[Subject Prev][Subject Next][Thread Prev][Thread Next][Subject Index][Thread Index]
Linux is hungry
- Subject: Linux is hungry
- From: "S.Gopinath" <sxg@xxxxxxxxx>
- Date: Mon, 19 Jul 1999 09:03:41 +0530
This is a multi-part message in MIME format.
------=_NextPart_000_0079_01BED1C5.98D067F0
Content-Type: multipart/alternative;
boundary="----=_NextPart_001_007A_01BED1C5.98D067F0"
------=_NextPart_001_007A_01BED1C5.98D067F0
Content-Type: text/plain;
charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable
Dear Friends,
I'm attachaing an article which discusses about Linux and NT. I think =
Linux Supporters needs to do a lot to popularize the usage of Linux at =
Entreproise level.=20
It requires improvement in SMP (suggested by many recent Benchmarks) =
and good integartion (in many respects).
Bye,
S.Gopinath
=20
=20
=20
PENGUIN'S BREW - AUGUST 1999 =20
-------------------------------------------------------------------- =
=20
Read Penguin's Brew in Performance Computing=20
FREE subscriptions available=20
-------------------------------------------------------------------- =
=20
=20
Darn The Torpedoes=20
Jon maddog Hall=20
As I write this month's article, it is the end of April. The =
roses are beginning to leaf out, and the thorns are becoming visible, as =
well. So it is not surprising that two events occurred that reveal =
weaknesses in Linux and the Linux community.=20
The first of these events was the publication of the D.H. =
Brown report comparing Linux with Windows NT v. 4.0 and commercial UNIX =
systems. The second event was Microsoft's funding of a test laboratory =
to benchmark Linux vs. Microsoft Windows NT 4.0.=20
D.H. Brown is a technical analyst firm that specializes in =
comparing OSs for suitability in various marketplaces, one of which is =
the enterprise-server market. I should point out that the commercial =
UNIX OSs have been under D.H. Brown's scrutiny for years - the firm has =
compared AIX, HP-UX, Solaris, Tru64 UNIX, and so on. Recently, Microsoft =
Windows NT 4.0 has been added to the list, and the report shows how the =
commercial UNIX OSs have matured to become candidates to replace the =
traditional, proprietary mainframe systems (MVS, MPE, and so on) in =
performing 24x7 functions, whereas Windows NT still falls short. =
Unfortunately, so does Linux.=20
Now, before I have 20,000 Linux people write me hate mail =
(or worse yet, get Slashdotted), let me explain what an enterprise =
server means, and why Linux fell short of the mark.=20
To run a system continuously, you have to detect failing =
hardware by looking at the error-message file and seeing soft hits =
increasing. Using disk drives as an example, let's say that your =
disk-drive controller is beginning to fail. Ideally, your system would =
mark the failing disk read-only, block additional writes to the disk, =
and start to write the data to another drive, while copying the failing =
disk blocks to a new drive as well. When the old disk is empty of any =
data files, you should be able to remove it from the file system, bring =
that disk offline, replace it with another disk, bring that disk online, =
bring it into the existing file system where the old disk was attached, =
and have the disk take its place along with all the other disks in that =
file system. The file systems should be able to handle reasonably sized =
disks and disk farms (in the case of Intel Linux, over 2GB in size) and =
should be able to have journaling, and not have to depend on fsck to =
check out a really large file system. An enterprise OS should be able to =
scale fairly well, so when you buy your next two or three CPUs and put =
them into your server system, you can take advantage of those CPUs' =
power. These features (and many others) are the ones D.H. Brown examines =
in its study.=20
D.H. Brown is very thorough when generating comparisons of =
OSs. They base these comparisons on input from MIS managers, weighing =
the OS's functionality with the importance given to features by these =
managers. For instance, if the managers rate TCP/IP performance as =
critical and NFS performance as secondary, the D.H. Brown report gives =
