Slow JMP data table opening?I've got a customer that has a JMP data table, running JMP Pro version 13, Windows 64 bit OS, 8GB RAM, and the table has about 15 million rows, with about 15 or 20 columns...mostly Character data types. Using WebEx, he reversed demo'd the file opening process for me today and it took about 10 minutes to open. Says it sometimes takes upwards of an hour. Any 'gotchas' we could look out for? I sent him to JMP tech support, in addition to trying Compress Columns. The Compress Columns seemed to help a bit, with an n = 1 of re-opening iterations. But nothing substantially different.
Any hints?
Re: Slow JMP data table opening?In 13 you can select specific columns before you open them in a new data table specifically for this situation of heavy character data. Perhaps you suggest that he could open only critical columns to his analysis and test load times to find a problematic memory intensive column. I just uploaded a video on this topic, however it's simply an option in the Open window once you select a jmp file, select Select Columns in the Open button drop down.
https://www.youtube.com/watch?v=HRoaowtyg8A
You can select specific Columns before opening a data table. If you don't plan to use all of the columns in your analysis, then you may want to create a new data table containing only the columns you’re interested in. This reduces the amount of memory that your data table uses. And if your data table has lots of columns, especially long character columns, it will require a significant amount of memory.
Re: Slow JMP data table opening?If it's a bunch of categorical stuff, you could try this add-in... just be sure they get rev1.
Compress to Labeled Code
Re: Slow JMP data table opening?Chris:
According to Windows File Properties: Size = about 154MB...and I saw this for myself.
The icon he clicked on was on his desktop so I'm assuming his hard drive.
Re: Slow JMP data table opening?Hi Peter,
2 things. Where is the file located? What is the file size?
If the columns are character, mostly, it will inherently take longer to load and more memory taken.
