Re: [lustre-discuss] Unexpected result with overstriping
John, I believe the lfs-setstripe man page is incorrect (or at least misleading) in this case. I recall seeing 2000 hardcoded as a maximum, so it appears to be picking that. Using "-C -1" to put a single stripe on each OST wouldn't have any benefit over "-c -1". IMHO, it would probably be more useful to have negative values represent number of stripes per OST. -Nathan From: lustre-discuss on behalf of John Bauer Sent: Friday, May 17, 2024 8:48 AM To: lustre-discuss Subject: [lustre-discuss] Unexpected result with overstriping External email: Use caution opening links or attachments Good morning all, I am playing around with overstriping a bit and I found a behavior that, to me, would seem unexpected. The documentation for -C -1 indicates that the file should be striped over all available OSTs. The pool, which happens to be the default, is ssd-pool which has 32 OSTs. I got a stripeCount of 2000. Is this as expected? pfe20.jbauer2 213> rm -f /nobackup/jbauer2/ddd.dat pfe20.jbauer2 214> lfs setstripe -C -1 /nobackup/jbauer2/ddd.dat pfe20.jbauer2 215> lfs getstripe /nobackup/jbauer2/ddd.dat /nobackup/jbauer2/ddd.dat lmm_stripe_count: 2000 lmm_stripe_size: 1048576 lmm_pattern: raid0,overstriped lmm_layout_gen:0 lmm_stripe_offset: 119 lmm_pool: ssd-pool obdidx objid objid group 119 523862870x31f59ef 0 123 523479470x31ec42b 0 127 527344870x324aa17 0 121 528393960x32643e4 0 131 527427090x324ca35 0 116 522426590x31d28e3 0 117 518311250x316e155 0 124 524252180x31ff202 0 125 524027220x31f9a22 0 106 527005810x32425a5 0 edited for brevity ___ lustre-discuss mailing list lustre-discuss@lists.lustre.org http://lists.lustre.org/listinfo.cgi/lustre-discuss-lustre.org
[lustre-discuss] FLR Mirroring for read performance
Greetings! During the helpful LUG tutorial from Rick Mohr on advanced lustre file layouts, it was mentioned that "lfs mirror" could be used to improve read performance. And the manual supports this, stating "files that are concurrently read by many clients (e.g. input decks, shared libraries, or executables) the aggregate parallel read performance of a single file can be improved by creating multiple mirrors of the file data". What method does Lustre use to ensure that multiple clients balance their read workloads from the multiple mirrors? Are there any tuning parameters that should be considered, other than making sure the "preferred" flag is NOT set on a single mirror, to help even out the read workload among the OSTs? Has anyone tested this and quantified the performance improvement? Thanks, Nathan ___ lustre-discuss mailing list lustre-discuss@lists.lustre.org http://lists.lustre.org/listinfo.cgi/lustre-discuss-lustre.org
Re: [lustre-discuss] 2.14 against mofed 5.5.1.0.3.2-rhel7.9
Michael, Likely your issue will be resolved in 2.15. You could try applying the patch from here: https://jira.whamcloud.com/browse/LU-15417 Build lustre on MOFED 5.5 -Nathan -Original Message- From: lustre-discuss On Behalf Of Michael DiDomenico via lustre-discuss Sent: Monday, March 7, 2022 10:16 AM To: lustre-discuss Subject: Re: [lustre-discuss] 2.14 against mofed 5.5.1.0.3.2-rhel7.9 External email: Use caution opening links or attachments and adding --with-o2ib=/usr/src/ofa_kernel/x86_64/2.10.0-1160.59.1.el7.x86_64 seems to have at least allowed configure to move forward and the subsequent make passed. that's as far as i can get today, i'll try mounts later On Mon, Mar 7, 2022 at 12:07 PM Michael DiDomenico wrote: > > digging into this a little further, it looks like > > from lustre-release/lnet/autoconf/lustre-lnet.m4 > > O2IBPATHS=$(eval $OFED_INFO | egrep -w > 'mlnx-ofed-kernel-dkms|mlnx-ofa_kernel-devel|compat-rdma-devel|kernel-ib-devel|ofa_kernel-devel' > | xargs $LSPKG | grep '\(/openib\|/ofa_kernel/default\|/ofa_kernel\)$' > | head -n1) > > this is the line that's failing for me. the first egrep returns > mlnx-ofa_kernel-devel, which is piped to rpm -ql, but the second grep > does not return anything > > On Mon, Mar 7, 2022 at 11:07 AM Michael DiDomenico > wrote: > > > > has anyone compiled 2.14 against the latest mofed 5.5.1.0.3.2? i > > just tried, but it kicked out with unable to find the ofed sources. > > previous build was 2.14 against mofed 5.4-1.0.3.0 which worked fine ___ lustre-discuss mailing list lustre-discuss@lists.lustre.org https://nam11.safelinks.protection.outlook.com/?url=http%3A%2F%2Flists.lustre.org%2Flistinfo.cgi%2Flustre-discuss-lustre.orgdata=04%7C01%7Cndauchy%40nvidia.com%7Cd508bfc677c74ba7145a08da005e3208%7C43083d15727340c1b7db39efd9ccc17a%7C0%7C0%7C637822701811996769%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C3000sdata=kWoGK6RgNn6LAPFY4E5bw1dLLapqeb5fvb2bpX2lqDI%3Dreserved=0 ___ lustre-discuss mailing list lustre-discuss@lists.lustre.org http://lists.lustre.org/listinfo.cgi/lustre-discuss-lustre.org