LSI SAS/SATA HBA's and the Fuel (or other IP35 Systems)

SGI hardware problems, solutions, tips, hacks, etc.
Forum rules
Any posts concerning pirated software or offering to buy/sell/trade commercial software are subject to removal.
User avatar
mapesdhs
Posts: 2496
Joined: Mon Nov 10, 2003 4:17 pm
Location: Edinburgh, Scotland
Contact:

Re: LSI SAS/SATA HBA's and the Fuel (or other IP35 Systems)

Unread postby mapesdhs » Wed Jun 03, 2009 9:07 am

Just some general info for you all...

T'was my bday on 19th May, so I asked for some dosh from parents/brother, bought myself a 1TB SATA for
my PC (movie archiving), though I'll test it when I can in a Fuel with an LSI card, see how it fares.

Anyway, my PC has a 146GB Maxtor Atlas 15K II as a system disk, same model I use in almost all my
SGIs and the main drive I recommend for those who want the best in an SGI (plenty available if anyone's
interested! ;D). I thought a performance comparison with the SATA might be interesting, the classic
space vs. speed issue so often discussed. How well does the old SCSI drive compare? I like the Maxtor
because of its fast access time (ideal as a system disk, ie. searching files, etc.), but despite its
age (released November 2004) it holds up rather well against the SATA, except for burst read where
presumably the faster/larger cache RAM in the modern SATA has an advantage. The following is from
testing with HDTach, standard read test, in order of max sustained speed:

Code: Select all

                                               Max        Min      Average     Burst    Random Access
                                            (MB/sec)   (MB/sec)   (MB/sec)   (MB/sec)        (ms)

Seagate 146GB 15K ST3146855LC SCA:             135         81      113.9       196.4         5.7
Fujitsu 300GB 15K MBA3300NC                    131         69      107.3       223,2         7.1
Samsung SpinPoint F1 1TB SATA/300 HD103UJ:     115         55       89.8       235.9        16.5
Maxtor Atlas 15K II 146GB 8K147J0 SCA:         105         63       89.3       177.7         5.7
Seagate 18GB 15K ST318452LC SCA:                61         45       55.0       128.8         5.9
Fujitsu 18GB 15K MAM3184MC SCA:                 57         44       53.4       144.2         6.0
Seagate 9GB 15K BF00963643 SCA:                 42         36       38.6       128.1         6.3


Note the average sustained speed is almost identical between the Samsung and the Maxtor (shows how long it's taken
for IDE/SATA to catch up), but check out the average access times! Also rather interesting, I first tested the Samsung
using the PC mbd's onboard Marvell RAID controller, which gave a substantially lower Burst rate (155MB/sec) and a
slightly lower average rate that matched the Maxtor (89.3). Looks like the choice of SATA controller in a PC is important
(the mbd's NVIDIA controller is much better).

For a real world test, I copied a 6.7GB VOB file from the Maxtor to the Samsung, which completed in 80 seconds, avg
speed 84MB/sec.

The Samsung isn't the fastest SATA around, but it's pretty high up the scale and is a decent tradeoff between speed and
price (in UKP, 65 + 6 shipping from CCL Online).


Are there any gotchas I should watch out for though when testing the Samsung with the Fuel?

Ian.
Last edited by mapesdhs on Wed Jun 03, 2009 3:46 pm, edited 4 times in total.

User avatar
Dr. Dave
Posts: 2311
Joined: Fri Feb 13, 2004 10:37 pm
Location: Ottawa, Canada >burp<

Re: LSI SAS/SATA HBA's and the Fuel (or other IP35 Systems)

Unread postby Dr. Dave » Wed Jun 03, 2009 10:02 am

Regarding newer drives, high platter densities mean that the areal transfer rates have gone way up on consumer drives lately - but as your test has noted the access times (and latencies) would still be better on the 15k drives. The 10k SATA drives are likely in the middle.

As for testing gotchas, I'd be concerned that for large contiguous files the SATA drive will post good numbers, but for small file accesses with lots of random seeks, the 15k SCSI drive will likely still win out by a fair margin. As a note, I always try to use a 15k drive as the root drive on any of the later-model SGI's I have - and for the Fuel this combo with a large consumer SATA/IDE drive for bulk storage (ie: homes) seems to be the best of both worlds. The root drive doesn't have to be more than 18GB or so to give you ample room for just about anything you'd want to do if you set it up this way, and 18GB 15k drives are pretty cheap.

So keeping this in mind, probably the two tests I'd try is a huge contiguous file transfer, and maybe a 'find' function across the disk to replicate random seek timing effects, using a SATA drive which has a duplicate filesystem to ensure the tests are equivalent. 'Diskperf' is actually a pretty good tool for easily showing this, though since it uses only the portion of the disk that the temp file is on, the numbers only represent a portion of the transfer curve for the platter.
:O3000: <> :O3000: :O2000: :Tezro: :Fuel: x2+ :Octane2: :Octane: x3 :1600SW: x2 :O2: x2+ :Indigo2IMP: :Indigo2: x2 :Indigo: x3 :Indy: x2+

Once you step up to the big iron, you learn all about physics, electrical standards, and first aid - usually all in the same day

User avatar
mapesdhs
Posts: 2496
Joined: Mon Nov 10, 2003 4:17 pm
Location: Edinburgh, Scotland
Contact:

Re: LSI SAS/SATA HBA's and the Fuel (or other IP35 Systems)

Unread postby mapesdhs » Wed Jun 03, 2009 10:46 am

Dr. Dave writes:
> Regarding newer drives, high platter densities mean that the areal transfer rates have gone way up on consumer drives lately
> - but as your test has noted the access times (and latencies) would still be better on the 15k drives. The 10k SATA drives are
> likely in the middle.

