Looking for feedback: If you were going to make the best improvement and do
it in steps what would you recommend to your client?



My client has a Custom Access application. Version 1 of the program ran
good. Version 2 runs much slower which I assume is because it handles more
information/tables. We are looking at trying to improve the performance with
either hardware upgrades and/or converting the access app. to SQL based. My
client's budget is limited due to the amount he's soaked into the
development of other things.

From what I can see there are 3 possible bottlenecks. Server, Network speed
and the application.



My client Current network: 20 users; (12 max using the access app.)

SBS 2000, Dual Xeon P3 866. SCSI 160, 1.5 gb ram, 100mb Ethernet.



Considering Dell server w/raid 1 & 5. 2gb ram, dual Xeon 3.0 (2mb cache).
Gigabit network. SBS 2003.



My personal thoughts are that converting the Database program to a SQL based
app would show and best increase but I'm looking for some opinions.



Thanks all.

db

Re: Looking for opinions by Cris

Cris
Sat Jul 16 18:35:04 CDT 2005

This is a multi-part message in MIME format.

------=_NextPart_000_010F_01C58A35.150731E0
Content-Type: text/plain;
charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable

Well given that money is an issue...
Why are you spending probably close to 5000 plus time for the upgrade? =
Yes that will probably put a nice penny in your pocket but not sure it =
will do anything for the client.


Why not up the ram in the current server? will that server take 3 GB =
of RAM??
Gigabit Switch (assuming internal wiring will support it) is really =
cheap from Dell right now (Dell PowerConnect 2724 24 port Managed =
Gigabit Switch $412.00)
Gigabit NIC in the Server (3Com Gigabit Card at Dell 112.00)
Gigabit NICs in the workstation (Linksys Gigabit Cards at Dell 23.00 =
each) 460.00 for 20 adapters.
So for Say a little over a $1000 (plus your time) you've eliminated 2 of =
the 3 bottlenecks you identified. Now the client has more money =
available for programmers to convert the program to SQL and you solve =
Number 3

There is a potential issue here with the single disk on the current =
server as everything is happening on that single disk. Setting up =
perfmon could help you determine if the disk simply over worked...raid 1 =
would not help there...raid 5 could if hardware based

--=20
Cris Hanna [SBS-MVP]
-------------------------------------------------
Please do not email me directly for assistance. Reply only in the =
Newsgroups for the benefit of everyone
"Doug B -db-" <dburrows_nospam-@dbnational.com> wrote in message =
news:i6fCe.155246$PR6.134346@tornado.texas.rr.com...
Looking for feedback: If you were going to make the best improvement =
and do=20
it in steps what would you recommend to your client?



My client has a Custom Access application. Version 1 of the program =
ran=20
good. Version 2 runs much slower which I assume is because it handles =
more=20
information/tables. We are looking at trying to improve the =
performance with=20
either hardware upgrades and/or converting the access app. to SQL =
based. My=20
client's budget is limited due to the amount he's soaked into the=20
development of other things.

From what I can see there are 3 possible bottlenecks. Server, Network =
speed=20
and the application.



My client Current network: 20 users; (12 max using the access app.)

SBS 2000, Dual Xeon P3 866. SCSI 160, 1.5 gb ram, 100mb Ethernet.



Considering Dell server w/raid 1 & 5. 2gb ram, dual Xeon 3.0 (2mb =
cache).=20
Gigabit network. SBS 2003.



My personal thoughts are that converting the Database program to a SQL =
based=20
app would show and best increase but I'm looking for some opinions.



Thanks all.

db



------=_NextPart_000_010F_01C58A35.150731E0
Content-Type: text/html;
charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable

<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.0 Transitional//EN">
<HTML><HEAD>
<META http-equiv=3DContent-Type content=3D"text/html; =
charset=3Diso-8859-1">
<META content=3D"MSHTML 6.00.2900.2668" name=3DGENERATOR>
<STYLE></STYLE>
</HEAD>
<BODY bgColor=3D#ffffff>
<DIV><FONT face=3DArial size=3D2>Well given that money is an =
issue...</FONT></DIV>
<DIV><FONT face=3DArial size=3D2>Why are you spending probably close to =
5000 plus=20
time for the upgrade?&nbsp;&nbsp; Yes that will probably put a nice =
penny in=20
your pocket but not sure it will do anything for the =
client.</FONT></DIV>
<DIV><FONT face=3DArial size=3D2></FONT>&nbsp;</DIV>
<DIV><FONT face=3DArial size=3D2></FONT>&nbsp;</DIV>
<DIV><FONT face=3DArial size=3D2>Why not up the ram in the current=20
server?&nbsp;&nbsp; will that server take 3 GB of RAM??</FONT></DIV>
<DIV><FONT face=3DArial size=3D2>Gigabit Switch (assuming internal =
wiring will=20
support it) is really cheap from Dell right now&nbsp; (Dell PowerConnect =

2724&nbsp; 24 port Managed Gigabit Switch $412.00)</FONT></DIV>
<DIV><FONT face=3DArial size=3D2>Gigabit NIC in the Server&nbsp; (3Com =
Gigabit Card=20
at Dell&nbsp; 112.00)</FONT></DIV>
<DIV><FONT face=3DArial size=3D2>Gigabit NICs in the workstation =
(Linksys Gigabit=20
Cards at Dell&nbsp; 23.00 each)&nbsp; 460.00 for 20 =
adapters.</FONT></DIV>
<DIV><FONT face=3DArial size=3D2>So for Say a little over a $1000 (plus =
your time)=20
you've eliminated 2 of the 3 bottlenecks you identified.&nbsp;&nbsp; Now =
the=20
client has more money available for programmers to convert the program =
to SQL=20
and you solve Number 3</FONT></DIV>
<DIV><FONT face=3DArial size=3D2></FONT>&nbsp;</DIV>
<DIV><FONT face=3DArial size=3D2>There is a potential issue here with =
the single=20
disk on the current server as everything is happening on that single=20
disk.&nbsp;&nbsp; Setting up perfmon could help you determine if the =
disk simply=20
over worked...raid 1 would not help there...raid 5 could if hardware=20
based</FONT></DIV>
<DIV><BR>-- <BR>Cris Hanna=20
[SBS-MVP]<BR>-------------------------------------------------<BR>Please =
do not=20
email me directly for assistance.&nbsp; Reply only in the Newsgroups for =
the=20
benefit of everyone</DIV>
<BLOCKQUOTE=20
style=3D"PADDING-RIGHT: 0px; PADDING-LEFT: 5px; MARGIN-LEFT: 5px; =
BORDER-LEFT: #000000 2px solid; MARGIN-RIGHT: 0px">
<DIV>"Doug B -db-" &lt;<A=20
=
href=3D"mailto:dburrows_nospam-@dbnational.com">dburrows_nospam-@dbnation=
al.com</A>&gt;=20
wrote in message <A=20
=
href=3D"news:i6fCe.155246$PR6.134346@tornado.texas.rr.com">news:i6fCe.155=
246$PR6.134346@tornado.texas.rr.com</A>...</DIV>Looking=20
for feedback: If you were going to make the best improvement and do =
<BR>it in=20
steps what would you recommend to your client?<BR><BR><BR><BR>My =
client has a=20
Custom Access application. Version 1 of the program ran <BR>good. =
Version 2=20
runs much slower which I assume is because it handles more=20
<BR>information/tables. We are looking at trying to improve the =
performance=20
with <BR>either hardware upgrades and/or converting the access app. to =
SQL=20
based. My <BR>client's budget is limited due to the amount he's soaked =
into=20
the <BR>development of other things.<BR><BR>From what I can see there =
are 3=20
possible bottlenecks. Server, Network speed <BR>and the=20
application.<BR><BR><BR><BR>My client Current network: 20 users; (12 =
max using=20
the access app.)<BR><BR>SBS 2000, Dual Xeon P3 866. SCSI 160, 1.5 gb =
ram,=20
100mb Ethernet.<BR><BR><BR><BR>Considering Dell server w/raid 1 &amp; =
5. 2gb=20
ram, dual Xeon 3.0 (2mb cache). <BR>Gigabit network. SBS=20
2003.<BR><BR><BR><BR>My personal thoughts are that converting the =
Database=20
program to a SQL based <BR>app would show and best increase but I'm =
looking=20
for some opinions.<BR><BR><BR><BR>Thanks=20
all.<BR><BR>db<BR><BR><BR></BLOCKQUOTE></BODY></HTML>

------=_NextPart_000_010F_01C58A35.150731E0--


Re: Looking for opinions by Merv

Merv
Sat Jul 16 19:29:03 CDT 2005

What would it take ($$) to convert the app to SQL? I think there's a good
chance that more of the problem is within Access and not your current
hardware. Is Version 2 of the App slow for just one or two users or is
there a threshold of the number of users that needs to kick in before the
app slows down?

Do you have a fast workstation with 1 MB or so of RAM in it that you could
put the current Version 2 on and then share it out to see if a faster
processor will help?

--
Merv Porter [SBS MVP]
===================================
"Doug B -db-" <dburrows_nospam-@dbnational.com> wrote in message
news:i6fCe.155246$PR6.134346@tornado.texas.rr.com...
> Looking for feedback: If you were going to make the best improvement and
> do
> it in steps what would you recommend to your client?
>
>
>
> My client has a Custom Access application. Version 1 of the program ran
> good. Version 2 runs much slower which I assume is because it handles more
> information/tables. We are looking at trying to improve the performance
> with
> either hardware upgrades and/or converting the access app. to SQL based.
> My
> client's budget is limited due to the amount he's soaked into the
> development of other things.
>
> From what I can see there are 3 possible bottlenecks. Server, Network
> speed
> and the application.
>
>
>
> My client Current network: 20 users; (12 max using the access app.)
>
> SBS 2000, Dual Xeon P3 866. SCSI 160, 1.5 gb ram, 100mb Ethernet.
>
>
>
> Considering Dell server w/raid 1 & 5. 2gb ram, dual Xeon 3.0 (2mb cache).
> Gigabit network. SBS 2003.
>
>
>
> My personal thoughts are that converting the Database program to a SQL
> based
> app would show and best increase but I'm looking for some opinions.
>
>
>
> Thanks all.
>
> db
>
>
>



Re: Looking for opinions by Doug

Doug
Sun Jul 17 03:24:53 CDT 2005

For a short time I moved the Ver. 2 App. to a Athlon64 3200 (running WinXP64
beta) with 1gb ram. They saw some improvement in the performance but the
share was limited to 10 users. Out of the 12 users who use it there are 5 or
6 who enter data while the others lookup. I haven't got any feedback that it
slows down when more users are on. I know that when the users are looking up
a record it can take at least :30 secs to fill the form.