extra weight to TCP/IP by weighing its factors higher, and assigning =
lower weight to NFS capabilities. Also, D.H. Brown (like some other =
analysts) does not count third-party add-ons integrated into a system by =
users. It measures only functions that are shipped by the OS =
manufacturer. D.H. Brown also wants to see the systems in use at a =
customer's location, hopefully in a stable state. Finally, in conditions =
such as scalability across multiple CPUs, D.H. Brown looks for standard =
benchmarks that have been run and verified.=20
Linux has projects alive on many fronts. Some actually fill =
perceived holes in Linux's capability as an enterprise server. For =
instance, I am aware of at least two process-migration and =
checkpoint/restart projects that would lessen the vulnerability of =
long-running programs. There is also a project that will deliver 64-bit =
file systems and journaled file systems.=20
When comparing OS scalability, D.H. Brown took significant =
steps to capture the newest features by using Caldera's newest release =
that incorporates the 2.2 kernel, but they did not run the benchmarks =
themselves. Instead, they swept the Internet looking for other companies =
that have done benchmarks. Unfortunately, there were none to be found, =
so when ranking scalability, Linux got a no-show compared with the other =
operating systems. Benchmarks obviously rely on hardware as well as =
software performance, and so should be done by the computer companies =
supporting that hardware. At this point, it is up to the Compaqs, IBMs, =
SGIs, and Suns to do the tests, and be visible.=20
I spoke with Linus [Torvalds] about this report (which he =
has seen). We both had more or less the same reaction to it, which was =
first to be angry about the report, but then realize that Linux has =
never been positioned as an enterprise server of the type D.H. Brown =
tests. Linux has always been positioned as a desktop or low-end server =
machine, and only recently as a replacement for high-end =
super-computers. In these environments, the D.H. Brown report praises =
Linux, so apparently we are right on target.=20
I thought of the D.H. Brown report as a suggested project =
list for Linux. Let's take a look at what the commercial community =
needs, and then provide them with it at Linux speed. Let's integrate =
this great new functionality coming out, get those benchmarks done and =
then ask D.H. Brown to update their report (which they have already =
agreed to do) to reflect the additional information.=20
Gulliver Awakes
The second event that occurred this month happened around =
the time that Linus and I were on a panel at an AIIM conference in =
Atlanta. As the panel members met and became familiar with each other, =
the lone Microsoft representative on the panel seemed agitated. He =
grabbed his notebook computer and went online to retrieve a press =
release that Microsoft had put out that day, which he then showed to =
Linus and me.=20
In April, Microsoft hired a test lab to run benchmarks of =
Windows NT and Linux. Windows NT emerged triumphant, scoring two or =
three times the throughput that Linux was able to produce. When the =
Microsoft representative brought this up in front of the audience, the =
only remark that I could make was that I had participated in many =
benchmarks, and to paraphrase Mark Twain, there are only three types of =
lies: Lies, Damn Lies, and Benchmarks. Nevertheless, benchmarks should =
be taken seriously.=20
Well-done benchmarks have several benefits. They help =
customers determine what size systems they should be buying for a =
particular purpose. If a customer's current system was rated at a =
particular throughput, then when the new system is rated as being three =
times faster, it gives the customer a chance to know that (in ideal =
circumstances) they should expect three times the load capabilities.=20
A second benefit to benchmarking is to show the OS and =
hardware people where the bottlenecks are. Participating in benchmark =
activity helps the developers see where they can improve the code and =
tighten up the system.=20
Much has been said on the particular comparison of Windows =
