Page 9 of 10 FirstFirst ... 78910 LastLast
Results 81 to 90 of 94

Thread: What Would You Like To See Next?

  1. #81

    Default

    I'd like to notice compositing screen managers benchmarked anytime games tend to be benchmarked far too. I do definitely not enjoy games often yet I personally use the particular WM on a regular basis, plus the video individuals do influence the item a good deal, therefore a new WM benchmark could be additional strongly related me.

    I know at least kwin carries a show_fps plugin as well as scripting assistance, then it mustn't be hard to help create a new benchmark for it. Not sure wether compiz yet others have got scripting assistance, or the method that you would produce a benchmark devoid of of which.

  2. #82
    Join Date
    Jan 2010
    Location
    Ghent
    Posts
    217

    Default Fuse-vs-Rump-vs-Native file system performances

    I would like to see some file system benchmarks on fuse-vs-native (there were some on ZFS, but also others could be interesting).

    A really interesting new technology is Rump, which provides NetBSD drivers in userspace on many different OSes, including Linux. fs-utils provides utilities to handle the file systems the Rump userland kernel provides.

    https://github.com/anttikantee/buildrump.sh
    https://github.com/stacktic/fs-utils

    Binary packages of Rump and fs-utils for a variety of Linux distros can be found at:
    http://download.opensuse.org/reposit...me:/staal1978/

    Disclaimer: I did the OBS packaging so I might be biased.

  3. #83

    Default

    Different JS benchmarks for Firefox 64/Firefox 32/Firefox 32 in 64bit Linux. I'm interested in Firefox from ftp.mozilla.org

  4. #84
    Join Date
    Aug 2013
    Posts
    3

    Default 32 bit vs 64 : HDD / SSD read-write speed

    In the 32 bit vs 64 bit tests that get posted on Phoronix (example) please also include:

    32 bit vs 64 bit operating systems, both installed on 64 bit hardware, which one is faster for reading / writing internal hard disk drive / solid state drive?

    -as mostly that's the bottleneck in general computing.

  5. #85
    Join Date
    Sep 2012
    Posts
    23

    Default Filesystem benchmark request

    I've bought an SSD and looking for the right setup to use that has RAID and LVM capabilities. I'm thinking of BTRFS or MDRAID5+LVM+EXT4|XFS so what I'd like to see as a benchmark is something that is equal to BTRFS. Specifically:

    BTRFS w/ RAID5
    MDRAID 5 + LVM + EXT4
    MDRAID 5 + LVM + XFS

    I ask for this b/c I know that LVM can have up to a 30% impact to performance, so testing BRTFS directly against EXT4/XFS, etc. is not really a fair comparison to how things will perform in the real world once you start adding the capabilities that BTRFS brings using LVM/MDRAID. It would be nice to have a benchmark that tests the difference scenarios to get a system with these types of capabilities.

  6. #86

    Default

    Quote Originally Posted by sheldonl View Post
    I've bought an SSD and looking for the right setup to use that has RAID and LVM capabilities. I'm thinking of BTRFS or MDRAID5+LVM+EXT4|XFS so what I'd like to see as a benchmark is something that is equal to BTRFS. Specifically:

    BTRFS w/ RAID5
    MDRAID 5 + LVM + EXT4
    MDRAID 5 + LVM + XFS

    I ask for this b/c I know that LVM can have up to a 30% impact to performance, so testing BRTFS directly against EXT4/XFS, etc. is not really a fair comparison to how things will perform in the real world once you start adding the capabilities that BTRFS brings using LVM/MDRAID. It would be nice to have a benchmark that tests the difference scenarios to get a system with these types of capabilities.
    Unfortunately it's not too often I get review samples of HDD/SSDs, let alone in 3+ identical disk configurations, so such RAID5 configurations are unlikely to be tested.

  7. #87
    Join Date
    Sep 2012
    Posts
    23

    Default

    Quote Originally Posted by Michael View Post
    Unfortunately it's not too often I get review samples of HDD/SSDs, let alone in 3+ identical disk configurations, so such RAID5 configurations are unlikely to be tested.
    As long as you have 3 HDD of the same RPM (need not be SSD's) you should be ok. Just create the same sized partition on each of them if they aren't the same size. It's not terribly scientific but it's good enough to give you an idea.

  8. #88
    Join Date
    Sep 2012
    Posts
    23

    Default

    Quote Originally Posted by Michael View Post
    Unfortunately it's not too often I get review samples of HDD/SSDs, let alone in 3+ identical disk configurations, so such RAID5 configurations are unlikely to be tested.
    I should mention that I'm not going to do RAID5 on one SSD of course, but it gives me the opportunity/excuse to reinstall my whole system which has 6 other drives.

  9. #89
    Join Date
    Sep 2012
    Posts
    23

    Default

    Quote Originally Posted by Michael View Post
    Unfortunately it's not too often I get review samples of HDD/SSDs, let alone in 3+ identical disk configurations, so such RAID5 configurations are unlikely to be tested.
    BTW Michael, you did do a ZFS vs EXT4+LVM on RAID benchmark back in April, do you not still have those disks?

  10. #90
    Join Date
    Dec 2013
    Posts
    17

    Default Regarding Openbenchmarking.org

    It would be nice if in the Phoronix test profiles page, next to each test a small column was displayed showing the platforms on which the test is supported.
    It would be better if we could filter by platform, with support for multiple filters (say, to be able to select all tests that are supported on both Windows and Linux).

    Here is a mockup showing what i am talking about :


Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •