Dear Sirs,
we tried to generate a very long sequence for test purposes and we noticed that the Sequence windows starts becoming very unresponsive after every user input. After a while everything returns normal, until the next input. The behaviour looks strange because, while the GUI is unresponsive, the CPU consumption for Clarity process is less than 3% and also the hard disk access light indicates light load.
We managed to reproduce the behaviour on 2 PC with Clarity 2.6.6.574 , Clarity 2.7.4.543 , Clarity 3.0.2.244 with Dani Master GC driver or Agilent 6890 driver, setting the sequence as you can see in the attached image.
Additional information:
Test PC 1 : CPU Intel Core 2 Quad Q9400, 3.25 GB RAM , OS: WinXP 32bit SP3
Test PC 2 : CPU Intel i5, 4 GB RAM , OS: Win7 64bit
Always used a "new method" for the tests.
Best regards,
Guido Pagani
Clarity Sequence window unresponsive with very long sequence
Re: Clarity Sequence window unresponsive with very long sequence
Dear Guido,
this is a long known problem, the reason is check for duplicate file names within the sequence. It becomes apparent also during run of such sequence during opening of the chromatograms. The time needed increases quadratically with number of injections - it is recomended to limit number of injections in sequence to several hundreds.
Ivan Vins
this is a long known problem, the reason is check for duplicate file names within the sequence. It becomes apparent also during run of such sequence during opening of the chromatograms. The time needed increases quadratically with number of injections - it is recomended to limit number of injections in sequence to several hundreds.
Ivan Vins
Ivan Vinš
Re: Clarity Sequence window unresponsive with very long sequence
Dear Conference,
We notice something strange using sequence file (*.seq).
Some customers reported interface freezing for 10/15 sec., in few (old*) PCs. This happens when they open sequence window, when they start the sequence and when sequence table changes the status of the rows, after a single analysis run finished.
We remember that these freezing problems are strongly linked with the total numbers of injections the sequence will do. But in this case the sequence files have only 3 or 4 rows with a total injections number less than 20.
We noticed that its sequence files are very big (~10 MB)!!!
This could be due to the fact that the customer worked for long time with this sequence files. He modified sequence files adding or removing or changing the rows in the table.
The large files also contain (average) ~13000 sequence audit trail rows!
To verify the relation between the file size and the freezing problem, we “Save As” the large sequence file. The new one isn’t so big (shrunk from 10MB to 10KB!!) and doesn’t have the freezing problems.
We make some tests to better understand if the size is only influenced by the sequence audit trail rows.
We create a new sequence file and we save it -> Kb 0.
We add a row and we save it -> KB 2.
We add another row and we save it -> KB 4.
We change the repetition number and we save it -> KB 5
We change the first vial and the last vial and we save it -> KB 6
We add another row and we save it -> KB 8.
We remove the last row I added and we save it -> KB 8.
We remove another row and we save it -> KB 8.
After this test one question spontaneously arrives. If I continuously change the file (also without starting the analysis), does the sequence file size rise?
We think that some users that work only on a single sequence file, could notice these slowing down problems.
Do you notice the same problems?
Thank you and best regards,
Domenico Agostino
* the old PCs I talked about always meet your recommended hardware requirements.
We notice something strange using sequence file (*.seq).
Some customers reported interface freezing for 10/15 sec., in few (old*) PCs. This happens when they open sequence window, when they start the sequence and when sequence table changes the status of the rows, after a single analysis run finished.
We remember that these freezing problems are strongly linked with the total numbers of injections the sequence will do. But in this case the sequence files have only 3 or 4 rows with a total injections number less than 20.
We noticed that its sequence files are very big (~10 MB)!!!
This could be due to the fact that the customer worked for long time with this sequence files. He modified sequence files adding or removing or changing the rows in the table.
The large files also contain (average) ~13000 sequence audit trail rows!
To verify the relation between the file size and the freezing problem, we “Save As” the large sequence file. The new one isn’t so big (shrunk from 10MB to 10KB!!) and doesn’t have the freezing problems.
We make some tests to better understand if the size is only influenced by the sequence audit trail rows.
We create a new sequence file and we save it -> Kb 0.
We add a row and we save it -> KB 2.
We add another row and we save it -> KB 4.
We change the repetition number and we save it -> KB 5
We change the first vial and the last vial and we save it -> KB 6
We add another row and we save it -> KB 8.
We remove the last row I added and we save it -> KB 8.
We remove another row and we save it -> KB 8.
After this test one question spontaneously arrives. If I continuously change the file (also without starting the analysis), does the sequence file size rise?
We think that some users that work only on a single sequence file, could notice these slowing down problems.
Do you notice the same problems?
Thank you and best regards,
Domenico Agostino
* the old PCs I talked about always meet your recommended hardware requirements.
Agostino Domenico
-
Daniel Mentlik
- DataApex Support
- Posts: 354
- Joined: Fri Mar 27, 2009 3:15 pm
Re: Clarity Sequence window unresponsive with very long sequence
Dear Domenico,
the behaviour you are observing is correct and indeed has something to do with the number of rows in the Sequence Audit trail. The sequence is not meant to be uses indefinitely, but a new sequence file is expected for a new series of samples. Simply deleting already measured rows and adding new ones to the same sequence does not delete the audit trail, and it can raise to substantial size.
During my testing:
- each entry in the audit trail produces approximately 0.5 kB of data more in the sequence file.
- each new row adds approximately 0.7 kB of data; it does not matter whether the row is completely filled in or partially empty, the size of the row is still those 0.7 kB
I would suggest to the users either:
- creating a new sequence file for each measurement set by using the File-New command (the pre-made sequence file can be saved as a template.seq file in the COMMON subdirectory of Clarity installation directory to save some work with the new sequence)
- using the Save As function on the actual sequence file (during the Save As operation the audit trail in the newly saved sequence file is deleted)
- disabling the additional entries inserted into the sequence file audit trail (thus its size will not rise that much during the usage). This can be done in the Audit Trail Settings dialog / Sequence tab, by unchecking all checkboxes there. Please note that the messages that the sequence was saved will still appear and that the sequence is saved after each row was measured, so this will only reduce the size of the file to one third approximately.
Best Regards,
Daniel
the behaviour you are observing is correct and indeed has something to do with the number of rows in the Sequence Audit trail. The sequence is not meant to be uses indefinitely, but a new sequence file is expected for a new series of samples. Simply deleting already measured rows and adding new ones to the same sequence does not delete the audit trail, and it can raise to substantial size.
During my testing:
- each entry in the audit trail produces approximately 0.5 kB of data more in the sequence file.
- each new row adds approximately 0.7 kB of data; it does not matter whether the row is completely filled in or partially empty, the size of the row is still those 0.7 kB
I would suggest to the users either:
- creating a new sequence file for each measurement set by using the File-New command (the pre-made sequence file can be saved as a template.seq file in the COMMON subdirectory of Clarity installation directory to save some work with the new sequence)
- using the Save As function on the actual sequence file (during the Save As operation the audit trail in the newly saved sequence file is deleted)
- disabling the additional entries inserted into the sequence file audit trail (thus its size will not rise that much during the usage). This can be done in the Audit Trail Settings dialog / Sequence tab, by unchecking all checkboxes there. Please note that the messages that the sequence was saved will still appear and that the sequence is saved after each row was measured, so this will only reduce the size of the file to one third approximately.
Best Regards,
Daniel
Daniel Mentlík
DataApex
DataApex