NT and Linux, and I will not delve into the details here. However, I =
will say that when doing a benchmark that is going to be widely =
published, the independent agency (assuming you are using one) has an =
obligation to do the best tuning possible of both platforms. I will also =
say that a set of industry-standard benchmarks showing Linux running on =
several different platforms, with a range of CPUs scaling up, would be a =
much better set than Microsoft had in this instance. Again, this should =
be done (in my humble opinion) by the equipment manufacturers, who have =
access to the hardware to =20
set up the large testing configurations needed, and who are =
(for the most part) neutral about which OS wins. The Linux community =
should ensure there are as many Linux tuning resources available as =
there are Windows NT tuning resources for these benchmarks. Most =
importantly, these benchmarks should be done not only to show which =
operating system is best, but to help end users determine the size of =
the systems they need, and help developers determine where the OS needs =
to be improved.=20
A side benefit of this gathering of tuning resources might =
be the documentation of what tuning should be done to a particular =
distribution of Linux to get it tuned properly for use as a database =
platform, file and print Server, or Web server, perhaps getting system =
administrators to within 90 percent of a fully tuned system, without =
having to read ten books and view 50 Web sites.=20
I feel that the system vendors (Compaq, HP, IBM, Sun =
Microsystems, and others) that have pledged support for Linux should =
make the investment to not only perform these benchmarks, but feed =
detailed information back to the Linux development community, so the =
community can improve Linux's throughput. It would be outstanding if =
these benchmarking efforts could be scheduled in conjunction with some =
large Linux event (such as Linux World in August). This might allow the =
attending developers to actually participate in tuning the system, and =
get instant feedback on what needs to be modified for the best =
performance.=20
Two for Two
Here we have two incidents of commercial entities looking at =
Linux from the eyes of the commercial community, utilizing analysts and =
benchmarks. There will be more of this type of analysis, and we have to =
be ready for it and participate in it. It is not enough at this point =
just to state anecdotal stories and hope they will swing users to Linux. =
We will have to be ready with true, factual, measured reports, with the =
results readily available. Then people can make their own informed =
comparisons.=20
=20
Jon maddog Hall is the executive director of Linux =
International and the senior leader of the UNIX Software Group at Compaq =
Corp.=20
=20
BACK TO TOP=20
Features | Opinions | Columns | Reviews | Subscribe | =
Feedback=20
Linux-IT | Hot Products=20
=20
=A91999 Miller Freeman, Inc.=20
=20
=20
------=_NextPart_001_007A_01BED1C5.98D067F0
Content-Type: text/html;
charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable
<!DOCTYPE HTML PUBLIC "-//W3C//DTD W3 HTML//EN">
<HTML>
<HEAD>
<META content=3Dtext/html;charset=3Diso-8859-1 =
http-equiv=3DContent-Type>
<META content=3D'"MSHTML 4.72.3110.7"' name=3DGENERATOR>
</HEAD>
<BODY bgColor=3D#ffffff>
<DIV><FONT color=3D#000000 size=3D2>Dear Friends,</FONT></DIV>
<DIV><FONT color=3D#000000 size=3D2></FONT> </DIV>
<DIV><FONT color=3D#000000 size=3D2> I'm attachaing an =
article which=20
discusses about Linux and NT. I think Linux Supporters needs to do a lot =
to=20
popularize the usage of Linux at Entreproise level. </FONT></DIV>
<DIV><FONT color=3D#000000 size=3D2>It requires improvement in SMP =
(suggested by=20
many recent Benchmarks) and good integartion (in many=20
respects).</FONT></DIV>
<DIV><FONT color=3D#000000 size=3D2></FONT> </DIV>
<DIV><FONT color=3D#000000 size=3D2>Bye,</FONT></DIV>
<DIV><FONT color=3D#000000 size=3D2>S.Gopinath</FONT></DIV>
<DIV><FONT color=3D#000000 size=3D2></FONT><!-- Begin Logo/Banner Table =
--><A=20
name=3DTop></A>
<TABLE border=3D0 cellPadding=3D2 cellSpacing=3D2 width=3D620>
<TBODY>
