Testing recommendations for new hardware
Greetings - I will be purchasing a new server within the next month (PE T610) that will be used as a Samba file server (and some additional functions) for a small office. It will be the primary storage server for the work of 8 staff members, running Centos 6. I have seen over the last few years of following this list people talking about testing hardware prior to it being put into production, but have not paid much attention to what tests are done for what circumstances. So I thought I would ask for recommendations on what kinds of hardware tests I should perform on my new server prior to putting it into production. I am not a system administrator by education, but have been managing our existing Linux server (PE 2600) for almost 7 years, so I will take any recommendations offered and do some additional research to understand what I need to do. Thanks. Jeff www.meridianenv.com ___ Linux-PowerEdge mailing list Linux-PowerEdge@dell.com https://lists.us.dell.com/mailman/listinfo/linux-poweredge Please read the FAQ at http://lists.us.dell.com/faq
testing my servers performance
I want to know how I can test my servers full performance on the given hardware.As to how many maximum connections it can handle what will be latency etc etc. The servers can be web,ftp,streaming,ssh. -- Not sent from my iPhone or my Blackberry or anyone else's ___ Linux-PowerEdge mailing list Linux-PowerEdge@dell.com https://lists.us.dell.com/mailman/listinfo/linux-poweredge Please read the FAQ at http://lists.us.dell.com/faq
testing
making sure i'm getting through. ___ Linux-PowerEdge mailing list Linux-PowerEdge@dell.com https://lists.us.dell.com/mailman/listinfo/linux-poweredge Please read the FAQ at http://lists.us.dell.com/faq
Re: PERC6/i vs. SAS6/iR performance testing?
I could be wrong, but my understanding is that the perc cards have BB write caches and that the SAS cards do not, which would certainly have a very significant effect in iops. --M On Jan 22, 2010, at 4:38 PM, Kaerka Phillips wrote: > Hello, > > I was wondering if anyone on the linux-poweredge list has done any > performance testing of same or similar hardware using the PERC6 vs. the SAS6 > controller cards? I have 7 R805's, two of which have PERC6/i cards > installed, and the other 5 have SAS6/iR cards installed. I've been working > on diagnosing a performance problem on two of those with the SAS cards, and > have now discovered that every R805 with the SAS cards has similar levels of > poor local disk i/o performance. The R805's with PERC6 cards are similar in > performance level to our R905's with PERC6's, and a little better than our > 2950's with PERC5's. > > I can provide my stats, but since they're somewhat extensive, the short > answer is that the R805's which have PERC6 cards perform in the 16k to 30k > i/o per second, depending on write or read type with iozone; but the ones > with the SAS cards can barely do 1000 i/o's per second. These R805's are > similar, but not identical though, here is the mix below. > > All are RHEL5.4, 64bit, and the bottom 3 run kernel-xen, the top two are the > standard RHEL5 64bit kernel > 2x R805's with 2x cpu, 8gb ram, SAS6/iR card, 2x sas 3gbps disks, raid-1 - > Avg. 284 i/o per second > 3x R805's with 2x cpu, 64gb ram, SAS6/iR card, 2x sas 3gbps disks, raid-1 - > Avg. 687 i/o per second > 2x R805's with 2x cpu, 64gb ram, PERC6/i card, 2x sas 3gbps disks, raid-1 - > Avg. 26k i/o per second > > I'm hoping that someone else out there has done some other similar > performance testing, and can either tell me what I'm doing wrong or also show > similar results? My test methods have been using iozone, dd, and rsync. > (iozone command line: iozone -s 1m -r 16 -i 0 -i 2 -I -O -t 75) > > Thanks, > > KBP > ___ > Linux-PowerEdge mailing list > Linux-PowerEdge@dell.com > https://lists.us.dell.com/mailman/listinfo/linux-poweredge > Please read the FAQ at http://lists.us.dell.com/faq ___ Linux-PowerEdge mailing list Linux-PowerEdge@dell.com https://lists.us.dell.com/mailman/listinfo/linux-poweredge Please read the FAQ at http://lists.us.dell.com/faq
PERC6/i vs. SAS6/iR performance testing?
Hello, I was wondering if anyone on the linux-poweredge list has done any performance testing of same or similar hardware using the PERC6 vs. the SAS6 controller cards? I have 7 R805's, two of which have PERC6/i cards installed, and the other 5 have SAS6/iR cards installed. I've been working on diagnosing a performance problem on two of those with the SAS cards, and have now discovered that every R805 with the SAS cards has similar levels of poor local disk i/o performance. The R805's with PERC6 cards are similar in performance level to our R905's with PERC6's, and a little better than our 2950's with PERC5's. I can provide my stats, but since they're somewhat extensive, the short answer is that the R805's which have PERC6 cards perform in the 16k to 30k i/o per second, depending on write or read type with iozone; but the ones with the SAS cards can barely do 1000 i/o's per second. These R805's are similar, but not identical though, here is the mix below. All are RHEL5.4, 64bit, and the bottom 3 run kernel-xen, the top two are the standard RHEL5 64bit kernel 2x R805's with 2x cpu, 8gb ram, SAS6/iR card, 2x sas 3gbps disks, raid-1 - Avg. 284 i/o per second 3x R805's with 2x cpu, 64gb ram, SAS6/iR card, 2x sas 3gbps disks, raid-1 - Avg. 687 i/o per second 2x R805's with 2x cpu, 64gb ram, PERC6/i card, 2x sas 3gbps disks, raid-1 - Avg. 26k i/o per second I'm hoping that someone else out there has done some other similar performance testing, and can either tell me what I'm doing wrong or also show similar results? My test methods have been using iozone, dd, and rsync. (iozone command line: iozone -s 1m -r 16 -i 0 -i 2 -I -O -t 75) Thanks, KBP ___ Linux-PowerEdge mailing list Linux-PowerEdge@dell.com https://lists.us.dell.com/mailman/listinfo/linux-poweredge Please read the FAQ at http://lists.us.dell.com/faq