Yup, that makes sense.

Btw, an update on the 'best' 15K drive for SGIs. Ok, so I'll admit I'm surprised. Today I received a Fujitsu 300GB 15K SCA SCSI
(which will now be the new system disk in my Fuel), a drive I bought 'cos the price was good and I needed more space in my
Fuel. First, here's the diskperf for a Maxtor 15K II 146GB:

Code: Select all

# req_size  fwd_wt  fwd_rd  bwd_wt  bwd_rd  rnd_wt  rnd_rd
#  (bytes)  (MB/s)  (MB/s)  (MB/s)  (MB/s)  (MB/s)  (MB/s)
#---------------------------------------------------------
      16384   54.97   63.18   46.53    4.31    8.07    4.65
      32768   75.20   88.89   65.67    9.07   14.60    8.85
      65536   89.38   97.03   75.82   20.21   23.55   15.83
     131072   89.35   97.11   72.84   26.67   34.56   25.64
     262144   89.54   96.84   73.22   45.07   48.67   39.29
     524288   89.03   73.91   78.00   53.26   59.15   53.85
    1048576   89.38   78.83   80.96   68.68   68.99   66.06
    2097152   90.35   77.97   77.15   78.24   75.46   75.88
    4194304   89.95   85.12   83.23   85.36   82.89   84.29


Ok, pretty good, but now check the Fujitsu (the model is MBA3300NC):

Code: Select all

# req_size  fwd_wt  fwd_rd  bwd_wt  bwd_rd  rnd_wt  rnd_rd
#  (bytes)  (MB/s)  (MB/s)  (MB/s)  (MB/s)  (MB/s)  (MB/s)
#---------------------------------------------------------
      16384   59.87   68.97   30.55   27.13   10.23    5.30
      32768   80.08   93.85   34.26   33.87   17.71   10.04
      65536   94.37  114.01   56.67   38.45   29.14   18.21
     131072   92.95  117.03   43.56   43.48   32.49   31.45
     262144  102.52  124.69   44.35   44.50   51.88   48.31
     524288  107.48  124.89   66.54   67.23   72.67   67.22
    1048576  109.85  124.88   89.21   90.17   91.55   86.17
    2097152  112.99  124.93  105.37  102.76  104.32   99.83
    4194304  114.38  124.92  110.83  109.11  111.25  109.44


Good grud! This is the first SCSI disk I've come across that can sustain more than 100MB/sec, but it's way over 100! Wow! 8)

So, I guess I'd now have to say this Fujitsu is the best possible SCSI disk for an SGI, but they're hard as hell to find (certainly
for a decent price anyway - I was damn lucky for sure), whereas I have more than 50 of the Maxtors available...

If anyone comes across a drive that beats the above Fujitsu, please let me know! Hmm, before I fit it into the Fuel, I'll connect
it to my PC and run HDTach, see what it comes up with. Should be interesting...


> As for testing gotchas, I'd be concerned that for large contiguous files the SATA drive will post good numbers,
> but for small file accesses with lots of random seeks, the 15k SCSI drive will likely still win out by a fair margin.

Absolutely true, which is why I'm fortunate in this regard since I'll be using the SATA disk to store large files,
typically 500MB to 2GB, so the slower access time isn't an issue.


> As a note, I always try to use a 15k drive as the root drive on any of the later-model SGI's I have ...

Despite the lower sustained max speed, a 15K helps in Octane aswell, boosting file searches, etc. by more than 30%.
Any O3K machine just gets more out of them given the use of an U160 bus, though of course Octane can aswell if
the disk is connected via better SCSI card.


> ... doesn't have to be more than 18GB or so to give you ample room for just about anything you'd want to do if you
> set it up this way, and 18GB 15k drives are pretty cheap.

(I even have a couple of 9GB 15Ks available! Wierd huh? I didn't know anyone made 9GB 15Ks)

An 18GB 15K is certainly cheap, but it's not very fast. Infact an 18GB 15K is typically slower than a reasonable
36GB or 73GB 10K for sustained read/write. For example, here's a typical 18GB 15K Seagate (all these tests were
done with a Fuel/600):

Code: Select all

# req_size  fwd_wt  fwd_rd  bwd_wt  bwd_rd  rnd_wt  rnd_rd
#  (bytes)  (MB/s)  (MB/s)  (MB/s)  (MB/s)  (MB/s)  (MB/s)
#---------------------------------------------------------
      16384   41.21   41.36    8.93    4.31    6.45    3.45
      32768   41.27   41.45   15.31    9.37   10.94    6.18
      65536   41.29   41.45   22.38   11.31   17.01   10.43
     131072   41.39   41.46   28.82   20.23   24.82   15.70
     262144   41.04   41.44   32.13   25.87   30.34   21.17
     524288   40.50   41.42   33.75   32.43   31.06   30.52
    1048576   39.70   41.44   37.78   33.61   31.94   33.82
    2097152   39.31   41.41   32.26   38.43   32.61   37.38
    4194304   38.75   41.38   35.42   38.66   31.80   38.51


That's pretty horrible. Would the access time really be all that good? I'm not so sure. Either way, compare the above
to a Fujitsu 73GB 10K:

Code: Select all

# req_size  fwd_wt  fwd_rd  bwd_wt  bwd_rd  rnd_wt  rnd_rd
#  (bytes)  (MB/s)  (MB/s)  (MB/s)  (MB/s)  (MB/s)  (MB/s)
#---------------------------------------------------------
      16384   54.35   55.74   22.08   21.95    9.11    3.88
      32768   61.80   70.82   25.12   25.99   15.21    7.34
      65536   70.20   80.40   42.42   29.54   23.92   13.28
     131072   76.87   83.65   29.09   29.08   24.06   21.62
     262144   82.66   88.36   31.38   29.07   37.65   32.90
     524288   86.61   88.42   48.59   43.52   52.78   45.43
    1048576   88.80   88.63   66.71   57.91   66.33   62.12
    2097152   88.78   88.55   80.09   69.35   75.50   70.96
    4194304   88.69   88.78   84.57   76.97   81.84   78.35


Blows the 18GB 15K away completely.

I'm sure the 18GB 15K was nice when drives were typically no more than 18GB max capacity, but at least for sustained read/write
they've long since been surpassed by newer 10K drives.

If I get a chance, I'll try one of the 18GB 15Ks with HDTach aswell, see what it says about average access time.

(EDIT: tested! Susained 18GB 15K speed isn't that good, but the access time, 6ms, is still way better than the SATA, and not that
much slower than modern 15Ks. See the table in my previous post, ammended with extra results)


> So keeping this in mind, probably the two tests I'd try is a huge contiguous file transfer, and maybe a 'find' function across the
> disk to replicate random seek timing effects, using a SATA drive which has a duplicate filesystem to ensure the tests are

Yes, I already have a 'find' test, consists of searching all of the contents of /usr/share (pre-copied onto a target drive before
a reboot prior to the test), 715MB of data, more than 54000 files, ie. searching for a non-existent file forces the search to
scan the entire archive. System is rebooted before each test of course. I've not tried this test with any newer models such as the
Maxtor mentioned earlier, but here's an example comparison of a 10K vs. a 15K in a Fuel/500:

Code: Select all

               Search Time
                (seconds)

    36GB 15K:     11.08
    36GB 10K:     18.32


ie. about 40% faster with a 15K.


> ... though since it uses only the portion of the disk that the temp file is on, the numbers only represent a portion of the
> transfer curve for the platter.

If the disk is blank via a freshly done mkfs, where on the drive will the file be placed when one does a mkfile to create
the test file? On the slowest part of the platters, or the fastest? Do all disks default to writing data in the same place when
they're initially empty?

Either way, I'm slowly building up a collection of diskperfs for various drives. When I have a decent number, I'll add the
info as links from my disks-for-sale page, and keep the entries shown even when I've none in stock (text will be in
italics) so the reference info is still accessible.

EDIT:

Just discovered one of my Seagate 146GB/15Ks is pretty fast too (faster than the Maxtor), ie. the ST3146855LC:

Code: Select all

# req_size  fwd_wt  fwd_rd  bwd_wt  bwd_rd  rnd_wt  rnd_rd
#  (bytes)  (MB/s)  (MB/s)  (MB/s)  (MB/s)  (MB/s)  (MB/s)
#---------------------------------------------------------
      16384   61.86   70.91   35.78   13.71   10.10    4.78
      32768   81.17   93.59   55.53   17.91   19.06    9.02
      65536   95.78  114.09   75.43   29.28   32.90   15.81
     131072  105.75  128.86   92.64   42.90   51.67   25.62
     262144  112.39  134.28  102.58   43.24   68.36   38.02
     524288  115.86  134.05   64.30   48.24   71.51   50.17
    1048576  117.45  134.01   85.31   64.54   92.22   59.12
    2097152  117.99  133.80  101.38   86.01   94.91   84.19
    4194304  118.53  128.88  112.93  108.64  110.52  107.86


Seems to be slightly faster than the Fujitsu.

Dr Dave, I added a few more results to the table in my last post (have a look). It seems you were right; though the
sustained rates with the older 15K drives is much less, the access time remain nice and fast.

Ian.

User avatar
mapesdhs
Posts: 2496
Joined: Mon Nov 10, 2003 4:17 pm
Location: Edinburgh, Scotland
Contact:

Re: LSI SAS/SATA HBA's and the Fuel (or other IP35 Systems)

Unread postby mapesdhs » Mon Jun 08, 2009 8:40 am

Quick update...

I've tested a 1TB SATA drive on a Fuel with an LSI SAS3442X-R SAS/SATA card and it works fine!

Check the hinv, and here's the diskperf/df output (the SATA drive is the one mounted on /0):

Code: Select all

#  req_size  fwd_wt  fwd_rd  bwd_wt  bwd_rd  rnd_wt  rnd_rd
#   (bytes)  (MB/s)  (MB/s)  (MB/s)  (MB/s)  (MB/s)  (MB/s)
#----------------------------------------------------------
      16384   64.18   65.03   47.89   33.18    4.44    2.27
      32768   75.78   99.78   88.04   37.53    8.73    4.42
      65536  106.05  111.59  106.76   42.43   16.23    8.48
     131072  112.12  114.28  107.58   49.08   27.45   15.77
     262144  106.30  113.90  106.95   56.80   41.58   27.56
     524288  106.15  113.59  106.60   55.57   57.88   44.60
    1048576  112.30  113.14  107.58   79.46   74.57   62.15
    2097152  111.39  110.91  106.38   85.67   86.97   80.43
    4194304  111.30  109.20  108.82  102.21   98.86   93.09

Filesystem             Type  kbytes     use     avail  %use Mounted on
/dev/root               xfs 71545548 10065788 61479760  15  /
/dev/dsk/dks0d2s7       xfs 143368692 88316288 55052404  62  /i
/dev/dsk/dks4d0s7       xfs 976641304  1058736 975582568   1  /0


So, the 1TB SATA gives 953GB of usable space (via 1MB = 1024KB, etc.). I'm just about to run a speed
test to compare access times to the system disk and Maxtor 15K data disk. More shortly...