Thanks for the feedback, keep it coming,
db
==============================
"Merv Porter [SBS-MVP]" <mwport@no_spam_hotmail.com> wrote in message
news:u1XXKamiFHA.3936@TK2MSFTNGP10.phx.gbl...
What would it take ($$) to convert the app to SQL? I think there's a good
chance that more of the problem is within Access and not your current
hardware. Is Version 2 of the App slow for just one or two users or is
there a threshold of the number of users that needs to kick in before the
app slows down?

Do you have a fast workstation with 1 MB or so of RAM in it that you could
put the current Version 2 on and then share it out to see if a faster
processor will help?

--
Merv Porter [SBS MVP]
===================================
"Doug B -db-" <dburrows_nospam-@dbnational.com> wrote in message
news:i6fCe.155246$PR6.134346@tornado.texas.rr.com...
> Looking for feedback: If you were going to make the best improvement and
> do
> it in steps what would you recommend to your client?
>
>
>
> My client has a Custom Access application. Version 1 of the program ran
> good. Version 2 runs much slower which I assume is because it handles more
> information/tables. We are looking at trying to improve the performance
> with
> either hardware upgrades and/or converting the access app. to SQL based.
> My
> client's budget is limited due to the amount he's soaked into the
> development of other things.
>
> From what I can see there are 3 possible bottlenecks. Server, Network
> speed
> and the application.
>
>
>
> My client Current network: 20 users; (12 max using the access app.)
>
> SBS 2000, Dual Xeon P3 866. SCSI 160, 1.5 gb ram, 100mb Ethernet.
>
>
>
> Considering Dell server w/raid 1 & 5. 2gb ram, dual Xeon 3.0 (2mb cache).
> Gigabit network. SBS 2003.
>
>
>
> My personal thoughts are that converting the Database program to a SQL
> based
> app would show and best increase but I'm looking for some opinions.
>
>
>
> Thanks all.
>
> db
>
>
>




Re: Looking for opinions by Doug

Doug
Sun Jul 17 03:49:06 CDT 2005

Thank you so much for your feedback,

There are 3 drives. Single OS drive and 2 raid 1 drives for data. I think I
would tend to agree that incorporating Gigabit speeds could help with the
data flow. What is your opinion on the performance increase obtained from
converting to SQL?
I have had an independent company setup a perfmon and give feedback, They
said "You can view the critical graphs on sheets Data-Processor1,
Data-Processor2. The server processors are having a hard time keeping up
with the demand." Personally I don't see that the processors are struggling
from what I see but I'm no "expert". They have recommended installing a
hardware firewall to turn off the windows firewall to save resources.

Thank you
db

===================================
"Cris Hanna [SBS-MVP]" <crisnospamhanna@computingnospampossibilities.net>
wrote in message news:eeBH$7liFHA.3568@tk2msftngp13.phx.gbl...
Well given that money is an issue...
Why are you spending probably close to 5000 plus time for the upgrade? Yes
that will probably put a nice penny in your pocket but not sure it will do
anything for the client.


Why not up the ram in the current server? will that server take 3 GB of
RAM??
Gigabit Switch (assuming internal wiring will support it) is really cheap
from Dell right now (Dell PowerConnect 2724 24 port Managed Gigabit Switch
$412.00)
Gigabit NIC in the Server (3Com Gigabit Card at Dell 112.00)
Gigabit NICs in the workstation (Linksys Gigabit Cards at Dell 23.00 each)
460.00 for 20 adapters.
So for Say a little over a $1000 (plus your time) you've eliminated 2 of the
3 bottlenecks you identified. Now the client has more money available for
programmers to convert the program to SQL and you solve Number 3

There is a potential issue here with the single disk on the current server
as everything is happening on that single disk. Setting up perfmon could
help you determine if the disk simply over worked...raid 1 would not help
there...raid 5 could if hardware based

--
Cris Hanna [SBS-MVP]
-------------------------------------------------
Please do not email me directly for assistance. Reply only in the
Newsgroups for the benefit of everyone
"Doug B -db-" <dburrows_nospam-@dbnational.com> wrote in message
news:i6fCe.155246$PR6.134346@tornado.texas.rr.com...
Looking for feedback: If you were going to make the best improvement and do
it in steps what would you recommend to your client?



My client has a Custom Access application. Version 1 of the program ran
good. Version 2 runs much slower which I assume is because it handles more
information/tables. We are looking at trying to improve the performance with
either hardware upgrades and/or converting the access app. to SQL based. My
client's budget is limited due to the amount he's soaked into the
development of other things.

From what I can see there are 3 possible bottlenecks. Server, Network speed
and the application.



My client Current network: 20 users; (12 max using the access app.)

SBS 2000, Dual Xeon P3 866. SCSI 160, 1.5 gb ram, 100mb Ethernet.



Considering Dell server w/raid 1 & 5. 2gb ram, dual Xeon 3.0 (2mb cache).
Gigabit network. SBS 2003.



My personal thoughts are that converting the Database program to a SQL based
app would show and best increase but I'm looking for some opinions.



Thanks all.

db



Re: Looking for opinions by Cris

Cris
Sun Jul 17 04:14:28 CDT 2005

This is a multi-part message in MIME format.

