473,434 Members | 1,431 Online
Bytes | Software Development & Data Engineering Community
Post Job

Home Posts Topics Members FAQ

Join Bytes to post your question to a community of 473,434 software developers and data experts.

Continuous writing/reading to MDB

I have an MDB that contains a single table. Checkweigher data is being
continuously written to this MDB from a dedicated workstation over a network
into the MDB on the server at the rate of about 120 records a minute. This
may increase in the future to about 200. The records are not large.

We have other workstations that run sessions on Terminal Server querying
this MDB, doing things like plotting the last hours data on a graph or
running other reports, or even just viewing raw data. No one else will be
writing to the MDB.

Can anyone see any potential problems.

I am concerned with the network connection being lost during the writing and
having a corrupted MDB. The issue of speed is another. We expect 70-80,000
records a day which means that there could be a substantial delay in
querying towards the end of the day. The MDB will be archived each day and
start empty the next day.

I don't see any locking issues.

Jeff
Nov 12 '05 #1
13 3080
When you are working with a file-server database, this is always a
consideration. Whether you should consider alternatives would depend on how
"mission-critical" this is -- that is, what would it cost your company for
it to be out of service while you Compact/Repair? What would it cost if you
couldn't successfully repair, and had to restore from a backup? How would
you "catch up" the updates from the backup to current?

You may tell management and they may decide it's not "all that big a deal"
and to go with the file-server (Jet) database; or you may hear a sharp
intake of breath from management. Have the comparison costs of full retail
SQL Server ready so they can see how many "hits" it would take to pay pack.
I have worked on a few databases that could have been done with file server
from a performance/user audience size point of view, but were done client
server because of their importance to the operation and the superior
reliability and recoverability of the server DB.

Larry Linson
Microsoft Access MVP

"Jeff Pritchard" <je************@asken.com.au> wrote in message
news:8H***************@news.optus.net.au...
I have an MDB that contains a single table. Checkweigher data is being
continuously written to this MDB from a dedicated workstation over a network into the MDB on the server at the rate of about 120 records a minute. This
may increase in the future to about 200. The records are not large.

We have other workstations that run sessions on Terminal Server querying
this MDB, doing things like plotting the last hours data on a graph or
running other reports, or even just viewing raw data. No one else will be
writing to the MDB.

Can anyone see any potential problems.

I am concerned with the network connection being lost during the writing and having a corrupted MDB. The issue of speed is another. We expect 70-80,000
records a day which means that there could be a substantial delay in
querying towards the end of the day. The MDB will be archived each day and
start empty the next day.

I don't see any locking issues.

Jeff

Nov 12 '05 #2
On Sat, 15 Nov 2003 01:54:12 GMT, "Jeff Pritchard"
<je************@asken.com.au> wrote:

A peak append rate of 3 - 5 records (or higher?) per second is rather
high. I would take an hour or so to write a simple proof-of-concept
app to make sure your system can handle that. Tick a timer every 200
msec and append a row. Then add 100,000 rows to that table and try
again.

I'm not that concerned about the SELECT queries. The power of indexing
should ensure it hardly matters whether you're querying 100 rows or
100,000. Except for some increased network traffic. You could even
have those guys hitting a separate database, which you populate every
so often with a data dump from the first database.

Access IMHO is not a 24/7 database. I would lean towards a SQL Server
database for this.

-Tom.

I have an MDB that contains a single table. Checkweigher data is being
continuously written to this MDB from a dedicated workstation over a network
into the MDB on the server at the rate of about 120 records a minute. This
may increase in the future to about 200. The records are not large.

We have other workstations that run sessions on Terminal Server querying
this MDB, doing things like plotting the last hours data on a graph or
running other reports, or even just viewing raw data. No one else will be
writing to the MDB.

Can anyone see any potential problems.