Oh, I also tested a SUN SAS drive which worked ok, a Hitachi 146GB 10K, model H101414SCSUN146G. Have
a few more to test this evening...

---

More results from the SAS drives! First, a SUN-badged Hitachi 146GB 10K (H101414SCSUN146GSA23):

Code: Select all

#  req_size  fwd_wt  fwd_rd  bwd_wt  bwd_rd  rnd_wt  rnd_rd
#   (bytes)  (MB/s)  (MB/s)  (MB/s)  (MB/s)  (MB/s)  (MB/s)
#----------------------------------------------------------
      16384   66.93   88.90   12.57    2.80    9.13    3.70
      32768   88.78   88.77    2.86    5.82   17.25    7.09
      65536   88.81   88.81   44.03   12.37   26.42   13.36
     131072   88.85   88.91   28.88   28.93   22.73   23.13
     262144   88.90   88.81   31.21   31.19   36.29   36.41
     524288   88.87   88.81   48.22   43.25   51.17   49.96
    1048576   88.87   88.81   71.19   60.99   65.04   64.45
    2097152   88.81   88.87   80.43   77.48   74.89   73.86
    4194304   88.76   88.77   81.77   81.00   80.67   80.02
    8388608   88.70   88.82   85.23   83.58   84.46   83.79



Here's a SUN-badged Seagate 146GB 10K SAS (ST9146802SS):

Code: Select all

#  req_size  fwd_wt  fwd_rd  bwd_wt  bwd_rd  rnd_wt  rnd_rd
#   (bytes)  (MB/s)  (MB/s)  (MB/s)  (MB/s)  (MB/s)  (MB/s)
#----------------------------------------------------------
      16384   89.90   90.43   46.32    6.93    8.37    3.60
      32768   90.28   91.42   79.54   11.84   15.99    7.06
      65536   90.46   91.60   81.42   19.29   27.38   12.87
     131072   90.76   91.67   80.76   40.48   42.99   22.81
     262144   90.22   91.64   80.61   36.54   57.97   36.35
     524288   90.73   91.56   48.78   49.01   51.38   52.39
    1048576   90.60   91.07   73.19   73.07   65.98   65.84
    2097152   90.60   91.41   82.14   82.87   75.56   76.49
    4194304   90.46   91.52   82.47   83.19   82.89   82.68


Nice! Now here's two of the Seagates in an XLV optimised for HD:

Code: Select all

#  req_size  fwd_wt  fwd_rd  bwd_wt  bwd_rd  rnd_wt  rnd_rd
#   (bytes)  (MB/s)  (MB/s)  (MB/s)  (MB/s)  (MB/s)  (MB/s)
#----------------------------------------------------------
    8294400  178.27  178.26  165.34  149.33  161.85  144.29
   16588800  177.98  178.83  166.97  157.04  167.76  158.16



And now top of the pack, the SAS version of the Seagate 15K mentioned in my earlier post (ST3146855SS):

Code: Select all

#  req_size  fwd_wt  fwd_rd  bwd_wt  bwd_rd  rnd_wt  rnd_rd
#   (bytes)  (MB/s)  (MB/s)  (MB/s)  (MB/s)  (MB/s)  (MB/s)
#----------------------------------------------------------
      16384   89.72   90.32   50.68    8.88   10.45    4.90
      32768  112.64  131.40   83.82   31.66   19.29    9.42
      65536  120.24  131.57  116.86   34.55   34.71   17.76
     131072  126.29  131.47  115.22   59.47   56.88   31.54
     262144  129.34  131.66  115.01   57.21   76.17   50.46
     524288  130.70  131.36   65.12   65.41   72.12   73.24
    1048576  130.75  131.41   86.91   87.12   93.33   93.63
    2097152  130.81  131.41  103.98  104.46  108.36  107.43
    4194304  130.64  131.56  115.35  116.02  118.92  117.31
    8388608  130.58  131.56  122.97  122.64  124.81  122.20


The SCSI version of this drive is significantly slower for sequential read/writes of smaller request sizes, but
much the same for larger sizes, though the SAS is more consistent.


Lastly, though not an optimal arrangement since it's using three different models, here's an XLV of four
146GB 10K SAS drives, optimised for HD:

Code: Select all

#  req_size  fwd_wt  fwd_rd  bwd_wt  bwd_rd  rnd_wt  rnd_rd
#   (bytes)  (MB/s)  (MB/s)  (MB/s)  (MB/s)  (MB/s)  (MB/s)
#----------------------------------------------------------
    8294400  216.83  312.60  212.35  231.79  212.48  256.24
   16588800  217.63  315.13  216.18  273.97  215.73  278.73


Either way, the LSI card works very well!

