seems a bit excessive... From personal experience I've only got good things to say about Western Digital drives and little good to say about Maxtor. However, my personal experience is statistically worthless. A good place to go and get answers is the SR reliability database.
  4. lockup

    problems with dual Opteron system

    alpha, you didn't come up with a good reason not to try the above so I assume you are doing it! Any luck?
  5. lockup

    problems with dual Opteron system

    Also, you may have some joy running the memory at 333 (PC2700) instead of 400 (PC3200), or in single channel mode, or both. I'm naturally suspicious of the memory in this instance! That's not to say the power supply won't become an issue later on. Bitter experience has lead me to spec high quality PSUs on all my recent builds.
  6. lockup

    problems with dual Opteron system

    These boards have been known to be picky with the memory. If possible, retest with memory from the following list: http://www.tyan.com/support/html/memory_s2875.html. Also, a BIOS update and reset might help you out: http://www.tyan.com/support/html/b_s2875.html Good luck!
  7. lockup

    What's the fastest processor available?

    I think I've now got a better understanding of your predicament. As you have no influence over the code, perhaps the best you can do is create some batch scripts that will do multiple runs in parallel and then opt for multicore Opteron/Woodcrest systems. As mentioned above, the only way to establish the fastest platform will be to test on each. Maybe a bit of begging/borrowing/stealing of hardware is required. Has your boss suggested a solution to your problem?
  8. lockup

    What's the fastest processor available?

    Well written code can mean different things to different people. It could be easy to understand at first read, or maintainable and easily extensible (or all of these things). It could occupy as few KB as possible. While all these attributes may be good, they ain't neccessarily fast. I design my code to perform well in the situations I think it will have the most work to do. I try to keep it strong in the first three attributes when I write it. Then, if performance is an issue I profile it to see where the hold-up is. In your case whatever you program in, its eventually going to be translated in to x86 machine code. FORTAN is predominantly used in scientific programming and hence I would expect its maths routines to be good and quick. However, there's no reason a good C++ library couldn't be just as quick, or quicker. Maybe that's because the people who want CFD solver routines are mostly using FORTRAN. C++ solvers do exist however, see here. Parallelisation would not compromise the drive to improve program efficiency on a single thread. It may have been easier to implement it earlier, but what the hey. To paraphrase you, if a code itself does not run inherently, throwing money at a new processor isn't going to help (much). If your code is rubbish, making it multithreaded could make it n times less rubbish on an n core system which might be the most efficient way to spend your time and money. You will need to work in a FORTRAN environment that has an integrated profiler application. Once you've run your code through it, it will show you the relative time spent in each function. The way it does this is implementation specific, but modern profilers will have a graphical interface and allow you to view the timing information in a number of different ways. Maybe you can get some eval versions - I have no recent experience of FORTRAN development environments. As an example I wrote a business analysis system for Vodafone in Java. Although it was accurate, it wasn't fast enough under the anticipated user load. By using a profiler my efforts were focussed on tuning the VM, optimising how and when I created Java objects and spending $200 on a library to improve the perfomance of one specific maths function. System was ten times quicker, everybody happy.
  9. lockup

    What's the fastest processor available?

    The choice of programming language, once the code is optimised, isn't going to make a vast difference on a single thread. As it happens I've programmed in FORTRAN and the others you mention. I could take issue with your assertion that FORTAN is a lower level language than C but really, it's not important. As far as I can make out the problem you wish to solve is, I have a maths-centric program that I want to run quicker This program has multiple elements Further elements are to be added of an as yet undetermined form Your solution is to get the fastest processor available, hence the title of this topic. That is indeed one solution. However, how much faster on a single thread can you get from where you are, 25%, 50% faster? And a year from now, how much faster will processors be on a single thread? You will be getting incremental changes when what you want is a step change in performance. The above is also assuming that the code is CPU bound, not IO bound. It would be pretty gutting to spend $1000 on the best CPU available, only to find that it makes 0% difference and you could have tripled the speed by getting a $500 hard drive. My suggestion is to profile existing code and see where the bottlenecks are. Then work towards removing those bottlenecks with a combination of tuned code, multi-threading and upgraded hardware. I reckon we'd all be keen to hear how you solve this one!
  10. lockup

    What's the fastest processor available?

    I think you need to run the program through a profiler and see where it is spending its time. If you're only using 25% CPU at full chat on a dual Opteron system, there's scope to make it twice as quick before you even consider making it multithreaded. Trust me on this one, profilers rock!
  11. lockup

    What's the fastest processor available?

    I have thought about that, but right now the way that the program is written (and running) is that it doesn't generate different output files for different input files. So, the only way that I would be able to do that would be to have it separated by directory and to run it all within itself. And to be quite honest, I haven't tried, but I worry that I might not get much benefit out of having multiple instances because of the IO transfers. (The result files are in excess of 8 GB). So if you double the output speed you'll be IO limited? Now this is beginning to look like a storagereview question! If it is possible to separate off work in to separate processes working in separate directories, that may well be a good path to follow in the short term. With little extra code that could be handled by a tactical batch file. Then you'll be looking at a multi-core multi-socket system with plenty of SCSI attached, job done. EDIT: Also, it shouldn't be too hard to add a couple of command line parameters that specify the input and output file paths.
  12. lockup

    What's the fastest processor available?

    If it's taking 90 minutes to do a single run, whatever it is, it sounds like it should be parallelized as soon as possible, not a year down the line! Single core performance is not going to grow at any where near the rate you need to get that down to a manageable time.
  13. I run nightly differential backups of one of my online volumes (300GB) over ADSL (2Mbit). That copes with about 10GB of changes per night outside of working hours. This is on a Windows platform, so I schedule xxcopy a really handy command line utility for file copies. You don't get the file history of tape, but you do get the confidence that the files have been backed up and are immediately accessible. Very handy for my particular scenario.
  14. I'm currently having issues with 8 Maxtor Maxline III 300GB SATA drives dropping off a 3Ware 9550SX-8LP controller. Both Maxtor and 3Ware support seem to be glacially slow at the moment. If anybody out there has run these drives (s/n 7L300S0) on a 9550 controller successfully or otherwise, I'd love to know. Cheers!
  15. lockup

    RAID5 vs RAID50 on 3Ware

    As it happens I will have a separate boot drive with 2x300GB in RAID1. If I have the time to initialise the array twice I'll run up some comparison IOMeter numbers and post them back here. Fingers crossed...