<TR><!-- Begin Logo -->
<TD width=3D150><A =
href=3D"http://www.performancecomputing.com/";><IMG=20
alt=3D"Performance Computing" border=3D0 height=3D49=20
src=3D"cid:005501bed197$7b588be0$31a764a4@xxxxxxxxxxxxxxx"=20
width=3D150></A></TD><!-- End Logo --><!-- Gutter -->
<TD width=3D5></TD><!-- Begin Ad Banner -->
<TD colSpan=3D5 width=3D470>
<CENTER><A=20
=
href=3D"http://www.ads.mfi.com/cgi-bin/accipiter/adclick.exe/CID=3D000004=
6b0000000000000000/URI=3DLINUX-IT/PENGUIN/9908.SHTML/SITE=3DPERFORMANCECO=
MPUTING.COM/KEYWORD=3D2ND?46206664"=20
target=3D_blank><IMG=20
alt=3D"Like a SUN system for a lot less money (heck, get =
TWO)"=20
border=3D0=20
=
src=3D"cid:005701bed197$7b5a1280$31a764a4@xxxxxxxxxxxxxxx";></A><!-- =
http://www.ads.mfi.com/cgi-bin/accipiter/adhtml.exe/Uri=3DLinux-IT/pengui=
n/9908.shtml/Site=3Dperformancecomputing.com/Keyword=3D2nd?46206664 -->=20
</CENTER></TD><!-- End Ad Banner =
--></TR></TBODY></TABLE><!-- End Logo/Banner Table --><BR><!-- Begin =
Main Table -->
<TABLE border=3D0 cellPadding=3D2 cellSpacing=3D2 width=3D620>
<TBODY>
<TR><!-- Begin Navigation Bar -->
<TD align=3Dright rowSpan=3D6 vAlign=3Dtop width=3D110><IMG =
border=3D0=20
height=3D25 =
src=3D"cid:005901bed197$7b5a1280$31a764a4@xxxxxxxxxxxxxxx"=20
width=3D110> <IMG alt=3D"" border=3D0 height=3D609=20
src=3D"cid:005b01bed197$7f0455d0$31a764a4@xxxxxxxxxxxxxxx"=20
useMap=3D#navbar.columns width=3D111> <MAP =
name=3Dnavbar.columns><AREA=20
alt=3DFeatures coords=3D0,0,111,23 href=3D"/features/" =
shape=3DRECT=20
title=3DFeatures><AREA alt=3DOpinions =
coords=3D0,24,110,46=20
href=3D"/opinions/" shape=3DRECT title=3DOpinions><AREA =
alt=3DColumns=20
coords=3D0,48,111,71 href=3D"/columns/" shape=3DRECT=20
title=3DColumns><AREA alt=3D"Cross Thoughts" =
coords=3D-1,72,109,84=20
href=3D"/columns/indexcol.shtml#Cross" shape=3DRECT=20
title=3D"Cross Thoughts"><AREA alt=3DCurrents =
coords=3D1,87,111,99=20
href=3D"/columns/indexcol.shtml#Currents" shape=3DRECT=20
title=3DCurrents><AREA alt=3D"Daemons And Dragons"=20
coords=3D0,101,110,122 =
href=3D"/columns/indexcol.shtml#Daemons"=20
shape=3DRECT title=3D"Daemons And Dragons"><AREA =
alt=3DIntegration=20
coords=3D1,124,110,137 =
href=3D"/columns/indexcol.shtml#Integration"=20
shape=3DRECT title=3DIntegration><AREA alt=3D"Java =
Advisor"=20
coords=3D-1,137,110,149 =
href=3D"/columns/indexcol.shtml#Java"=20
shape=3DRECT title=3D"Java Advisor"><AREA alt=3D"Net =
Worth"=20
coords=3D-2,149,109,163 =
href=3D"/columns/indexcol.shtml#Net"=20
shape=3DRECT title=3D"Net Worth"><AREA alt=3DNetopian=20
coords=3D-1,163,110,176 =
href=3D"/columns/indexcol.shtml#Netopian"=20
shape=3DRECT title=3DNetopian><AREA alt=3D"Packets & =
Protocols"=20
coords=3D0,178,110,200 =
href=3D"/columns/indexcol.shtml#Packets"=20
shape=3DRECT title=3D"Packets & Protocols"><AREA=20
alt=3D"Penguin's Brew" coords=3D-2,200,109,216 =
href=3D"homepage.shtml"=20
shape=3DRECT title=3D"Penguin's Brew"><AREA alt=3D"Perl =
Advisor"=20
coords=3D0,216,109,232 =
href=3D"/columns//indexcol.shtml#Perl"=20
shape=3DRECT title=3D"Perl Advisor"><AREA alt=3D"Rules =
Of The Game"=20
coords=3D1,231,111,253 =
href=3D"/columns/indexcol.shtml#Rules"=20
shape=3DRECT title=3D"Rules Of The Game"><AREA =
alt=3D"Web Advisor"=20
coords=3D1,255,111,268 =
href=3D"/columns/indexcol.shtml#Web"=20
shape=3DRECT title=3D"Web Advisor"><AREA alt=3DReviews=20
coords=3D-1,270,108,293 href=3D"/reviews/" shape=3DRECT=20
title=3DReviews><AREA alt=3DSubscribe =
coords=3D0,296,109,318=20
href=3D"/sub.shtml" shape=3DRECT title=3DSubscribe><AREA =
alt=3DFeedback=20
coords=3D0,320,110,343 href=3D"/feedback.shtml" =
shape=3DRECT=20
title=3DFeedback><AREA alt=3DFYI coords=3D-1,462,111,495 =
=
href=3D"http://ww2.infoxpress.com/mfi/performancecomputing/"=20
shape=3DRECT title=3DFYI><AREA alt=3D"UNIX Review"=20
coords=3D-13,497,109,524 href=3D"/unixreview/" =
shape=3DRECT=20
title=3D"UNIX Review"><AREA alt=3Dunixintegration.com=20
coords=3D-1,525,110,556 href=3D"/unixintegration/" =
shape=3DRECT=20
title=3Dunixintegration.com><AREA alt=3D"What's =
Happening?"=20
coords=3D-2,556,111,607 href=3D"/happenin/" shape=3DRECT =
title=3D"What's Happening?"><AREA alt=3DLinux-IT=20