------=_NextPart_000_02F9_01C58A86.0667E4D0
Content-Type: text/plain;
charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable

typically...
A custom Access DB is not going to be dual processor aware, based on my =
experience

And an Access DB on the server is not a true Client Server product...all =
the processing is on the workstations

What is the OS and specs of the machines accessing the DB?

There are numerous articles out there which will compare SQL to Access

I'd really question the perfmon feedback. If you accepted their =
report...did they get additional work??


--=20
Cris Hanna [SBS-MVP]
-------------------------------------------------
Please do not email me directly for assistance. Reply only in the =
Newsgroups for the benefit of everyone
"Doug B -db-" <dburrows_nospam-@dbnational.com> wrote in message =
news:6YoCe.176309$6g3.70666@tornado.texas.rr.com...
Thank you so much for your feedback,

There are 3 drives. Single OS drive and 2 raid 1 drives for data. I =
think I=20
would tend to agree that incorporating Gigabit speeds could help with =
the=20
data flow. What is your opinion on the performance increase obtained =
from=20
converting to SQL?
I have had an independent company setup a perfmon and give feedback, =
They=20
said "You can view the critical graphs on sheets Data-Processor1,=20
Data-Processor2. The server processors are having a hard time keeping =
up=20
with the demand." Personally I don't see that the processors are =
struggling=20
from what I see but I'm no "expert". They have recommended installing =
a=20
hardware firewall to turn off the windows firewall to save resources.

Thank you
db

=
=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=
=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D
"Cris Hanna [SBS-MVP]" =
<crisnospamhanna@computingnospampossibilities.net>=20
wrote in message news:eeBH$7liFHA.3568@tk2msftngp13.phx.gbl...
Well given that money is an issue...
Why are you spending probably close to 5000 plus time for the upgrade? =
Yes=20
that will probably put a nice penny in your pocket but not sure it =
will do=20
anything for the client.


Why not up the ram in the current server? will that server take 3 GB =
of=20
RAM??
Gigabit Switch (assuming internal wiring will support it) is really =
cheap=20
from Dell right now (Dell PowerConnect 2724 24 port Managed Gigabit =
Switch=20
$412.00)
Gigabit NIC in the Server (3Com Gigabit Card at Dell 112.00)
Gigabit NICs in the workstation (Linksys Gigabit Cards at Dell 23.00 =
each)=20
460.00 for 20 adapters.
So for Say a little over a $1000 (plus your time) you've eliminated 2 =
of the=20
3 bottlenecks you identified. Now the client has more money =
available for=20
programmers to convert the program to SQL and you solve Number 3

There is a potential issue here with the single disk on the current =
server=20
as everything is happening on that single disk. Setting up perfmon =
could=20
help you determine if the disk simply over worked...raid 1 would not =
help=20
there...raid 5 could if hardware based

--=20
Cris Hanna [SBS-MVP]
-------------------------------------------------
Please do not email me directly for assistance. Reply only in the=20
Newsgroups for the benefit of everyone
"Doug B -db-" <dburrows_nospam-@dbnational.com> wrote in message=20
news:i6fCe.155246$PR6.134346@tornado.texas.rr.com...
Looking for feedback: If you were going to make the best improvement =
and do
it in steps what would you recommend to your client?



My client has a Custom Access application. Version 1 of the program =
ran
good. Version 2 runs much slower which I assume is because it handles =
more
information/tables. We are looking at trying to improve the =
performance with
either hardware upgrades and/or converting the access app. to SQL =
based. My
client's budget is limited due to the amount he's soaked into the
development of other things.

From what I can see there are 3 possible bottlenecks. Server, Network =
speed
and the application.



My client Current network: 20 users; (12 max using the access app.)

SBS 2000, Dual Xeon P3 866. SCSI 160, 1.5 gb ram, 100mb Ethernet.



Considering Dell server w/raid 1 & 5. 2gb ram, dual Xeon 3.0 (2mb =
cache).
Gigabit network. SBS 2003.



My personal thoughts are that converting the Database program to a SQL =
based
app would show and best increase but I'm looking for some opinions.



Thanks all.

db=20


------=_NextPart_000_02F9_01C58A86.0667E4D0
Content-Type: text/html;
charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable

<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.0 Transitional//EN">
<HTML><HEAD>
<META http-equiv=3DContent-Type content=3D"text/html; =
charset=3Diso-8859-1">
<META content=3D"MSHTML 6.00.2900.2668" name=3DGENERATOR>
<STYLE></STYLE>
</HEAD>
<BODY bgColor=3D#ffffff>
<DIV><FONT face=3DArial size=3D2>typically...</FONT></DIV>
<DIV><FONT face=3DArial size=3D2>A custom Access DB is not going =
to&nbsp;be dual=20
processor aware, based on my experience</FONT></DIV>
<DIV><FONT face=3DArial size=3D2></FONT>&nbsp;</DIV>
<DIV><FONT face=3DArial size=3D2>And an Access DB on the server is not a =
true Client=20
Server product...all the processing is on the workstations</FONT></DIV>
<DIV><FONT face=3DArial size=3D2></FONT>&nbsp;</DIV>
<DIV><FONT face=3DArial size=3D2>What is the OS and specs of the =
machines accessing=20
the DB?</FONT></DIV>
<DIV><FONT face=3DArial size=3D2></FONT>&nbsp;</DIV>
<DIV><FONT face=3DArial size=3D2>There are numerous articles out there =
which will=20
compare SQL to Access</FONT></DIV>
<DIV><FONT face=3DArial size=3D2></FONT>&nbsp;</DIV>
<DIV><FONT face=3DArial size=3D2>I'd really question the perfmon=20
feedback.&nbsp;&nbsp; If you accepted their report...did they get =
additional=20
work??</FONT></DIV>
<DIV><FONT face=3DArial size=3D2></FONT>&nbsp;</DIV>
<DIV><BR>-- <BR>Cris Hanna=20
[SBS-MVP]<BR>-------------------------------------------------<BR>Please =
do not=20
email me directly for assistance.&nbsp; Reply only in the Newsgroups for =
the=20
benefit of everyone</DIV>
<BLOCKQUOTE=20
style=3D"PADDING-RIGHT: 0px; PADDING-LEFT: 5px; MARGIN-LEFT: 5px; =
BORDER-LEFT: #000000 2px solid; MARGIN-RIGHT: 0px">
<DIV>"Doug B -db-" &lt;<A=20
=
href=3D"mailto:dburrows_nospam-@dbnational.com">dburrows_nospam-@dbnation=
al.com</A>&gt;=20
wrote in message <A=20
=
href=3D"news:6YoCe.176309$6g3.70666@tornado.texas.rr.com">news:6YoCe.1763=
09$6g3.70666@tornado.texas.rr.com</A>...</DIV>Thank=20
you so much for your feedback,<BR><BR>There are 3 drives. Single OS =
drive and=20
2 raid 1 drives for data. I think I <BR>would tend to agree that =
incorporating=20
Gigabit speeds could help with the <BR>data flow. What is your opinion =
on the=20
performance increase obtained from <BR>converting to SQL?<BR>I have =
had an=20
independent company setup a perfmon and give feedback, They <BR>said =
"You can=20
view the critical graphs on sheets Data-Processor1, =
<BR>Data-Processor2.&nbsp;=20
The server processors are having a hard time keeping up <BR>with the=20
demand."&nbsp; Personally I don't see that the processors are =
struggling=20
<BR>from what I see but I'm no "expert". They have recommended =
installing a=20
<BR>hardware firewall to turn off the windows firewall to save=20
resources.<BR><BR>Thank=20
=
you<BR>db<BR><BR>=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=
=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D<BR>"Cris Hanna =
[SBS-MVP]"=20
&lt;<A=20
=
href=3D"mailto:crisnospamhanna@computingnospampossibilities.net">crisnosp=
amhanna@computingnospampossibilities.net</A>&gt;=20
<BR>wrote in message <A=20
=
href=3D"news:eeBH$7liFHA.3568@tk2msftngp13.phx.gbl">news:eeBH$7liFHA.3568=
@tk2msftngp13.phx.gbl</A>...<BR>Well=20
given that money is an issue...<BR>Why are you spending probably close =
to 5000=20
plus time for the upgrade?&nbsp;&nbsp; Yes <BR>that will probably put =
a nice=20
penny in your pocket but not sure it will do <BR>anything for the=20
client.<BR><BR><BR>Why not up the ram in the current =
server?&nbsp;&nbsp; will=20
that server take 3 GB of <BR>RAM??<BR>Gigabit Switch (assuming =
internal wiring=20
will support it) is really cheap <BR>from Dell right now&nbsp; (Dell=20
PowerConnect 2724&nbsp; 24 port Managed Gigabit Switch =
<BR>$412.00)<BR>Gigabit=20
NIC in the Server&nbsp; (3Com Gigabit Card at Dell&nbsp; =
112.00)<BR>Gigabit=20
NICs in the workstation (Linksys Gigabit Cards at Dell&nbsp; 23.00 =
each)=20
<BR>460.00 for 20 adapters.<BR>So for Say a little over a $1000 (plus =
your=20
time) you've eliminated 2 of the <BR>3 bottlenecks you =
identified.&nbsp;&nbsp;=20
Now the client has more money available for <BR>programmers to convert =
the=20
program to SQL and you solve Number 3<BR><BR>There is a potential =
issue here=20
with the single disk on the current server <BR>as everything is =
happening on=20
that single disk.&nbsp;&nbsp; Setting up perfmon could <BR>help you =
determine=20
if the disk simply over worked...raid 1 would not help =
<BR>there...raid 5=20
could if hardware based<BR><BR>-- <BR>Cris Hanna=20
=
[SBS-MVP]<BR>-------------------------------------------------<BR>Please =
do=20
not email me directly for assistance.&nbsp; Reply only in the =
<BR>Newsgroups=20
for the benefit of everyone<BR>"Doug B -db-" &lt;<A=20
=
href=3D"mailto:dburrows_nospam-@dbnational.com">dburrows_nospam-@dbnation=
al.com</A>&gt;=20
wrote in message <BR><A=20
=
href=3D"news:i6fCe.155246$PR6.134346@tornado.texas.rr.com">news:i6fCe.155=
246$PR6.134346@tornado.texas.rr.com</A>...<BR>Looking=20
for feedback: If you were going to make the best improvement and =
do<BR>it in=20
steps what would you recommend to your client?<BR><BR><BR><BR>My =
client has a=20
Custom Access application. Version 1 of the program ran<BR>good. =
Version 2=20
runs much slower which I assume is because it handles=20
more<BR>information/tables. We are looking at trying to improve the=20
performance with<BR>either hardware upgrades and/or converting the =
access app.=20
to SQL based. My<BR>client's budget is limited due to the amount he's =
soaked=20
into the<BR>development of other things.<BR><BR>From what I can see =
there are=20
3 possible bottlenecks. Server, Network speed<BR>and the=20
application.<BR><BR><BR><BR>My client Current network: 20 users; (12 =
max using=20
the access app.)<BR><BR>SBS 2000, Dual Xeon P3 866. SCSI 160, 1.5 gb =
ram,=20
100mb Ethernet.<BR><BR><BR><BR>Considering Dell server w/raid 1 &amp; =
5. 2gb=20
ram, dual Xeon 3.0 (2mb cache).<BR>Gigabit network. SBS=20
2003.<BR><BR><BR><BR>My personal thoughts are that converting the =
Database=20
program to a SQL based<BR>app would show and best increase but I'm =
looking for=20
some opinions.<BR><BR><BR><BR>Thanks all.<BR><BR>db=20
<BR><BR></BLOCKQUOTE></BODY></HTML>

