45mb base64 encoded PDF msgs

[Jerry Sawa]

We’re looking at a Cardiac Monitor system that will be sending HL7 base64 encoded PDF msgs outbound to our Epic system.  The msgs can be as large as 45mb.

This makes me nervous as I’m not sure what issues (if any) Cloverleaf may experience.  Disk space, SMAT DB size, Recovery DB size, processing time, etc.

Has anyone dealt with any systems that sends these size msgs thru Cloverleaf?

I’m leaning towards just using our OnBase document mgmt system to retrieve the PDFs from the Cardiac Monitoring system and having OnBase send links to the documents to Epic.

• This topic was modified 1 year, 5 months ago by Jerry Sawa.
• This topic was modified 1 year, 5 months ago by Jerry Sawa.

Topic

[Todd Hamilton]

Jerry,

For a data conversion project I worked on recently, I used cloverleaf to send large PDF files to Epic through Epic Interconnect based interface.  I did run into a minor size issue that I never resolved.  (Just ran out of time). The size issue happened around 90MB files and was most certainly a limit of our specific Epic interconnect implementation and not Cloverleaf.

Something to consider, if your OnBase implementation, server, and team are AMAZING.  Then by all means do this through OnBase.  If not, you might want to consider Epic Interconnect.

todd.hamilton.omaha@gmail.com
(402) 660-2787

[Jerry Sawa]

Thanks Todd.  I was so concerned with Cloverleaf that I hadn’t even considered Interconnect issues.  I think we’ll go with OnBase.

[Peter Heggie]

We have a document interface that occasionally sends base64 encoded PDFs that are larger than 15M, and we have filtered those out because the target system has issues handling that size.

The Cloverleaf interface can handle such a message but operationally, if the interface is down for any length of time, and input messages build up, the process start up can take a lot of time and other resources to read them all in. The process has exhausted memory in the past.

If you turn recovery off, you avoid many of these issues. And it would be better if you can avoid filtering or translating, although that is hard to avoid. You certainly want to run this through a dedicated process. You may want to setup a batch process to periodically restart the process.

Peter Heggie

[Jerry Sawa]

Thanks Peter.  If we do end up using Cloverleaf, I’ll definitely keep that in mind.

[Rob Lindsey]

On thing to due if you have to do any translations, is to write it out to a directory, store the location and name in the message or Userdata then pick it up on the way out the door.

Rob

[Jerry Sawa]

Thanks everyone.  Much appreciated.  I’ll keep all of your suggestions under advisement.

[Author]

Replies