Ian.

ramq
Posts: 771
Joined: Fri Jan 25, 2008 6:06 am
Location: Sweden

Re: LSI SAS/SATA HBA's and the Fuel (or other IP35 Systems)

Unread postby ramq » Tue Jun 09, 2009 1:27 am

Darn. You beat me to the SAS drive test.
All because my Fuel is still at your warehouse... ;)
:O3200: :Fuel: :Indy: :O3x02L:

User avatar
mapesdhs
Posts: 2496
Joined: Mon Nov 10, 2003 4:17 pm
Location: Edinburgh, Scotland
Contact:

Re: LSI SAS/SATA HBA's and the Fuel (or other IP35 Systems)

Unread postby mapesdhs » Tue Jun 09, 2009 2:51 am

ramq wrote:Darn. You beat me to the SAS drive test.
All because my Fuel is still at your warehouse... ;)


Yeah, sorry about that; couldn't post anything last week, spent all my time tidying up the place ready for
a Landlord/property inspection (rented house). Most inconvenient...

Ian.

ramq
Posts: 771
Joined: Fri Jan 25, 2008 6:06 am
Location: Sweden

Re: LSI SAS/SATA HBA's and the Fuel (or other IP35 Systems)

Unread postby ramq » Tue Jun 09, 2009 4:36 am

I've got a couple of Seagate ST973402SS drives I want to evaluate. :D
:O3200: :Fuel: :Indy: :O3x02L:

User avatar
mapesdhs
Posts: 2496
Joined: Mon Nov 10, 2003 4:17 pm
Location: Edinburgh, Scotland
Contact:

Re: LSI SAS/SATA HBA's and the Fuel (or other IP35 Systems)

Unread postby mapesdhs » Tue Jun 09, 2009 6:26 am

ramq wrote:I've got a couple of Seagate ST973402SS drives I want to evaluate. :D


They should perform the same as the ST9146802SS mentioned earlier, as it's the same model
range (Savio 10K II), ie. 90MB/sec for one drive, just under 180MB/sec for two in XLV.

Ian.

ramq
Posts: 771
Joined: Fri Jan 25, 2008 6:06 am
Location: Sweden

Re: LSI SAS/SATA HBA's and the Fuel (or other IP35 Systems)

Unread postby ramq » Tue Jun 09, 2009 12:05 pm

Most probably, yes.
I've got access to those drives as well, but since you already covered them...
:O3200: :Fuel: :Indy: :O3x02L:

User avatar
hamei
Posts: 10433
Joined: Tue Feb 24, 2004 4:10 pm
Location: over the rainbow

Re: LSI SAS/SATA HBA's and the Fuel (or other IP35 Systems)

Unread postby hamei » Fri Jun 19, 2009 11:57 pm

A slight further update on SATA DVD-RW drives. Ran thru some cdrecord atempts as follows : have two DC-writers, one an LG on a Yamadapter and the other a SATA Pioneer. First the Yamadapet on scsi bus :

Code: Select all

fewel 16# cdrecord -v dev=1,6,0 driveropts=burnfree -dao open_inventor.iso
Cdrecord-ProDVD-ProBD-Clone 2.01.01a35 (mips-sgi-irix6.5) Copyright (C) 1995-2007 Jörg Schilling
TOC Type: 1 = CD-ROM
scsidev: '1,6,0'
scsibus: 1 target: 6 lun: 0
Using libscg version 'schily-0.9'.
Driveropts: 'burnfree'
SCSI buffer size: 64512
atapi: 0
Device type    : Removable CD-ROM
Version        : 2
Response Format: 2
Capabilities   :
Vendor_info    : 'HL-DT-ST'
Identifikation : 'DVDRAM GSA-4167B'
Revision       : 'DL11'
Device seems to be: Generic mmc2 DVD-R/DVD-RW/DVD-RAM.
Current: CD-R
Profile: DVD-RAM
Profile: DVD-R sequential recording
Profile: DVD-R/DL sequential recording
Profile: DVD-R/DL layer jump recording
Profile: DVD-RW sequential recording
Profile: DVD-RW restricted overwrite
Profile: DVD+RW
Profile: DVD+R
Profile: DVD+R/DL
Profile: DVD-ROM
Profile: CD-R (current)
Profile: CD-RW
Profile: CD-ROM
Profile: Removable Disk
Using generic SCSI-3/mmc   CD-R/CD-RW driver (mmc_cdr).
Driver flags   : MMC-3 SWABAUDIO BURNFREE
Supported modes: TAO PACKET SAO SAO/R96P SAO/R96R RAW/R16 RAW/R96P RAW/R96R
Drive buf size : 1053696 = 1029 KB
Drive pbuf size: 1966080 = 1920 KB
Drive DMA Speed: 3238 kB/s 18x CD 2x DVD 0x BD
FIFO size      : 4194304 = 4096 KB
Track 01: data   286 MB       
Total size:      328 MB (32:33.62) = 146522 sectors
Lout start:      328 MB (32:35/47) = 146522 sectors
Current Secsize: 512
ATIP info from disk:
  Indicated writing power: 5
Disk Is not unrestricted
Disk Is not erasable
  Disk sub type: Medium Type A, high Beta category (A+) (3)
  ATIP start of lead in:  -11634 (97:26/66)
  ATIP start of lead out: 359846 (79:59/71)
Disk type:    Short strategy type (Phthalocyanine or similar)
Manuf. index: 3
Manufacturer: CMC Magnetics Corporation