------=_NextPart_000_02F9_01C58A86.0667E4D0--


Re: Looking for opinions by Doug

Doug
Mon Jul 18 03:54:06 CDT 2005

This is a multi-part message in MIME format.

------=_NextPart_000_0047_01C58B4C.64CC1980
Content-Type: text/plain;
charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable

I question it too. I have the report results on a .csv file if you'd =
like to see it.

Most workstations at P4, XP w/512 mb. Some of them had been 256 mb and =
upgrading them to 512 did help on the workstation performance but not =
much improvement on acquiring the data/query results.

After their analysis they came back with a $28,000 proposal but there is =
no way they will get any of that work. Especially since many aspects of =
it aren't even necessary.

I really appreciate your comments Cris. Thank you for your time.

db

=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=
=3D=3D=3D=3D=3D=3D=3D=3D
"Cris Hanna [SBS-MVP]" =
<crisnospamhanna@computingnospampossibilities.net> wrote in message =
news:%236sew$qiFHA.3960@TK2MSFTNGP12.phx.gbl...
typically...
A custom Access DB is not going to be dual processor aware, based on =
my experience

And an Access DB on the server is not a true Client Server =
product...all the processing is on the workstations

What is the OS and specs of the machines accessing the DB?

There are numerous articles out there which will compare SQL to Access

I'd really question the perfmon feedback. If you accepted their =
report...did they get additional work??


--=20
Cris Hanna [SBS-MVP]
-------------------------------------------------
Please do not email me directly for assistance. Reply only in the =
Newsgroups for the benefit of everyone
"Doug B -db-" <dburrows_nospam-@dbnational.com> wrote in message =
news:6YoCe.176309$6g3.70666@tornado.texas.rr.com...
Thank you so much for your feedback,

There are 3 drives. Single OS drive and 2 raid 1 drives for data. I =
think I=20
would tend to agree that incorporating Gigabit speeds could help =
with the=20
data flow. What is your opinion on the performance increase obtained =
from=20
converting to SQL?
I have had an independent company setup a perfmon and give feedback, =
They=20
said "You can view the critical graphs on sheets Data-Processor1,=20
Data-Processor2. The server processors are having a hard time =
keeping up=20
with the demand." Personally I don't see that the processors are =
struggling=20
from what I see but I'm no "expert". They have recommended =
installing a=20
hardware firewall to turn off the windows firewall to save =
resources.

Thank you
db

=
=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=
=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D
"Cris Hanna [SBS-MVP]" =
<crisnospamhanna@computingnospampossibilities.net>=20
wrote in message news:eeBH$7liFHA.3568@tk2msftngp13.phx.gbl...
Well given that money is an issue...
Why are you spending probably close to 5000 plus time for the =
upgrade? Yes=20
that will probably put a nice penny in your pocket but not sure it =
will do=20
anything for the client.


Why not up the ram in the current server? will that server take 3 =
GB of=20
RAM??
Gigabit Switch (assuming internal wiring will support it) is really =
cheap=20
from Dell right now (Dell PowerConnect 2724 24 port Managed =
Gigabit Switch=20
$412.00)
Gigabit NIC in the Server (3Com Gigabit Card at Dell 112.00)
Gigabit NICs in the workstation (Linksys Gigabit Cards at Dell =
23.00 each)=20
460.00 for 20 adapters.
So for Say a little over a $1000 (plus your time) you've eliminated =
2 of the=20
3 bottlenecks you identified. Now the client has more money =
available for=20
programmers to convert the program to SQL and you solve Number 3

There is a potential issue here with the single disk on the current =
server=20
as everything is happening on that single disk. Setting up perfmon =
could=20
help you determine if the disk simply over worked...raid 1 would not =
help=20
there...raid 5 could if hardware based

--=20
Cris Hanna [SBS-MVP]
-------------------------------------------------
Please do not email me directly for assistance. Reply only in the=20
Newsgroups for the benefit of everyone
"Doug B -db-" <dburrows_nospam-@dbnational.com> wrote in message=20
news:i6fCe.155246$PR6.134346@tornado.texas.rr.com...
Looking for feedback: If you were going to make the best improvement =
and do
it in steps what would you recommend to your client?



