Slow WAN import vs ? <your suggestion goes here>

Questions
1

Hi everyone,

I have a file hosted at a hosting provider. I do not have access to the server itself or its file system. I need to update 100k + records without triggering auto-entry options (this leads me towards an import to update the records).

This is a one time update, not something that I need to code to happen at interval x.

I have exported the data locally to my machine to re-import it, so I guess I could attempt exporting it to the server, in a file format other than fmp12 (because server cannot import from fmp12 files) and import that back in, having to make sure the server replicates the proper found sets and figuring out how to map that (I guess I’ll have to export to locally to be able to map the import script step).

As you may guess, a WAN import of 100k records is… slower than what I would like it to be.

Is the server-side process I describe what you would turn to if, like me, you wanted to avoid a slow-moving progress bar?

Thanks for any input you may share.

2

What is the size of the records in MB and available bandwidth over WAN?

3

The downloaded data file is 26MB.

The WAN bandwidth (if you refer to something like speedof.me) reads 73Mbps downstream and 12Mbps upstream.

4

For this type of maintenance, I download the file locally and do the update/import operation there. Then load it back on FMS via ‘Upload to Host’. Normally done overnight, when I can take the files offline.

I am running that cycle with a 8 GB file over a 30 Mbps line (VPN connection). Takes about 30 min for down or upload and works just fine.