coords=3D0,350,112,412 href=3D"/Linux-IT/" shape=3DRECT=20
title=3DLinux-IT><AREA alt=3D"Hot Products" =
coords=3D0,412,111,459=20
href=3D"/hotproducts/" shape=3DRECT title=3D"Hot =
Products"><AREA=20
coords=3D0,0,0,0 shape=3DRECT></MAP>
<CENTER>
<FORM=20
=
action=3D/scripts/wais/SFcapture.pl?Full~Site~Search+/srchtop.shtml+/srch=
bot.shtml+=20
method=3Dpost>
<INPUT name=3Ddatabase type=3Dhidden=20
value=3Dlocal//htdocs/percomp/wais-src/Full_Site_Search> =
<IMG=20
alt=3DSearch border=3D0 height=3D17=20
src=3D"cid:005d01bed197$7f05dc70$31a764a4@xxxxxxxxxxxxxxx"; =
width=3D110>=20
<INPUT name=3Darg0 size=3D12><BR>
<H4></H4></FORM></CENTER></TD><!-- End Navigation bar =
--><!-- Gutter -->
<TD rowSpan=3D5 width=3D10></TD><!-- Column Name -->
<TD bgColor=3D#3c700e colSpan=3D3 width=3D460>
<CENTER><FONT color=3Dwhite face=3D"ARIAL, =
HELVETICA"><STRONG>PENGUIN'S=20
BREW - AUGUST 1999</STRONG></FONT></CENTER></TD><!-- Gutter =
-->
<TD rowSpan=3D5 width=3D10></TD><!-- Side Column -->
<TD rowSpan=3D5 vAlign=3Dtop width=3D40><IMG border=3D0 =
height=3D900=20
src=3D"cid:005f01bed197$7f05dc70$31a764a4@xxxxxxxxxxxxxxx"; =
width=3D4>
<HR>
<FONT color=3D#009900 face=3D"ARIAL, HELVETICA" =
size=3D-1>Read Penguin's=20
Brew in <EM>Performance Computing</EM>=20
<P><A =
href=3D"http://www.performancecomputing.com/sub.shtml";>FREE=20
subscriptions available</A></FONT>=20
<HR>
</TD></TR>
<TR><!-- Story -->
<TD colSpan=3D3 vAlign=3Dtop width=3D460><FONT face=3D"ARIAL, =
HELVETICA"=20
size=3D+3><!-- Title --><STRONG>Darn The Torpedoes</STRONG> =
</FONT>
<P><!-- Deck --><!-- Byline --><EM>Jon maddog Hall</EM>=20
<P><!-- Story Body --><FONT size=3D+4>A</FONT>s I write this =
month's=20
article, it is the end of April. The roses are beginning to =
leaf=20
out, and the thorns are becoming visible, as well. So it is =
not=20
surprising that two events occurred that reveal weaknesses =
in Linux=20
and the Linux community.=20
<P>The first of these events was the publication of the D.H. =
Brown=20
report comparing Linux with <IMG align=3Dright =
alt=3D"Penguin's Brew"=20
border=3D0 height=3D200=20
src=3D"cid:006101bed197$7f05dc70$31a764a4@xxxxxxxxxxxxxxx"; =
width=3D188>=20
Windows NT v. 4.0 and commercial UNIX systems. The second =
event was=20
Microsoft's funding of a test laboratory to benchmark Linux =
vs.=20
Microsoft Windows NT 4.0.=20
<P>D.H. Brown is a technical analyst firm that specializes =
in=20
comparing OSs for suitability in various marketplaces, one =
of which=20
is the enterprise-server market. I should point out that the =
commercial UNIX OSs have been under D.H. Brown's scrutiny =
for years=20
- the firm has compared AIX, HP-UX, Solaris, Tru64 UNIX, and =
so on.=20
Recently, Microsoft Windows NT 4.0 has been added to the =
list, and=20
the report shows how the commercial UNIX OSs have matured to =
become=20
candidates to replace the traditional, proprietary mainframe =
systems=20
(MVS, MPE, and so on) in performing 24x7 functions, whereas =
Windows=20
NT still falls short. Unfortunately, so does Linux.=20
<P>Now, before I have 20,000 Linux people write me hate mail =
(or=20
worse yet, get Slashdotted), let me explain what an =
enterprise=20
server means, and why Linux fell short of the mark.=20
<P>To run a system continuously, you have to detect failing =
hardware=20
by looking at the error-message file and seeing soft hits=20
increasing. Using disk drives as an example, let's say that =
your=20
disk-drive controller is beginning to fail. Ideally, your =
system=20
would mark the failing disk read-only, block additional =
writes to=20
the disk, and start to write the data to another drive, =
while=20
copying the failing disk blocks to a new drive as well. When =
the old=20
disk is empty of any data files, you should be able to =
remove it=20
from the file system, bring that disk offline, replace it =
with=20
another disk, bring that disk online, bring it into the =
existing=20
file system where the old disk was attached, and have the =
disk take=20
its place along with all the other disks in that file =
system. The=20
file systems should be able to handle reasonably sized disks =
and=20
disk farms (in the case of Intel Linux, over 2GB in size) =