    Capacity  Blklen/Sparesz.  Type
           0             2048  No Media Present or Unknown Capacity
Blocks total: 359846 Blocks current: 359846 Blocks remaining: 213324
cdrecord: DMA speed too slow (OK for 15x). Cannot write at speed 48x.
Starting to write CD/DVD/BD at speed 48 in real SAO mode for single session.
Last chance to quit, starting real write    0 seconds. Operation starts.
Waiting for reader process to fill input buffer ... input buffer ready.
BURN-Free is ON.
Performing OPC...
Sending CUE sheet...
cdrecord: WARNING: Drive returns wrong startsec (0) using -150
Writing pregap for track 1 at -150
Starting new track at sector: 0
Track 01:  286 of  286 MB written (fifo 100%) [buf  90%]   6.0x.
WARNING: padding up to secsize (by 206 bytes).
Track 01: Total bytes read/written: 300076850/300077056 (146522 sectors).
Writing  time:  141.947s
Average write speed  13.8x.
Min drive buffer fill was 7%
Fixating...
Fixating time:    3.266s
cdrecord: fifo had 4727 puts and 4727 gets.
cdrecord: fifo was 0 times empty and 4112 times full, min fill was 95%.


now the unsuccessful attenmpt on the SATA bus :

Code: Select all

fewel 6# cdrecord -v dev=2,2,0 driveropts=burnfree -dao open_inventor.iso
Cdrecord-ProDVD-ProBD-Clone 2.01.01a35 (mips-sgi-irix6.5) Copyright (C) 1995-2007 Jörg Schilling
TOC Type: 1 = CD-ROM
scsidev: '2,2,0'
scsibus: 2 target: 2 lun: 0
Using libscg version 'schily-0.9'.
Driveropts: 'burnfree'
SCSI buffer size: 64512
atapi: 0
Device type    : Removable CD-ROM
Version        : 0
Response Format: 2
Capabilities   :
Vendor_info    : 'PIONEER '
Identifikation : 'DVD-RW  DVR-216 '
Revision       : '1.09'
Device seems to be: Generic mmc2 DVD-R/DVD-RW/DVD-RAM.
Current: CD-R
Profile: DVD+R/DL
Profile: DVD+R
Profile: DVD+RW
Profile: DVD-R/DL layer jump recording
Profile: DVD-R/DL sequential recording
Profile: DVD-RW sequential recording
Profile: DVD-RW restricted overwrite
Profile: DVD-RAM
Profile: Removable Disk
Profile: DVD-R sequential recording
Profile: DVD-ROM
Profile: CD-RW
Profile: CD-R (current)
Profile: CD-ROM
Using generic SCSI-3/mmc   CD-R/CD-RW driver (mmc_cdr).
Driver flags   : MMC-3 SWABAUDIO BURNFREE
Supported modes: TAO PACKET SAO SAO/R96P SAO/R96R RAW/R16 RAW/R96P RAW/R96R LAYER_JUMP
Drive buf size : 1136640 = 1110 KB
FIFO size      : 4194304 = 4096 KB
Track 01: data   286 MB       
Total size:      328 MB (32:33.62) = 146522 sectors
Lout start:      328 MB (32:35/47) = 146522 sectors
Current Secsize: 2048
ATIP info from disk:
  Indicated writing power: 5
Disk Is not unrestricted
Disk Is not erasable
  Disk sub type: Medium Type A, high Beta category (A+) (3)
  ATIP start of lead in:  -11634 (97:26/66)
  ATIP start of lead out: 359846 (79:59/71)
Disk type:    Short strategy type (Phthalocyanine or similar)
Manuf. index: 3
Manufacturer: CMC Magnetics Corporation

    Capacity  Blklen/Sparesz.  Type
           0             2048  Unformated or Blank Media
Blocks total: 359846 Blocks current: 359846 Blocks remaining: 213324
Starting to write CD/DVD/BD at speed 40 in real SAO mode for single session.
Last chance to quit, starting real write    0 seconds. Operation starts.
Waiting for reader process to fill input buffer ... input buffer ready.
BURN-Free is ON.
Performing OPC...
Sending CUE sheet...
Writing pregap for track 1 at -150
Starting new track at sector: 0
Track 01:    0 of  286 MB written.cdrecord: I/O error. write_g1: scsi sendcmd: retryable error
CDB:  2A 00 00 00 01 93 00 00 1F 00
status: 0xff 0x0 (Reserved)
Sense Bytes:
Sense Key: 0xFFFFFFFF [], Segment 0
Sense Code: 0x00 Qual 0x00 (no additional sense information) Fru 0x0
Sense flags: Blk 0 (not valid)
cmd finished after 10.995s timeout 200s

