メインコンテンツへスキップ

⚠️ Please note that this topic or post has been archived. The information contained here may no longer be accurate or up-to-date. ⚠️

File processing using clusters.

コメント

5件のコメント

  • NNN634243134066291815
    It seems like there is an easy solution for your friend that will work right now, although I would like this feature as well.

    He already has at least two towers (otherwise a cluster wouldn't be a cluster) and two copies of all the files (original and backup). Why not process half the job on one computer and half the job on the other? That cuts the processing time down to 7 hours. Than backup only those new files which depending on file size might take 2 hours for a total of 9 hours, shaving 5 hours off the total.

    Otherwise you could look into gigabit or better NAS systems which would also be an easy and cost effective solution. The bottleneck would seem to be at the end you probably (I'm assuming) have to copy all these files to your clients computer or drive.
    0
  • Ulf Liljegren
    What we are working is Capture One 4.0, no secret there.
    One of the major difference in 4.0 is the processing speed. We have increased the processing speed with about 300% with maintained quality. I think this will have more improvement and practical appliance then making the application cluster aware. (But it would be cool! 😊
    0
  • JamesNYC
    Yes faster processing sped would be nice and as I was shown last year at Photo Expo in NYC the faster processing already exists in the BETA version of V.4.0
    Now all we need is to have it release to the masses.
    Which should not be in the too distant future since Phase One reps have been doing Demos of V.4.0 here in NYC for the past 3+ months.

    The fact is that having the ability to create a cluster for processing would be a huge help for a number of digital techs and others that are dealing with file sizes and numbers of files that most users never have to deal with.

    Consider 300 gigs a day that need to be processed over night for a bitchy client.

    Lets get V.4.0 out and add clustering for V.4.0.2
    Hey even an Open Source option would be helpful, but please pick up the pace on the software release end of things.

    JamesNYC
    0
  • Mark711
    Have you looked into X Grid in Mac OS X Server?


    Distributed computing
    Xgrid makes it easy to turn an ad hoc group of Macs into a low-cost supercomputer by streamlining the process of assembling nodes, submitting jobs, and retrieving results. Scientists, animators, and digital content creators now have the opportunity to easily run a single job across multiple computers at once, dramatically improving performance and responsiveness.



    The description gives me the impression that any app can spread the processing across multiple machines, Mac Only I Presume, since it is X Grid that handles the distribution of the processing not the app.

    I run OS X Server but haven't messed with X Grid.

    Good Luck
    0
  • rap_digital
    X-grid really isn't the solution in this case. With only 2 agents available it would be faster and easy to split the job in half and run two batches. You may find that having the files on a network raid drive will save time moving them around. Also something like can help manage keyboards and mice.
    0

投稿コメントは受け付けていません。