and should=20
be able to have journaling, and not have to depend on fsck =
to check=20
out a really large file system. An enterprise OS should be =
able to=20
scale fairly well, so when you buy your next two or three =
CPUs and=20
put them into your server system, you can take advantage of =
those=20
CPUs' power. These features (and many others) are the ones =
D.H.=20
Brown examines in its study.=20
<P>D.H. Brown is very thorough when generating comparisons =
of OSs.=20
They base these comparisons on input from MIS managers, =
weighing the=20
OS's functionality with the importance given to features by =
these=20
managers. For instance, if the managers rate TCP/IP =
performance as=20
critical and NFS performance as secondary, the D.H. Brown =
report=20
gives extra weight to TCP/IP by weighing its factors higher, =
and=20
assigning lower weight to NFS capabilities. Also, D.H. Brown =
(like=20
some other analysts) does not count third-party add-ons =
integrated=20
into a system by users. It measures only functions that are =
shipped=20
by the OS manufacturer. D.H. Brown also wants to see the =
systems in=20
use at a customer's location, hopefully in a stable state. =
Finally,=20
in conditions such as scalability across multiple CPUs, D.H. =
Brown=20
looks for standard benchmarks that have been run and =
verified.=20
<P>Linux has projects alive on many fronts. Some actually =
fill=20
perceived holes in Linux's capability as an enterprise =
server. For=20
instance, I am aware of at least two process-migration and=20
checkpoint/restart projects that would lessen the =
vulnerability of=20
long-running programs. There is also a project that will =
deliver=20
64-bit file systems and journaled file systems.=20
<P>When comparing OS scalability, D.H. Brown took =
significant steps=20
to capture the newest features by using Caldera's newest =
release=20
that incorporates the 2.2 kernel, but they did not run the=20
benchmarks themselves. Instead, they swept the Internet =
looking for=20
other companies that have done benchmarks. Unfortunately, =
there were=20
none to be found, so when ranking scalability, Linux got a =
no-show=20
compared with the other operating systems. Benchmarks =
obviously rely=20
on hardware as well as software performance, and so should =
be done=20
by the computer companies supporting that hardware. At this =
point,=20
it is up to the Compaqs, IBMs, SGIs, and Suns to do the =
tests, and=20
be visible.=20
<P>I spoke with Linus [Torvalds] about this report (which he =
has=20
seen). We both had more or less the same reaction to it, =
which was=20
first to be angry about the report, but then realize that =
Linux has=20
never been positioned as an enterprise server of the type =
D.H. Brown=20
tests. Linux has always been positioned as a desktop or =
low-end=20
server machine, and only recently as a replacement for =
high-end=20
super-computers. In these environments, the D.H. Brown =
report=20
praises Linux, so apparently we are right on target.=20
<P>I thought of the D.H. Brown report as a suggested project =
list=20
for Linux. Let's take a look at what the commercial =
community needs,=20
and then provide them with it at Linux speed. Let's =
integrate this=20
great new functionality coming out, get those benchmarks =
done and=20
then ask D.H. Brown to update their report (which they have =
already=20
agreed to do) to reflect the additional information.=20
<P>
<H3><FONT face=3D"ARIAL, HELVETICA">Gulliver =
Awakes</FONT></H3>
<P>The second event that occurred this month happened around =
the=20
time that Linus and I were on a panel at an AIIM conference =
in=20
Atlanta. As the panel members met and became familiar with =
each=20
other, the lone Microsoft representative on the panel seemed =
agitated. He grabbed his notebook computer and went online =
to=20
retrieve a press release that Microsoft had put out that =
day, which=20
he then showed to Linus and me.=20
<P>In April, Microsoft hired a test lab to run benchmarks of =
Windows=20
NT and Linux. Windows NT emerged triumphant, scoring two or =
three=20
times the throughput that Linux was able to produce. When =
the=20
