Advice for a Novice

Hunter J.

Registered User.
Local time
Today, 05:37
Joined
Feb 1, 2004
Messages
24
Hello,

I could use some advice if at all possible. I have created a database that will report production information off a specific machine on the production floor.
Here are some my questions/concerns:
1. I have the potential to log 400k records a day. Most of them are bits 1 or 0, but that is a lot of records I think. Is there any way to automatically run a report at the end of or start of the day and then save the report in a different folder with a unique name for later use?
2. A very large portion of the data that is logged is 0 and I could stand to remove these from the database and this would cut down on size/number of records and speed reporting up.

Ok, now I am trying to improve it and I can reduce the amount of data logged but I am not sure how to bring it all together and that is what I am working on today/weekend. This would make some of the above a mute point but is there any advice to help on the two concerns above until I finish my improvements?

Thanks,
Hunter J.
Novice on Access
 
1)
Hmm not quite sure what you require here. Possibly an export as a text file which the db holds a reference too - use the date reversed : yyyymmddhhnnss (format function)
Alternatively if its only an ID, date/time and a results (byte or bit) even though you have loads of records, it won't/shouldn't take up that much space. Granted its a lot to wade through.

2)
Depends how you are reading the data in. If via code then ignore the line with 0 (as you don't want them) so you never store the data.
Or hold it anyway as a reference.
Or run a delete query after importing, to remove those records. Note that the space on the hd would still be used and the mdb would need a repair and compression.


If you can run tests, then do a test for a week (three mdbs each processing the same data) so you can compare times/processing/sizes etc


Vince
 
Well, to begin with, the format/meaning of this data set is not clear from your description. In the final analysis, ALL data sets are 1's and 0's inside the computer. What is significant about the pattern that makes you want to record it in the first place?

The rest of your answer depends on getting this data set into a form that is somehow meaningful. Without knowing the meaning, we are going to be lost. I'm not about to tell you how to get to point B when I don't know where point A is. I could do serious damage to your project despite my (relatively) good intentions.
 
Sorry it took so long to get back.

Let see if I can explain what I am trying to do. I have created two tables that data is automatically logged into by a PLC. One table is for the current product that my equipment is running and this data is start/stop date and time. In the other table I log production information that is also date and time stamped. I am trying to group the production information to the product that was running during that time.
In the products table you can see TagIndex 28 started at 11:04:06 and stopped at 11:19:18. In the Float table there is data that was logged during this time frame that I would like to associate with TagIndex 28. Any help is appreciated.



Thanks,
Hunter J
Novice at Access
 

Attachments

I cannot download your zip due to site firewall rules. (I'm at a military site.)

If you are saying that most of the data you get from the computer-controlled device is null bytes and that it continually sends you streams of nulls when it is doing a task but isn't yet finished with it, then you have a data filtering problem. It would be possible to write a little snippet of VBA code that opened your file and got rid of null sequences before it did the import. Read the file in text mode, find sequences of some number of bytes of 0 and start deleting them as you find them. Write the remnants to a second file. Import the SECOND file (after it has been pre-scanned.)
 

Users who are viewing this thread

Back
Top Bottom