I am concerned with the network connection being lost during the writing and
having a corrupted MDB. The issue of speed is another. We expect 70-80,000
records a day which means that there could be a substantial delay in
querying towards the end of the day. The MDB will be archived each day and
start empty the next day.

I don't see any locking issues.

Jeff


Nov 12 '05 #3
Hi Larry

Given that the data is archived each day, there is only the possibility of
losing a days data. The data is for monitoring weights only. They have run
for years without this and their approach is that it is not mission
critical. While not ideal to lose a days data, it is not going to stop
production.

We would only compact when it is archived. Even then probably not necessary
as we would simply replace the MDB with an empty copy.

The thought of setting up MSDE on the workstation was raised but then they
can't legally access this from multiple workstations, I think.

Do you think there will be a problem with others querying while data is
being continually written?

Jeff
"Larry Linson" <bo*****@localhost.not> wrote in message
news:hp*****************@nwrddc02.gnilink.net...
When you are working with a file-server database, this is always a
consideration. Whether you should consider alternatives would depend on how "mission-critical" this is -- that is, what would it cost your company for
it to be out of service while you Compact/Repair? What would it cost if you couldn't successfully repair, and had to restore from a backup? How would
you "catch up" the updates from the backup to current?

You may tell management and they may decide it's not "all that big a deal"
and to go with the file-server (Jet) database; or you may hear a sharp
intake of breath from management. Have the comparison costs of full retail
SQL Server ready so they can see how many "hits" it would take to pay pack. I have worked on a few databases that could have been done with file server from a performance/user audience size point of view, but were done client
server because of their importance to the operation and the superior
reliability and recoverability of the server DB.

Larry Linson
Microsoft Access MVP

"Jeff Pritchard" <je************@asken.com.au> wrote in message
news:8H***************@news.optus.net.au...
I have an MDB that contains a single table. Checkweigher data is being
continuously written to this MDB from a dedicated workstation over a

network
into the MDB on the server at the rate of about 120 records a minute. This may increase in the future to about 200. The records are not large.

We have other workstations that run sessions on Terminal Server querying
this MDB, doing things like plotting the last hours data on a graph or
running other reports, or even just viewing raw data. No one else will be writing to the MDB.

Can anyone see any potential problems.

I am concerned with the network connection being lost during the writing

and
having a corrupted MDB. The issue of speed is another. We expect 70-80,000 records a day which means that there could be a substantial delay in
querying towards the end of the day. The MDB will be archived each day and start empty the next day.

I don't see any locking issues.

Jeff


Nov 12 '05 #4
On Sat, 15 Nov 2003 05:13:49 GMT, "Jeff Pritchard"
<je************@asken.com.au> wrote:

You may want to read up on MSDE. It is a server database, not a client
one. It can be accessed by multiple users at the same time. It is
tuned for 5 concurrent users - above that number it will put requests
in a queue. You can likely get away with 1 writer and 10 - 15
occasional readers without too much of a problem.

-Tom.

Hi Larry

Given that the data is archived each day, there is only the possibility of
losing a days data. The data is for monitoring weights only. They have run
for years without this and their approach is that it is not mission
critical. While not ideal to lose a days data, it is not going to stop
production.

We would only compact when it is archived. Even then probably not necessary
as we would simply replace the MDB with an empty copy.

The thought of setting up MSDE on the workstation was raised but then they
can't legally access this from multiple workstations, I think.

Do you think there will be a problem with others querying while data is
being continually written?

Jeff

<clip>

Nov 12 '05 #5
Hi Tom

Yes I realise MSDE is multi-user, just wasn't sure about the number of users
you could legally have accessing it. Isn't there a legal limit?

The practical limit is another issue. The 10-15 readers is probably okay at
the moment. Definitely would only be one writer.