write track data: error after 825344 bytes
cdrecord: A write error occured.
cdrecord: Please properly read the error message above.
cdrecord: I/O error. test unit ready: scsi sendcmd: retryable error
CDB:  00 00 00 00 00 00
status: 0x2 (CHECK CONDITION)
Sense Bytes: 70 00 06 00 00 00 00 0E 00 00 00 00 29 00 00 00 00 00 00 00 00 00 00 00
Sense Key: 0x6 Unit Attention, Segment 0
Sense Code: 0x29 Qual 0x00 (power on, reset, or bus device reset occurred) Fru 0x0
Sense flags: Blk 0 (not valid)
cmd finished after 0.031s timeout 200s
cdrecord: I/O error. test unit ready: scsi sendcmd: retryable error
CDB:  00 00 00 00 00 00
status: 0x2 (CHECK CONDITION)
Sense Bytes: 70 00 02 00 00 00 00 0E 00 00 00 00 04 01 00 00 00 00 00 00 00 05 00 00
Sense Key: 0x2 Not Ready, Segment 0
Sense Code: 0x04 Qual 0x01 (logical unit is in process of becoming ready) Fru 0x0
Sense flags: Blk 0 (not valid)
cmd finished after 0.030s timeout 200s
Writing  time:   31.244s
Average write speed 121.2x.
Fixating...
cdrecord: I/O error. flush cache: scsi sendcmd: retryable error
CDB:  35 00 00 00 00 00 00 00 00 00
status: 0x2 (CHECK CONDITION)
Sense Bytes: 70 00 02 00 00 00 00 0E 00 00 00 00 04 01 00 00 00 00 00 00 00 05 00 00
Sense Key: 0x2 Not Ready, Segment 0
Sense Code: 0x04 Qual 0x01 (logical unit is in process of becoming ready) Fru 0x0
Sense flags: Blk 0 (not valid)
cmd finished after 0.030s timeout 200s
Trouble flushing the cache
Fixating time:   14.586s
cdrecord: fifo had 77 puts and 14 gets.
cdrecord: fifo was 0 times empty and 1 times full, min fill was 92%.


The CD "written" on SATA is unreadable. Any SCSI experts can gather some clues from this ? It seems like it is close but jaan-jaap can save his virtual cigars still :(

lausey
Posts: 6
Joined: Tue Aug 08, 2006 8:51 am
Location: Boston, MA

Re: LSI SAS/SATA HBA's and the Fuel (or other IP35 Systems)

Unread postby lausey » Mon Jun 22, 2009 7:20 pm

I've worked a fair amount with LSI SAS chipsets over the last few years - some notes:

I don't have a Fuel so I can't do any tests with the card in an SGI.

1. The 1068 A0 rev is indeed a prototype chip, can't run new versions of firmware. I would not suggest buying any cards without knowing it had an A3 rev or greater 1064/1068 chipset, the previous revs were protos and were never released to the public.
2. Most of the SAS3041X-R boards I've seen have the A3 rev chip on them. (These are all low profile cards and come with a low profile bracket or full height bracket.
3. There was an issue with SATA dvd drives that I believe was fixed in the newest versions of the firmware for the 1064/1068 A3 and 1064e/1068e (pci express versions of the same chip) B2 & B3. No idea if this would fix the problem in IRIX though.
4. There should be no problems using the firmware posted on the LSI website for the Sun & HP OEM versions of the cards as long as you pick the correct chipset type, you can flash in a PC booted into DOS with the files on the LSI website. These flash updates also update the fcode for using the cards in a Sun SPARC system.

User avatar
mapesdhs
Posts: 2496
Joined: Mon Nov 10, 2003 4:17 pm
Location: Edinburgh, Scotland
Contact:

Re: LSI SAS/SATA HBA's and the Fuel (or other IP35 Systems)

Unread postby mapesdhs » Wed Jan 05, 2011 2:52 am

Quick update for those who may have missed my other post...

My Fuel now has a 120GB SSD, full details here.

Meanwhile, the fastest mechanical disk I've tested with an LSI SAS/SATA card is the Seagate 450GB 15K SAS (15K.7 series),
model ST3450857SS (I have several for sale if anyone's interested). Here's the diskperf, tested with a Fuel/900 using a SAS3442X-R:

Code: Select all

#  req_size  fwd_wt  fwd_rd  bwd_wt  bwd_rd  rnd_wt  rnd_rd
#   (bytes)  (MB/s)  (MB/s)  (MB/s)  (MB/s)  (MB/s)  (MB/s)
#----------------------------------------------------------
       4096   38.83   40.26   19.03   13.10    3.11    1.34
       8192   68.98   70.94   34.82   20.83    6.22    2.67
      16384  110.34  112.64   61.36   31.14   12.28    5.35
      32768  154.20  156.89   99.76   42.16   22.76   10.52
      65536  194.83  178.83  143.22   57.72   41.58   19.79
     131072  206.30  198.62  170.62   63.65   69.34   35.18
     262144  206.03  207.51  163.79   95.62   98.67   59.84
     524288  205.83  207.88   95.40   96.06   95.99   92.94
    1048576  206.58  207.95  149.61  145.27  127.19  123.02
    2097152  206.24  207.90  152.67  150.97  158.85  156.52
    4194304  206.77  207.91  184.26  171.54  180.34  176.95
    8388608  206.14  207.91  193.90  189.59  191.76  189.09



And here's a diskperf for two of them, using a SAS3080X card in a Fuel/900:

Code: Select all

#  req_size  fwd_wt  fwd_rd  bwd_wt  bwd_rd  rnd_wt  rnd_rd
#   (bytes)  (MB/s)  (MB/s)  (MB/s)  (MB/s)  (MB/s)  (MB/s)
#----------------------------------------------------------
    8306688  219.31  313.41  210.87  241.13  209.23  240.08
   16613376  220.86  315.04  214.12  281.44  215.24  280.99
   33226752  220.73  315.53  218.37  296.50  218.12  297.58



Do check the SSD results in my other post though, the results are interesting to say the least.

Ian.

User avatar
mia
Posts: 1055
Joined: Wed Feb 19, 2003 1:54 pm

Re: LSI SAS/SATA HBA's and the Fuel (or other IP35 Systems)

Unread postby mia » Sat Oct 06, 2012 5:40 pm

Not bad, my turn.