My client has a Custom Access application. Version 1 of the program =
ran
good. Version 2 runs much slower which I assume is because it =
handles more
information/tables. We are looking at trying to improve the =
performance with
either hardware upgrades and/or converting the access app. to SQL =
based. My
client's budget is limited due to the amount he's soaked into the
development of other things.

From what I can see there are 3 possible bottlenecks. Server, =
Network speed
and the application.



My client Current network: 20 users; (12 max using the access app.)

SBS 2000, Dual Xeon P3 866. SCSI 160, 1.5 gb ram, 100mb Ethernet.



Considering Dell server w/raid 1 & 5. 2gb ram, dual Xeon 3.0 (2mb =
cache).
Gigabit network. SBS 2003.



My personal thoughts are that converting the Database program to a =
SQL based
app would show and best increase but I'm looking for some opinions.



Thanks all.

db=20


------=_NextPart_000_0047_01C58B4C.64CC1980
Content-Type: text/html;
charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable

<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.0 Transitional//EN">
<HTML><HEAD>
<META http-equiv=3DContent-Type content=3D"text/html; =
charset=3Diso-8859-1">
<META content=3D"MSHTML 6.00.2900.2668" name=3DGENERATOR>
<STYLE></STYLE>
</HEAD>
<BODY bgColor=3D#ffffff>
<DIV><FONT face=3DArial size=3D2>I question it too. I have the report =
results on a=20
.csv file if you'd like to see it.</FONT></DIV>
<DIV><FONT face=3DArial size=3D2></FONT>&nbsp;</DIV>
<DIV><FONT face=3DArial size=3D2>Most workstations at P4, XP w/512 mb. =
Some of them=20
had been 256 mb and upgrading them to 512 did help on the workstation=20
performance but not much improvement on acquiring the data/query=20
results.</FONT></DIV>
<DIV><FONT face=3DArial size=3D2></FONT>&nbsp;</DIV>
<DIV><FONT face=3DArial size=3D2>After their analysis they came back =
with a $28,000=20
proposal but there is no way they will get any of that work. Especially =
since=20
many aspects of it aren't even necessary.</FONT></DIV>
<DIV><FONT face=3DArial size=3D2></FONT>&nbsp;</DIV>
<DIV><FONT face=3DArial size=3D2>I really appreciate your comments Cris. =
Thank you=20
for your time.</FONT></DIV>
<DIV><FONT face=3DArial size=3D2></FONT>&nbsp;</DIV>
<DIV><FONT face=3DArial size=3D2>db</FONT></DIV>
<DIV><FONT face=3DArial size=3D2></FONT>&nbsp;</DIV>
<DIV><FONT face=3DArial =
size=3D2>=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=
=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D</FONT></DIV>
<BLOCKQUOTE dir=3Dltr=20
style=3D"PADDING-RIGHT: 0px; PADDING-LEFT: 5px; MARGIN-LEFT: 5px; =
BORDER-LEFT: #000000 2px solid; MARGIN-RIGHT: 0px">
<DIV>"Cris Hanna [SBS-MVP]" &lt;<A=20
=
href=3D"mailto:crisnospamhanna@computingnospampossibilities.net">crisnosp=
amhanna@computingnospampossibilities.net</A>&gt;=20
wrote in message <A=20
=
href=3D"news:%236sew$qiFHA.3960@TK2MSFTNGP12.phx.gbl">news:%236sew$qiFHA.=
3960@TK2MSFTNGP12.phx.gbl</A>...</DIV>
<DIV><FONT face=3DArial size=3D2>typically...</FONT></DIV>
<DIV><FONT face=3DArial size=3D2>A custom Access DB is not going =
to&nbsp;be dual=20
processor aware, based on my experience</FONT></DIV>
<DIV><FONT face=3DArial size=3D2></FONT>&nbsp;</DIV>
<DIV><FONT face=3DArial size=3D2>And an Access DB on the server is not =
a true=20
Client Server product...all the processing is on the =
workstations</FONT></DIV>
<DIV><FONT face=3DArial size=3D2></FONT>&nbsp;</DIV>
<DIV><FONT face=3DArial size=3D2>What is the OS and specs of the =
machines=20
accessing the DB?</FONT></DIV>
<DIV><FONT face=3DArial size=3D2></FONT>&nbsp;</DIV>
<DIV><FONT face=3DArial size=3D2>There are numerous articles out there =
which will=20
compare SQL to Access</FONT></DIV>
<DIV><FONT face=3DArial size=3D2></FONT>&nbsp;</DIV>
<DIV><FONT face=3DArial size=3D2>I'd really question the perfmon=20
feedback.&nbsp;&nbsp; If you accepted their report...did they get =
additional=20
work??</FONT></DIV>
<DIV><FONT face=3DArial size=3D2></FONT>&nbsp;</DIV>
<DIV><BR>-- <BR>Cris Hanna=20
=
[SBS-MVP]<BR>-------------------------------------------------<BR>Please =
do=20
not email me directly for assistance.&nbsp; Reply only in the =
Newsgroups for=20
the benefit of everyone</DIV>
<BLOCKQUOTE=20
style=3D"PADDING-RIGHT: 0px; PADDING-LEFT: 5px; MARGIN-LEFT: 5px; =
BORDER-LEFT: #000000 2px solid; MARGIN-RIGHT: 0px">
<DIV>"Doug B -db-" &lt;<A=20
=
href=3D"mailto:dburrows_nospam-@dbnational.com">dburrows_nospam-@dbnation=
al.com</A>&gt;=20
wrote in message <A=20
=
href=3D"news:6YoCe.176309$6g3.70666@tornado.texas.rr.com">news:6YoCe.1763=
09$6g3.70666@tornado.texas.rr.com</A>...</DIV>Thank=20
you so much for your feedback,<BR><BR>There are 3 drives. Single OS =
drive=20
and 2 raid 1 drives for data. I think I <BR>would tend to agree that =

