I'm seeing in NASPT benchmark mysterious performance figures for File Copy to NAS - all the other figures are where you would expect them to be, but this figure is anomalously way low. Stripe size isn't enough to explain the figure, and (a bit of a preview) it doesn't get better under SAN performance. Any ideas?
Also anyone have hands on experience tuning Openfiler for 3Ware raid controllers? Lessons learned?
From the diagram, it appears this storage sub system will be connected into a server via Fiber? can you please clarify? also is connecting the fiber into your network provided your switches support fiber an option? or only connected via ethernet?
From the diagram, it appears this storage sub system will be connected into a server via Fiber? can you please clarify? also is connecting the fiber into your network provided your switches support fiber an option? or only connected via ethernet?
Part two of the series will answer all of these questions.
But the plan is to connect the SAN server via fiber to a DAS server ( my mainline Windows 7 box ), the Windows 7 DAS server sees the SAN as a disk, albeit a very large one - a disk that can be shared like any. That is how the storage is shared over the network.
The scenario that you mention, a switch that handles both your tcp/ip traffic and fiber traffic doesn't really work, they are two different protocols, tcp/ip and fcp ( fibre channel protocol ). Sort like expecting Air Traffic Control to handle traffic jams on Highway 95.
Switches exist for fiber but we aren't using one (outside of the 1K budget), there might be switches that handle iscsi to fcp translation, I don't know. But again, expensive. You can run ethernet on a fiber cable, but that isn't fibre channel.
Given the performance results are below that of a small Sandforce SSD, I'd be very interested to see if something like Flashcache (http://planet.admon.org/flashcache-c...-in-ssd-disks/) would boost your performance results (especially small file size) for an additional $100?
Also, I'm not sure how the on-board Intel ICH-10R performance compares with the RAID card you used, but I have read a few articles where the ICH-10R paired with a decent CPU has superior performance to dedicated RAID cards: http://www.tomshardware.com/reviews/...a,2374-11.html
However, I suppose the limited number of SATA ports (6) would restrict the maximum capacity of your array to about a third of the 35TB that you are shooting for.
Given the performance results are below that of a small Sandforce SSD, I'd be very interested to see if something like Flashcache (http://planet.admon.org/flashcache-c...-in-ssd-disks/) would boost your performance results (especially small file size) for an additional $100?
Also, I'm not sure how the on-board Intel ICH-10R performance compares with the RAID card you used, but I have read a few articles where the ICH-10R paired with a decent CPU has superior performance to dedicated RAID cards: http://www.tomshardware.com/reviews/...a,2374-11.html
However, I suppose the limited number of SATA ports (6) would restrict the maximum capacity of your array to about a third of the 35TB that you are shooting for.
Supposedly ICH10R will handle port multipliers, addonics has a 5x1 ($50), but I'm still limited to 3Gbps per port right? Won't performance fall dramatically if I put 20 drives on 4 PMs?
The Tom article you shared ( Thanks ) does recommend in the conclusion using a RAID card for a dedicated array, better consistent performance supposedly
I'm running SSD as the system drive, my swap is on SSD...it was a part reused from a previous build, A-Data 32GB. I'll take a look at flash cache, a kernel module? Wonder if it'll work on Openfiler.. Appreciate the pointer
Processor + X58/ICH10R Motherboard would also bust my $150 Budget for the MB/Proc.
If money didn't matter, give me one of those sweet new gen Areca 24Port SAS/SATA 4gig cache RAID cards, man I'd like to see that fly. Areca, if you are reading this, I'd be happy to review one of those puppies, provided I could keep it...
Been tuning performance for part 3 of the article, looking at the DAS server ( which gateways my SAN Node ), the SAN jets, but I lose alot when shared - I'm wondering if there are any ways to squeeze more out of SMB2 under Windows 7.
Supposedly ICH10R will handle port multipliers, addonics has a 5x1 ($50), but I'm still limited
to 3Gbps per port right? Won't performance fall dramatically if I put 20 drives on 4 PMs?
If you want to go cheaper, an ICH-9R may have similar performance. Regarding the port multiplier question - if you are able to find a compatible PM then even with 5 drives connected to each PM, the theoretical max bandwidth per drive with 5 drives connected is still 75MB/s. If you add a separate SATA controller dedicated as the system drive then perhaps you could spread the load across all 6 ports on the ICH-10R.
Please do post back and let us know your experiences with flashcache.
I'm seeing in NASPT benchmark mysterious performance figures for File Copy to NAS - all the other figures are where you would expect them to be, but this figure is anomalously way low. Stripe size isn't enough to explain the figure, and (a bit of a preview) it doesn't get better under SAN performance. Any ideas?
Also anyone have hands on experience tuning Openfiler for 3Ware raid controllers? Lessons learned?
hello Greg
great informative article there.
just a few questions there. i see that the results of benchmark in the nas file copy results.
do you think it may be a network interface bandwidth issue.
you said your mother board has two nic ports have you tried link aggregation ?can that be done in openfiler?will that improve the read and write throughput.
I'm seeing in NASPT benchmark mysterious performance figures for File Copy to NAS - all the other figures are where you would expect them to be, but this figure is anomalously way low. Stripe size isn't enough to explain the figure, and (a bit of a preview) it doesn't get better under SAN performance. Any ideas?
Also anyone have hands on experience tuning Openfiler for 3Ware raid controllers? Lessons learned?
great article there Greg very informative.
one thing which i don't understand though is why is there so much of diffrence in hd playback results and file copy from nas results?
one more thing is you have stated the the motherboard has two nics.
have you tried link aggregation.does openfiler support this?
will that effectively increase read throughput?
one thing which i don't understand though is why is there so much of diffrence in hd playback results and file copy from nas results?
one more thing is you have stated the the motherboard has two nics.
have you tried link aggregation.does openfiler support this?
will that effectively increase read throughput?
HD playback is the continuous "playback" of a single large file, where file copy is based on a large number of small reads across multiple files, the most demanding form of I/O - which is impacted by tcp/ip tuning, stripe size and disk speeds ( caching doesn't help, either on the RAID card, or in the machine because the reads are too small ). HD playback benefits from read ahead, caching and the large stripe size.
Openfiler supports link aggregation and it works fine, but does not impact Old Shuck's throughput largely because all file access/read/write all go through the fibre channel interface as a SAN. When I ran it as a NAS before converting it to SAN, I did not test aggregated throughput, but I see no reason for there not to be improved performance in a multi-node environment.
If you look at the gallery for Ol'Shuck, you will see the final NASPT performance chart.
Build Your Own Fibre Channel SAN
Hybrid Mode