Microsoft representative brought this up in front of the =
audience,=20
the only remark that I could make was that I had =
participated in=20
many benchmarks, and to paraphrase Mark Twain, there are =
only three=20
types of lies: Lies, Damn Lies, and Benchmarks. =
Nevertheless,=20
benchmarks should be taken seriously.=20
<P>Well-done benchmarks have several benefits. They help =
customers=20
determine what size systems they should be buying for a =
particular=20
purpose. If a customer's current system was rated at a =
particular=20
throughput, then when the new system is rated as being three =
times=20
faster, it gives the customer a chance to know that (in =
ideal=20
circumstances) they should expect three times the load =
capabilities.=20
=20
<P>A second benefit to benchmarking is to show the OS and =
hardware=20
people where the bottlenecks are. Participating in benchmark =
activity helps the developers see where they can improve the =
code=20
and tighten up the system.=20
<P>Much has been said on the particular comparison of =
Windows NT and=20
Linux, and I will not delve into the details here. However, =
I will=20
say that when doing a benchmark that is going to be widely=20
published, the independent agency (assuming you are using =
one) has=20
an obligation to do the best tuning possible of both =
platforms. I=20
will also say that a set of industry-standard benchmarks =
showing=20
Linux running on several different platforms, with a range =
of CPUs=20
scaling up, would be a much better set than Microsoft had in =
this=20
instance. Again, this should be done (in my humble opinion) =
by the=20
equipment manufacturers, who have access to the hardware to=20
<TABLE align=3Dright cellPadding=3D5 cellSpacing=3D5 =
width=3D200>
<TBODY>
<TR>
<TD><A=20
=
href=3D"http://www.performancecomputing.com/Linux-IT/";><IMG=20
alt=3DLinux-IT border=3D0 height=3D101=20
=
src=3D"cid:006301bed197$7f076310$31a764a4@xxxxxxxxxxxxxxx"=20
width=3D200></A> =
</TD></TR></TBODY></TABLE> set up the large=20
testing configurations needed, and who are (for the most =
part)=20
neutral about which OS wins. The Linux community should =
ensure there=20
are as many Linux tuning resources available as there are =
Windows NT=20
tuning resources for these benchmarks. Most importantly, =
these=20
benchmarks should be done not only to show which operating =
system is=20
best, but to help end users determine the size of the =
systems they=20
need, and help developers determine where the OS needs to be =
improved.=20
<P>A side benefit of this gathering of tuning resources =
might be the=20
documentation of what tuning should be done to a particular=20
distribution of Linux to get it tuned properly for use as a =
database=20
platform, file and print Server, or Web server, perhaps =
getting=20
system administrators to within 90 percent of a fully tuned =
system,=20
without having to read ten books and view 50 Web sites.=20
<P>I feel that the system vendors (Compaq, HP, IBM, Sun=20
Microsystems, and others) that have pledged support for =
Linux should=20
make the investment to not only perform these benchmarks, =
but feed=20
detailed information back to the Linux development =
community, so the=20
community can improve Linux's throughput. It would be =
outstanding if=20
these benchmarking efforts could be scheduled in conjunction =
with=20
some large Linux event (such as Linux World in August). This =
might=20
allow the attending developers to actually participate in =
tuning the=20
system, and get instant feedback on what needs to be =
modified for=20
the best performance.=20
<P>
<H3><FONT face=3D"ARIAL, HELVETICA">Two for Two</FONT></H3>
<P>Here we have two incidents of commercial entities looking =
at=20
Linux from the eyes of the commercial community, utilizing =
analysts=20
and benchmarks. There will be more of this type of analysis, =
and we=20
have to be ready for it and participate in it. It is not =
enough at=20
this point just to state anecdotal stories and hope they =
will swing=20
users to Linux. We will have to be ready with true, factual, =
measured reports, with the results readily available. Then =
people=20
can make their own informed comparisons.=20