Jeff

"Tom van Stiphout" <to*****@no.spam.cox.net> wrote in message
news:pe********************************@4ax.com...
On Sat, 15 Nov 2003 05:13:49 GMT, "Jeff Pritchard"
<je************@asken.com.au> wrote:

You may want to read up on MSDE. It is a server database, not a client
one. It can be accessed by multiple users at the same time. It is
tuned for 5 concurrent users - above that number it will put requests
in a queue. You can likely get away with 1 writer and 10 - 15
occasional readers without too much of a problem.

-Tom.

Hi Larry

Given that the data is archived each day, there is only the possibility oflosing a days data. The data is for monitoring weights only. They have runfor years without this and their approach is that it is not mission
critical. While not ideal to lose a days data, it is not going to stop
production.

We would only compact when it is archived. Even then probably not necessaryas we would simply replace the MDB with an empty copy.

The thought of setting up MSDE on the workstation was raised but then theycan't legally access this from multiple workstations, I think.

Do you think there will be a problem with others querying while data is
being continually written?

Jeff

<clip>

Nov 12 '05 #6
Good suggestion. I will set up a database on my peer-to-peer and test it at
different rates.

I doubt they would ever get higher that 3, 4 at most. That is probably at
least a couple of years away as their manufacturing equip can't do that at
the moment. If that happens we can revisit the problem and maybe make some
changes then.

Jeff

"Tom van Stiphout" <to*****@no.spam.cox.net> wrote in message
news:pv********************************@4ax.com...
On Sat, 15 Nov 2003 01:54:12 GMT, "Jeff Pritchard"
<je************@asken.com.au> wrote:

A peak append rate of 3 - 5 records (or higher?) per second is rather
high. I would take an hour or so to write a simple proof-of-concept
app to make sure your system can handle that. Tick a timer every 200
msec and append a row. Then add 100,000 rows to that table and try
again.

I'm not that concerned about the SELECT queries. The power of indexing
should ensure it hardly matters whether you're querying 100 rows or
100,000. Except for some increased network traffic. You could even
have those guys hitting a separate database, which you populate every
so often with a data dump from the first database.

Access IMHO is not a 24/7 database. I would lean towards a SQL Server
database for this.

-Tom.

I have an MDB that contains a single table. Checkweigher data is being
continuously written to this MDB from a dedicated workstation over a networkinto the MDB on the server at the rate of about 120 records a minute. Thismay increase in the future to about 200. The records are not large.

We have other workstations that run sessions on Terminal Server querying
this MDB, doing things like plotting the last hours data on a graph or
running other reports, or even just viewing raw data. No one else will be
writing to the MDB.

Can anyone see any potential problems.

I am concerned with the network connection being lost during the writing andhaving a corrupted MDB. The issue of speed is another. We expect 70-80,000records a day which means that there could be a substantial delay in
querying towards the end of the day. The MDB will be archived each day andstart empty the next day.

I don't see any locking issues.

Jeff

Nov 12 '05 #7
"Jeff Pritchard" wrote
Yes I realise MSDE is multi-user, just
wasn't sure about the number of users
you could legally have accessing it. Isn't
there a legal limit?
No, there is not. It enforces its intended use by inserting delays on each
(internal) "batch update process" above 5. If you add 'way too many users,
then performance will deteriorate to the point that you will give up in
disgust and not add any more. In earlier versions, it would only _perform_ 5
of those operations at a time, and under those conditions, I have seen
reliable reports of 25 concurrent users. I haven't seen similar reports for
the current version with its modified approach to user limits.
The practical limit is another issue. The
10-15 readers is probably okay at
the moment. Definitely would only be
one writer.


Don't take Tom's numbers as any kind of "absolute limit" -- performance will
depend on many factors, including the environment and the requirements,
design, and implementation. With experience, I am sure you can tweak better
performance for more users than you will get initially.

On the other hand, I am quite comfortable that Tom is correct that, in your
planned environment, MSDE would likely be more than adequate. And, if you
grow to the point that it is not, it is easy to replace it with Microsoft
SQL Server.