incorporating Gigabit speeds could help with the <BR>data flow. What =
is your=20
opinion on the performance increase obtained from <BR>converting to=20
SQL?<BR>I have had an independent company setup a perfmon and give =
feedback,=20
They <BR>said "You can view the critical graphs on sheets =
Data-Processor1,=20
<BR>Data-Processor2.&nbsp; The server processors are having a hard =
time=20
keeping up <BR>with the demand."&nbsp; Personally I don't see that =
the=20
processors are struggling <BR>from what I see but I'm no "expert". =
They have=20
recommended installing a <BR>hardware firewall to turn off the =
windows=20
firewall to save resources.<BR><BR>Thank=20
=
you<BR>db<BR><BR>=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=
=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D<BR>"Cris Hanna=20
[SBS-MVP]" &lt;<A=20
=
href=3D"mailto:crisnospamhanna@computingnospampossibilities.net">crisnosp=
amhanna@computingnospampossibilities.net</A>&gt;=20
<BR>wrote in message <A=20
=
href=3D"news:eeBH$7liFHA.3568@tk2msftngp13.phx.gbl">news:eeBH$7liFHA.3568=
@tk2msftngp13.phx.gbl</A>...<BR>Well=20
given that money is an issue...<BR>Why are you spending probably =
close to=20
5000 plus time for the upgrade?&nbsp;&nbsp; Yes <BR>that will =
probably put a=20
nice penny in your pocket but not sure it will do <BR>anything for =
the=20
client.<BR><BR><BR>Why not up the ram in the current =
server?&nbsp;&nbsp;=20
will that server take 3 GB of <BR>RAM??<BR>Gigabit Switch (assuming =
internal=20
wiring will support it) is really cheap <BR>from Dell right =
now&nbsp; (Dell=20
PowerConnect 2724&nbsp; 24 port Managed Gigabit Switch=20
<BR>$412.00)<BR>Gigabit NIC in the Server&nbsp; (3Com Gigabit Card =
at=20
Dell&nbsp; 112.00)<BR>Gigabit NICs in the workstation (Linksys =
Gigabit Cards=20
at Dell&nbsp; 23.00 each) <BR>460.00 for 20 adapters.<BR>So for Say =
a little=20
over a $1000 (plus your time) you've eliminated 2 of the <BR>3 =
bottlenecks=20
you identified.&nbsp;&nbsp; Now the client has more money available =
for=20
<BR>programmers to convert the program to SQL and you solve Number=20
3<BR><BR>There is a potential issue here with the single disk on the =
current=20
server <BR>as everything is happening on that single =
disk.&nbsp;&nbsp;=20
Setting up perfmon could <BR>help you determine if the disk simply =
over=20
worked...raid 1 would not help <BR>there...raid 5 could if hardware=20
based<BR><BR>-- <BR>Cris Hanna=20
=
[SBS-MVP]<BR>-------------------------------------------------<BR>Please =
do=20
not email me directly for assistance.&nbsp; Reply only in the =
<BR>Newsgroups=20
for the benefit of everyone<BR>"Doug B -db-" &lt;<A=20
=
href=3D"mailto:dburrows_nospam-@dbnational.com">dburrows_nospam-@dbnation=
al.com</A>&gt;=20
wrote in message <BR><A=20
=
href=3D"news:i6fCe.155246$PR6.134346@tornado.texas.rr.com">news:i6fCe.155=
246$PR6.134346@tornado.texas.rr.com</A>...<BR>Looking=20
for feedback: If you were going to make the best improvement and =
do<BR>it in=20
steps what would you recommend to your client?<BR><BR><BR><BR>My =
client has=20
a Custom Access application. Version 1 of the program ran<BR>good. =
Version 2=20
runs much slower which I assume is because it handles=20
more<BR>information/tables. We are looking at trying to improve the=20
performance with<BR>either hardware upgrades and/or converting the =
access=20
app. to SQL based. My<BR>client's budget is limited due to the =
amount he's=20
soaked into the<BR>development of other things.<BR><BR>From what I =
can see=20
there are 3 possible bottlenecks. Server, Network speed<BR>and the=20
application.<BR><BR><BR><BR>My client Current network: 20 users; (12 =
max=20
using the access app.)<BR><BR>SBS 2000, Dual Xeon P3 866. SCSI 160, =
1.5 gb=20
ram, 100mb Ethernet.<BR><BR><BR><BR>Considering Dell server w/raid 1 =
&amp;=20
5. 2gb ram, dual Xeon 3.0 (2mb cache).<BR>Gigabit network. SBS=20
2003.<BR><BR><BR><BR>My personal thoughts are that converting the =
Database=20
program to a SQL based<BR>app would show and best increase but I'm =
looking=20
for some opinions.<BR><BR><BR><BR>Thanks all.<BR><BR>db=20
<BR><BR></BLOCKQUOTE></BLOCKQUOTE></BODY></HTML>

------=_NextPart_000_0047_01C58B4C.64CC1980--