<P><BR><!-- Bio --><EM>Jon maddog Hall is the executive =
director of=20
<A href=3D"http://www.li.org/"; target=3Dx>Linux =
International</A> and=20
the senior leader of the UNIX Software Group at <A=20
href=3D"http://www.compaq.com/"; target=3Dx>Compaq Corp.</A> =
</EM>
<P></P></TD></TR>
<TR><!-- Boilerplate -->
<TD align=3Dmiddle colSpan=3D3 vAlign=3Dbottom width=3D390><FONT =
size=3D-1><A=20
=
href=3D"http://www.performancecomputing.com/Linux-IT/penguin/9908.shtml#T=
op">BACK=20
TO TOP</A></FONT>=20
<P><FONT face=3D"Arial, Helvetica" size=3D-2><A=20
=
href=3D"http://www.performancecomputing.com/features/";>Features</A> |=20
<A =
href=3D"http://www.performancecomputing.com/opinions/";>Opinions</A>=20
| <A =
href=3D"http://www.performancecomputing.com/columns/";>Columns</A>=20
| <A =
href=3D"http://www.performancecomputing.com/reviews/";>Reviews</A>=20
| <A=20
=
href=3D"http://www.performancecomputing.com/sub.shtml";>Subscribe</A> |=20
<A=20
=
href=3D"http://www.performancecomputing.com/feedback.shtml";>Feedback</A> =
<BR><A=20
=
href=3D"http://www.performancecomputing.com/Linux-IT/";>Linux-IT</A> |=20
<A =
href=3D"http://www.performancecomputing.com/hotproducts/";>Hot=20
Products</A> </FONT><BR><BR><FONT size=3D-1>©1999 <A=20
href=3D"http://www.mfi.com/";>Miller Freeman, =
Inc.</A></FONT><!-- End Boilerplate --> =
</P></TD></TR></TBODY></TABLE><!-- Begin Bottom Banner Table -->
<TABLE border=3D0 cellPadding=3D2 cellSpacing=3D2 width=3D625>
<TBODY>
<TR><!-- Begin Spacer -->
<TD width=3D150></TD><!--- Gutter -->
<TD width=3D5></TD><!-- Begin Ad Banner -->
<TD colSpan=3D5 width=3D470>
<CENTER><A=20
=
href=3D"http://www.ads.mfi.com/cgi-bin/accipiter/adclick.exe/CID=3D000004=
6b0000000000000000/URI=3DLINUX-IT/PENGUIN/9908.SHTML/SITE=3DPERFORMANCECO=
MPUTING.COM/KEYWORD=3D2ND?41836547"=20
target=3D_blank><IMG=20
alt=3D"Like a SUN system for a lot less money (heck, get =
TWO)"=20
border=3D0=20
=
src=3D"cid:005701bed197$7b5a1280$31a764a4@xxxxxxxxxxxxxxx";></A><!-- =
http://www.ads.mfi.com/cgi-bin/accipiter/adhtml.exe/Uri=3DLinux-IT/pengui=
n/9908.shtml/Site=3Dperformancecomputing.com/Keyword=3D2nd?41836547 -->=20
</CENTER></TD><!-- End Ad Banner =
--></TR></TBODY></TABLE><!-- End Bottom Banner Table --></DIV>
<DIV> </DIV>
<DIV><FONT color=3D#000000 size=3D2></FONT> </DIV></BODY></HTML>
------=_NextPart_001_007A_01BED1C5.98D067F0--
------=_NextPart_000_0079_01BED1C5.98D067F0
Content-Type: image/gif
Content-Transfer-Encoding: base64
Content-ID: <005501bed197$7b588be0$31a764a4@xxxxxxxxxxxxxxx>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------=_NextPart_000_0079_01BED1C5.98D067F0
Content-Type: image/gif
Content-Transfer-Encoding: base64
Content-ID: <005701bed197$7b5a1280$31a764a4@xxxxxxxxxxxxxxx>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------=_NextPart_000_0079_01BED1C5.98D067F0
Content-Type: image/gif
Content-Transfer-Encoding: base64
Content-ID: <005901bed197$7b5a1280$31a764a4@xxxxxxxxxxxxxxx>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------=_NextPart_000_0079_01BED1C5.98D067F0
Content-Type: image/gif
Content-Transfer-Encoding: base64
Content-ID: <005b01bed197$7f0455d0$31a764a4@xxxxxxxxxxxxxxx>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------=_NextPart_000_0079_01BED1C5.98D067F0
Content-Type: image/gif
Content-Transfer-Encoding: base64
Content-ID: <005d01bed197$7f05dc70$31a764a4@xxxxxxxxxxxxxxx>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=
------=_NextPart_000_0079_01BED1C5.98D067F0
Content-Type: image/gif
Content-Transfer-Encoding: base64
Content-ID: <005f01bed197$7f05dc70$31a764a4@xxxxxxxxxxxxxxx>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------=_NextPart_000_0079_01BED1C5.98D067F0
Content-Type: image/gif
Content-Transfer-Encoding: base64
Content-ID: <006101bed197$7f05dc70$31a764a4@xxxxxxxxxxxxxxx>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------=_NextPart_000_0079_01BED1C5.98D067F0
Content-Type: image/jpeg
Content-Transfer-Encoding: base64
Content-ID: <006301bed197$7f076310$31a764a4@xxxxxxxxxxxxxxx>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------=_NextPart_000_0079_01BED1C5.98D067F0--
------------------------------