Larry Linson
Microsoft Access MVP


Nov 12 '05 #8
je************@asken.com.au (Jeff Pritchard) wrote in
<8H***************@news.optus.net.au>:
I have an MDB that contains a single table. Checkweigher data is
being continuously written to this MDB from a dedicated
workstation over a network into the MDB on the server at the rate
of about 120 records a minute. This may increase in the future to
about 200. The records are not large.

We have other workstations that run sessions on Terminal Server
querying this MDB, doing things like plotting the last hours data
on a graph or running other reports, or even just viewing raw
data. No one else will be writing to the MDB.

Can anyone see any potential problems.

I am concerned with the network connection being lost during the
writing and having a corrupted MDB. The issue of speed is another.
We expect 70-80,000 records a day which means that there could be
a substantial delay in querying towards the end of the day. The
MDB will be archived each day and start empty the next day.


Well, you could eliminate the network connection problem by putting
the data file on the same machine where the writes are being
initiated, then let the WTS users connect to it from the terminal
server. Since they are read-only, loss of network connection
wouldn't be a problem.

If you can't give network access to this workstation, then perhaps
replication would work, with a replica on your WTS machine,
synchronized on a schedule. Unfortunately, Replication Manager
can't synch more often than every 15 minutes. I don't know if
Michael Kaplan's TSI tool could be used to initiate a
synchronization more frequently.

--
David W. Fenton http://www.bway.net/~dfenton
dfenton at bway dot net http://www.bway.net/~dfassoc
Nov 12 '05 #9
Putting the MDB on the workstation may be a go. I will keep this in mind as
we progress.

Jeff

"David W. Fenton" <dX********@bway.net.invalid> wrote in message
news:94***************************@24.168.128.86.. .
je************@asken.com.au (Jeff Pritchard) wrote in
<8H***************@news.optus.net.au>:
I have an MDB that contains a single table. Checkweigher data is
being continuously written to this MDB from a dedicated
workstation over a network into the MDB on the server at the rate
of about 120 records a minute. This may increase in the future to
about 200. The records are not large.

We have other workstations that run sessions on Terminal Server
querying this MDB, doing things like plotting the last hours data
on a graph or running other reports, or even just viewing raw
data. No one else will be writing to the MDB.

Can anyone see any potential problems.

I am concerned with the network connection being lost during the
writing and having a corrupted MDB. The issue of speed is another.
We expect 70-80,000 records a day which means that there could be
a substantial delay in querying towards the end of the day. The
MDB will be archived each day and start empty the next day.


Well, you could eliminate the network connection problem by putting
the data file on the same machine where the writes are being
initiated, then let the WTS users connect to it from the terminal
server. Since they are read-only, loss of network connection
wouldn't be a problem.

If you can't give network access to this workstation, then perhaps
replication would work, with a replica on your WTS machine,
synchronized on a schedule. Unfortunately, Replication Manager
can't synch more often than every 15 minutes. I don't know if
Michael Kaplan's TSI tool could be used to initiate a
synchronization more frequently.

--
David W. Fenton http://www.bway.net/~dfenton
dfenton at bway dot net http://www.bway.net/~dfassoc

Nov 12 '05 #10
Okay. Thanks for your comments. I am better armed for the battle now.

MSDE is a definite option if the MDB solution has a problem. I think we will
start with the MDB on the server (clients preference). If problems occur
moving it to the workstation may be the go, then MSDE.

