I ran the select count(*) from obs and got 28,571,890. So we do have a ton of them in there.
We have 48,284,579 and do not have speed issues so it is all in the configuration and hardware that you use.
Now another question... The person that set our EMR up in the beginning has the DB and JBOSS on the same server... GE doesn't recommend this setup over 50 users. We have 65. Is it difficult to move the JBOSS part of the EMR to another server?
Our CPS (for PM only) has a SQL server and then the application server is separate. We never have issues with slowness or anything really with CPS.
If you are experiencing performance issues with your DB and you are not running on SSD drives, I would HIGHLY recommend the switch. A single SSD drive usually has more IOPS than 20-30 spinning disks. Since you would never have a single drive for anything, running that drive in a R5 or R10 array of them is that much faster.
I do not know what the process would be to separate app and db but I do not imagine it would be terribly difficult. To determine if you need it, take a look at the CPU and memory usage on the server. If they are high, it might be time to separate. Disk IO would be another (major) bottleneck but it is slightly more difficult to determine if that is your issue. (you would have to know what your IO capabilities were when nothing was running and then determine what it was running at during peak load)
One more thought. When we were having speed issues (due to our snapshots) GE provided us with a script they called "getstatistics" which went really in depth to tell you what was slowing down your server. You really need a DBA to look at the output to give you an idea of what the output means. Open a case with them and see if you can get someone (preferably Hillary Hart, she knows it inside and out) to provide the script to you and interpret the output.
I might be able to locate the script but again, unless you know what you are looking for, it will not do much good.
If I call into GE's Platform group is Hillary a part of that group?
She is a DBA that works in the group. She is not a front line support person but if you open a ticket with support and mention that you are looking for that script and she was the one that provided it to us you might be able to have them pull her in as a resource.
I went ahead and started Archiving these a few days ago. Since then our performance has actually greatly improved (knock on wood). We were able to archive roughly 10k charts observations. Any chart that had over 1000 obs were set to archive and only keep 500 at a time.
Thank you for everyone's input on this!
Good news, thanks for letting us know.