Hello,
First of all, my congratulations to the idea of "virtual folders" is damn nice, thanks for making this.
I have catalog-ed most of my data into VVV and I think the number of files are probably huge for VVV, the size of the vvv file is around 50MB. If I click on a virtual folder & say search containing "xyz" text, the program is probably trying to pull a lot of data into memory (i see a lot of virtual memory & CPU being used). I suspect this is either because
1) the "Firebird" database isn't able to scale to this many records (pardon my ignorance) or
2) we have a scaling problem in the way the db queries are written (probably using joins, etc..) that is hurting the performance.
I definitely feel VVV could turn swift if we address this db/ related stuff. I currently have a pretty decent (recent) hardware with me & I happen to wait for, something like 5-7 minutes to get the results of my search query. I haven't had a chance to look at the source to see whats happening underneath, will do so on a free weekend. Thanks.
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
It is strange that VVV is so slow, but I never tried it with a file that is so large.
About possible causes:
1) Firebird is not likely to be the culprit. It is an enterprise-ready database that can handle gigabyte databases without problems.
2) It is possible that queries are not optimal.
Keep in mind that a "containing" search cannot use indexes, so the database is forced to read all the records. This can slow down things, but I don't think so much.
Are you using Linux, OS X or Windows?
Are you using the last VVV version (1.0)?
If it is not a problem, may you send me your .vvv file (zipped) so that I can test it directly?
the address is:
mailing (at) fsoft (dot) it
Fulvio
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Hello,
First of all, my congratulations to the idea of "virtual folders" is damn nice, thanks for making this.
I have catalog-ed most of my data into VVV and I think the number of files are probably huge for VVV, the size of the vvv file is around 50MB. If I click on a virtual folder & say search containing "xyz" text, the program is probably trying to pull a lot of data into memory (i see a lot of virtual memory & CPU being used). I suspect this is either because
1) the "Firebird" database isn't able to scale to this many records (pardon my ignorance) or
2) we have a scaling problem in the way the db queries are written (probably using joins, etc..) that is hurting the performance.
I definitely feel VVV could turn swift if we address this db/ related stuff. I currently have a pretty decent (recent) hardware with me & I happen to wait for, something like 5-7 minutes to get the results of my search query. I haven't had a chance to look at the source to see whats happening underneath, will do so on a free weekend. Thanks.
It is strange that VVV is so slow, but I never tried it with a file that is so large.
About possible causes:
1) Firebird is not likely to be the culprit. It is an enterprise-ready database that can handle gigabyte databases without problems.
2) It is possible that queries are not optimal.
Keep in mind that a "containing" search cannot use indexes, so the database is forced to read all the records. This can slow down things, but I don't think so much.
Are you using Linux, OS X or Windows?
Are you using the last VVV version (1.0)?
If it is not a problem, may you send me your .vvv file (zipped) so that I can test it directly?
the address is:
mailing (at) fsoft (dot) it
Fulvio