Jeff

"Larry Linson" <bo*****@localhost.not> wrote in message
news:Si******************@nwrddc02.gnilink.net...
"Jeff Pritchard" wrote
> Yes I realise MSDE is multi-user, just
> wasn't sure about the number of users
> you could legally have accessing it. Isn't
> there a legal limit?
No, there is not. It enforces its intended use by inserting delays on each
(internal) "batch update process" above 5. If you add 'way too many users,
then performance will deteriorate to the point that you will give up in
disgust and not add any more. In earlier versions, it would only _perform_

5 of those operations at a time, and under those conditions, I have seen
reliable reports of 25 concurrent users. I haven't seen similar reports for the current version with its modified approach to user limits.
> The practical limit is another issue. The
> 10-15 readers is probably okay at
> the moment. Definitely would only be
> one writer.
Don't take Tom's numbers as any kind of "absolute limit" -- performance

will depend on many factors, including the environment and the requirements,
design, and implementation. With experience, I am sure you can tweak better performance for more users than you will get initially.

On the other hand, I am quite comfortable that Tom is correct that, in your planned environment, MSDE would likely be more than adequate. And, if you
grow to the point that it is not, it is easy to replace it with Microsoft
SQL Server.

Larry Linson
Microsoft Access MVP

Nov 12 '05 #11
"Jeff Pritchard" wrote
MSDE is a definite option if the MDB
solution has a problem. I think we will
start with the MDB on the server (clients
preference). If problems occur moving it to the work-
station may be the go, then MSDE.


I'd suggest a separate copy for each user, of the database used to query the
data, linking the tables in the server-side MDB. That's true even with
Terminal Server, each user should have his/her own copy of the front-end.

Larry Linson
Microsoft Access MVP
Nov 12 '05 #12
They already do.

Each user has a separate copy of the front-end MDE. The actual database is 2
different front-end MDE files (different functionality for different areas),
separate MDEs for each user, linking to 3 different back-end MDB files.
Back-end files contain separate sets of data that don't have any
relationship to each other, or where there is no complication with
relationships (cascades etc).

The whole system works beautifully. This checkweigher stuff is a new
requirement. There will be a new back-end MDB for the checkweigher data.

Thanks for your comments Larry.

Jeff

"Larry Linson" <bo*****@localhost.not> wrote in message
news:rx******************@nwrddc03.gnilink.net...
"Jeff Pritchard" wrote
> MSDE is a definite option if the MDB
> solution has a problem. I think we will
> start with the MDB on the server (clients
> preference).
> If problems occur moving it to the work-
> station may be the go, then MSDE.


I'd suggest a separate copy for each user, of the database used to query

the data, linking the tables in the server-side MDB. That's true even with
Terminal Server, each user should have his/her own copy of the front-end.

Larry Linson
Microsoft Access MVP

Nov 12 '05 #13
I tried that test over my PTP network and wrote 100,000 records in 10
seconds. Tried a few other tests and speed is not a problem. Worked up to
500,000 records.

Jeff

"Tom van Stiphout" <to*****@no.spam.cox.net> wrote in message
news:pv********************************@4ax.com...
On Sat, 15 Nov 2003 01:54:12 GMT, "Jeff Pritchard"
<je************@asken.com.au> wrote:

A peak append rate of 3 - 5 records (or higher?) per second is rather
high. I would take an hour or so to write a simple proof-of-concept
app to make sure your system can handle that. Tick a timer every 200
msec and append a row. Then add 100,000 rows to that table and try
again.

I'm not that concerned about the SELECT queries. The power of indexing
should ensure it hardly matters whether you're querying 100 rows or
100,000. Except for some increased network traffic. You could even
have those guys hitting a separate database, which you populate every
so often with a data dump from the first database.

Access IMHO is not a 24/7 database. I would lean towards a SQL Server
database for this.

-Tom.

I have an MDB that contains a single table. Checkweigher data is being
continuously written to this MDB from a dedicated workstation over a networkinto the MDB on the server at the rate of about 120 records a minute. Thismay increase in the future to about 200. The records are not large.

We have other workstations that run sessions on Terminal Server querying
this MDB, doing things like plotting the last hours data on a graph or
running other reports, or even just viewing raw data. No one else will be
writing to the MDB.

Can anyone see any potential problems.