Code: Select all

jason$ mkfile 6g testfile ; diskperf -W -D -r 4k -m 4m testfile
#---------------------------------------------------------
# Disk Performance Test Results Generated By Diskperf V1.2
#
# Test name     : Unspecified
# Test date     : Sat Oct  6 16:54:54 2012
# Test machine  : IRIX64 hydra 6.5 07202013 IP35
# Test type     : XFS data subvolume
# Test path     : testfile
# Request sizes : min=4096 max=4194304
# Parameters    : direct=1 time=10 scale=1.000 delay=0.000
# XFS file size : 6442450944 bytes
#---------------------------------------------------------
# req_size  fwd_wt  fwd_rd  bwd_wt  bwd_rd  rnd_wt  rnd_rd
#  (bytes)  (MB/s)  (MB/s)  (MB/s)  (MB/s)  (MB/s)  (MB/s)
#---------------------------------------------------------
       4096   19.72   23.50   15.11   11.12    6.04    1.35
       8192   37.85   44.73   28.20   14.81   12.25    2.68
      16384   65.83   75.17   51.67   16.56   23.96    5.23
      32768  107.45  115.58   82.97   10.81   44.62   10.07
      65536  151.64  147.51  126.62   21.80   80.12   19.14
     131072  239.40  274.34  207.96   35.51   80.77   31.53
     262144  278.30  391.11  119.72   66.84  132.85   55.84
     524288  321.62  391.24  196.07  105.88  200.64   96.11
    1048576  350.83  391.28  201.78  158.65  258.66  157.86
    2097152  364.98  390.96  126.92  220.66  122.15  217.32
    4194304  373.96  391.04  197.45  284.72  195.51  277.28


4xIP53@1Ghz, xvm raid 0+1, 4xST9300653SS.
:Onyx2:

User avatar
mapesdhs
Posts: 2496
Joined: Mon Nov 10, 2003 4:17 pm
Location: Edinburgh, Scotland
Contact:

Re: LSI SAS/SATA HBA's and the Fuel (or other IP35 Systems)

Unread postby mapesdhs » Sun Oct 07, 2012 2:11 am

Cool!!

What controller setup was this using? One card? Two cards? And in which slots?

Ian.
I'm working on a charitable PC build for the Learn Engineering YouTube channel. Please PM/email/call if you'd like to contribute!
Donations of any kind of item I can sell to provide funds are also most welcome.
mapesdhs@yahoo.com
+44 (0)7434 635 121

User avatar
mia
Posts: 1055
Joined: Wed Feb 19, 2003 1:54 pm

Re: LSI SAS/SATA HBA's and the Fuel (or other IP35 Systems)

Unread postby mia » Sun Oct 07, 2012 7:10 pm

mapesdhs wrote:What controller setup was this using?

Code: Select all

Integral SCSI controller 6: Version SAS/SATA LS1068
  Disk drive: unit 0 on SCSI controller 6 (XVM Local Disk) (primary path)
  Disk drive: unit 1 on SCSI controller 6 (XVM Local Disk) (primary path)
Integral SCSI controller 3: Version SAS/SATA LS1068
  Disk drive: unit 0 on SCSI controller 3 (XVM Local Disk) (primary path)
  Disk drive: unit 1 on SCSI controller 3 (XVM Local Disk) (primary path)


mapesdhs wrote:And in which slots?

Code: Select all

Bus Slot Stat    Power Mode/Speed
[...snip...]
  3    2 0x00 0c  7.5W PCIX 133MHz
  4    1 0x00 0c  7.5W PCIX 133MHz


Let's take this thing up a notch, with raid0, same hardware configuration:

Code: Select all

jason$ mkfile 4g testfile ; sync ; diskperf -W -D -r 4k -m 4m testfile
#---------------------------------------------------------
# Disk Performance Test Results Generated By Diskperf V1.2
#
# Test name     : Unspecified
# Test date     : Sun Oct  7 18:56:56 2012
# Test machine  : IRIX64 hydra 6.5 07202013 IP35
# Test type     : XFS data subvolume
# Test path     : testfile
# Request sizes : min=4096 max=4194304
# Parameters    : direct=1 time=10 scale=1.000 delay=0.000
# XFS file size : 4294967296 bytes
#---------------------------------------------------------
# req_size  fwd_wt  fwd_rd  bwd_wt  bwd_rd  rnd_wt  rnd_rd
#  (bytes)  (MB/s)  (MB/s)  (MB/s)  (MB/s)  (MB/s)  (MB/s)
#---------------------------------------------------------
       4096   26.07   25.16   18.97   14.25   13.54    1.47
       8192   47.78   46.60   33.74   20.20   26.53    2.87
      16384   82.62   78.85   61.31   25.04   48.55    5.64
      32768  124.91  122.39   96.80   20.33   86.26   10.93
      65536  166.90  154.36  135.94   20.90  133.95   20.19
     131072  331.76  308.49  268.67   34.62  176.46   33.89
     262144  475.52  460.05  430.89   78.23  177.99   58.86
     524288  539.13  515.28  197.42  126.35  286.29  106.31
    1048576  622.51  600.20  398.52  178.15  422.77  184.51
    2097152  692.61  660.22  396.16  287.42  538.49  273.79
    4194304  727.27  685.00  244.63  387.93  243.50  408.68
:Onyx2:


Return to “SGI: Hardware”

Who is online

Users browsing this forum: mazzinia and 2 guests