|
[Veritas-bu] PCI bus and HBA throughput
2003-04-25 08:57:14
This is a multi-part message in MIME format.
------=_NextPart_000_0005_01C30B08.AA7C69B0
Content-Type: text/plain;
charset="us-ascii"
Content-Transfer-Encoding: quoted-printable
Hi Vijay
=20
Here is what we found using two other brands of PCI Bus 2Gbit HBA's in
SunFire 880's going to T9904B's:
1) To a single drive using 2Gbit HBA's, observed peak ~40MByte/sec,
regardless of 33 or 66Mhz bus
2) Writing 3 streams to 3 drives across the same 2Gbit HBA, observed =
peak
~100MByte/sec regardless of 33 or 66Mhz bus
During all tests source data was fully cached in RAM (machine had =
8GByte)
and no other processes running (machine had 8 CPU's)
=20
Wconcluded that the bottleneck was likely in the handling of the SAN
protocols across the HBA. Our basis for this conclusion was this:
- Writing zeroes (/dev/zero) to a single T9940B drive yielded a rate of
69MByte/sec which is basically the upper limit at which StorageTek =
claims
the drive can take data in; so we know it can go that fast.
- Pushing zeroes to 2 drives simultaneously can be done at an aggregate =
rate
of ~130MByte/sec on the 66MHz bus, closer to ~120MByte/sec on the 33Mhz =
bus.
- Going to 3 simultaneous drives with zeroes yields rates of =
~138MByte/sec
and 129MByte/sec respectivel for the 66MHz and 33Mhz buses.
In other words, the bus speed gave the 66Mhz bus a slight edge but =
hardly
worth bothering. It would be very interesting to see what the results =
look
like when you use multiple HBA's simultaneously...
=20
My advice: don't hang more than 2 drives of a single HBA.
=20
PaulW
www.pbnj-solutions.com <http://www.pbnj-solutions.com/> IT Solutions =
That
Stick
------=_NextPart_000_0005_01C30B08.AA7C69B0
Content-Type: text/html;
charset="us-ascii"
Content-Transfer-Encoding: quoted-printable
<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.0 Transitional//EN">
<HTML><HEAD>
<META HTTP-EQUIV=3D"Content-Type" CONTENT=3D"text/html; =
charset=3Dus-ascii">
<TITLE>Message</TITLE>
<META content=3D"MSHTML 6.00.2800.1170" name=3DGENERATOR>
<STYLE></STYLE>
</HEAD>
<BODY bgColor=3D#ffffff>
<DIV><FONT face=3DArial color=3D#0000ff size=3D2><SPAN =
class=3D426311712-25042003>Hi=20
Vijay</SPAN></FONT></DIV>
<DIV><FONT face=3DArial color=3D#0000ff size=3D2><SPAN=20
class=3D426311712-25042003></SPAN></FONT> </DIV>
<DIV><FONT face=3DArial color=3D#0000ff size=3D2><SPAN =
class=3D426311712-25042003>Here=20
is what we found using two other brands of PCI Bus 2Gbit HBA's in =
SunFire 880's=20
going to T9904B's:</SPAN></FONT></DIV>
<DIV><FONT face=3DArial color=3D#0000ff size=3D2><SPAN =
class=3D426311712-25042003>1) To=20
a single drive using 2Gbit HBA's, observed peak ~40MByte/sec, regardless =
of 33=20
or 66Mhz bus</SPAN></FONT></DIV>
<DIV><FONT face=3DArial color=3D#0000ff size=3D2><SPAN =
class=3D426311712-25042003>2)=20
Writing 3 streams to 3 drives across the same 2Gbit HBA, observed peak=20
~100MByte/sec regardless of 33 or 66Mhz bus</SPAN></FONT></DIV>
<DIV><FONT face=3DArial color=3D#0000ff size=3D2><SPAN =
class=3D426311712-25042003>During=20
all tests source data was fully cached in RAM (machine had 8GByte) and =
no other=20
processes running (machine had 8 CPU's)</SPAN></FONT></DIV>
<DIV><FONT face=3DArial color=3D#0000ff size=3D2><SPAN=20
class=3D426311712-25042003></SPAN></FONT> </DIV>
<DIV><FONT face=3DArial color=3D#0000ff size=3D2><SPAN=20
class=3D426311712-25042003>Wconcluded that the bottleneck was likely in =
the=20
handling of the SAN protocols across the HBA. Our basis for this=20
conclusion was this:</SPAN></FONT></DIV>
<DIV><FONT face=3DArial color=3D#0000ff size=3D2><SPAN =
class=3D426311712-25042003>-=20
Writing zeroes (/dev/zero) to a single T9940B drive yielded a rate of=20
69MByte/sec which is basically the upper limit at which StorageTek =
claims the=20
drive can take data in; so we know it can go that =
fast.</SPAN></FONT></DIV>
<DIV><FONT face=3DArial color=3D#0000ff size=3D2><SPAN =
class=3D426311712-25042003>-=20
Pushing zeroes to 2 drives simultaneously can be done at an =
aggregate rate=20
of ~130MByte/sec on the 66MHz bus, closer to ~120MByte/sec on the 33Mhz=20
bus.</SPAN></FONT></DIV>
<DIV><FONT face=3DArial color=3D#0000ff size=3D2><SPAN =
class=3D426311712-25042003>-=20
Going to 3 simultaneous drives with zeroes yields rates of ~138MByte/sec =
and=20
129MByte/sec respectivel for the 66MHz and 33Mhz =
buses.</SPAN></FONT></DIV>
<DIV><FONT face=3DArial color=3D#0000ff size=3D2><SPAN =
class=3D426311712-25042003>In=20
other words, the bus speed gave the 66Mhz bus a slight edge but hardly =
worth=20
bothering. It would be very interesting to see what the results =
look like=20
when you use multiple HBA's simultaneously...</SPAN></FONT></DIV>
<DIV><FONT face=3DArial color=3D#0000ff size=3D2><SPAN=20
class=3D426311712-25042003></SPAN></FONT> </DIV>
<DIV><FONT face=3DArial color=3D#0000ff size=3D2><SPAN =
class=3D426311712-25042003>My=20
advice: don't hang more than 2 drives of a single =
HBA.</SPAN></FONT></DIV>
<DIV><FONT face=3DArial color=3D#0000ff size=3D2><SPAN=20
class=3D426311712-25042003></SPAN></FONT> </DIV>
<DIV align=3Dleft><FONT face=3DArial size=3D2>PaulW</FONT></DIV>
<DIV align=3Dleft><FONT face=3DArial size=3D2><A=20
href=3D"http://www.pbnj-solutions.com/";>www.pbnj-solutions.com</A> IT =
Solutions=20
That Stick</FONT></DIV></BODY></HTML>
------=_NextPart_000_0005_01C30B08.AA7C69B0--
|
|
|