I am concerned with the network connection being lost during the writing andhaving a corrupted MDB. The issue of speed is another. We expect 70-80,000records a day which means that there could be a substantial delay in
querying towards the end of the day. The MDB will be archived each day andstart empty the next day.

I don't see any locking issues.

Jeff

Nov 12 '05 #14

This thread has been closed and replies have been disabled. Please start a new discussion.

Similar topics

19
by: Nicolas Pernetty | last post by:
Hello, I'm looking for any work/paper/ressource about continuous system simulation using Python or any similar object oriented languages (or even UML theory !). I'm aware of SimPy for...
0
by: peterleeds | last post by:
Could anyone explain why Access 2003 is continuously calculating a continuous form, when it works perfectly in previous versions? On reading previous messages I gather that overlapping controls...
4
by: Kathy | last post by:
What is the standard technique for handling the fields in the following scenario on a continuous form? Multiple Divisions. Each Division has multiple Buildings. Each Building has a Supervisor. ...
0
by: Jeremy Wallace | last post by:
Folks, Here's a write-up I did for our developer wiki. I don't know if the whole rest of the world has already figured out how to do this, but I hadn't ever seen it implemented, and had spent a...
8
by: Dave Potts | last post by:
Hi, I'm just starting a development project in Python having spent time in the Java world. I was wondering what tool advice you could give me about setting up a continuous integration...
2
by: Phil Stanton | last post by:
Is it possible to display different images on continuous forms. The path to the picture is held in a table, and the photos are held in another folder. I am trying to avoid using OLE Object due to...
6
by: Greg Strong | last post by:
Hello All, Is is possible to use an ADO recordset to populate an unbound continuous Subform? I've done some Googling without much luck, so this maybe impossible, but let me try to explain...
7
by: capnwhit | last post by:
Hello all, The program below has a bug... The program is supposed to convert a string to a continuous hex dump and back to a string. The output should be as follows: abcd 61626364 abcd
1
by: fugs4ever | last post by:
I would like to ask anyone who can help. How I can write a continuous text to a file (file.txt). Meaning, after writing a text to a file, the next time I write a text, it will be in the next line....
1
by: nemocccc | last post by:
hello, everyone, I want to develop a software for my android phone for daily needs, any suggestions?
0
by: Hystou | last post by:
Most computers default to English, but sometimes we require a different language, especially when relocating. Forgot to request a specific language before your computer shipped? No problem! You can...
0
Oralloy
by: Oralloy | last post by:
Hello folks, I am unable to find appropriate documentation on the type promotion of bit-fields when using the generalised comparison operator "<=>". The problem is that using the GNU compilers,...
0
jinu1996
by: jinu1996 | last post by:
In today's digital age, having a compelling online presence is paramount for businesses aiming to thrive in a competitive landscape. At the heart of this digital strategy lies an intricately woven...
1
by: Hystou | last post by:
Overview: Windows 11 and 10 have less user interface control over operating system update behaviour than previous versions of Windows. In Windows 11 and 10, there is no way to turn off the Windows...
0
agi2029
by: agi2029 | last post by:
Let's talk about the concept of autonomous AI software engineers and no-code agents. These AIs are designed to manage the entire lifecycle of a software development project—planning, coding, testing,...
0
isladogs
by: isladogs | last post by:
The next Access Europe User Group meeting will be on Wednesday 1 May 2024 starting at 18:00 UK time (6PM UTC+1) and finishing by 19:30 (7.30PM). In this session, we are pleased to welcome a new...
0
by: TSSRALBI | last post by:
Hello I'm a network technician in training and I need your help. I am currently learning how to create and manage the different types of VPNs and I have a question about LAN-to-LAN VPNs. The...
0
by: adsilva | last post by:
A Windows Forms form does not have the event Unload, like VB6. What one acts like?

By using Bytes.com and it's services, you agree to our Privacy Policy and Terms of Use.

To disable or enable advertisements and analytics tracking please visit the manage ads & tracking page.