Good Morning All,
Does anyone have or know where to obtain a list of acceptable execution times for the major Methods in Epicor ? I.e. Erp.Proxy.BO.CustShipImpl GetByID, rp.Proxy.BO.PartImpl GetByID. ( does Epicor have a standard ? )
We ( meaning I ) often get complaints from Users that Epicor “is Slow” ( shocking right ??) 
and of course to make a case with Epicor , you need to run a trace. We have been ( and in fact are still going through) the extreme end of the spectrum ( 2 min to load Part Tracker, 4 Min to load Cust Shipment Entry) . Those are easy, that’s to slow.
But after this dust settles, and someone insists that I enter a case to Epicor that states Time Phase runs slow (when it runs in 5 seconds) , they( the User) may not believe me when I say, " that is ok to me " . I don’t want to wast Epicor Support Staff time on dealing with what I see as trivial complaints. There is a reason that only 2 people in our Company can initiate cases with Epicor Support…If I could get some sort of expected performance ‘benchmarks’ from Epicor , It would certainly make things easier for us support staff people (Epicor also) .
As always, appreciative of this group of people to bounce things off of…
Dean
That is completely subjective and based on about a billion variables
Database Size
Amount of Memory
Cache Availability
Disk Speed
Index Freshness
Network Behavior
Network Load
Phase of the Moon
Amount of Dust in the Air
Earth Magnetic Levels…
Ok some of these are more than others but really there isn’t anyone that can say “X” is the right number. The best you can do is look at consistency, Over a period of time doing X takes an Average of Z seconds… and if it goes 2xZ or 3xZ that’s a problem.
Sorry for the philosophical answer but I don’t think there’s an objective one, at least not one that’s any good.
Hi Jose,
So you are what they call a first responder ??? Thank you for the reply !! You forgot-- Which Cloud Server you are on haha…
My C# skills are not that good but I have been trying to work out de-serializing the trace log to a Class ( then to a datatable ) to store and be able to review…
Ahh, so many things, so little time…
Thanks though, a lot of wishfull thinking on my part.
Dean
1 Like
Maybe not always first, but yeah I’m around a lot… I run the forum along with a small band of misfits (@jgiese.wci, @rbucek, @bderuvo, @EarlGrei) so I have to be here frequently 
and I enjoy helping (when I can)
@DeanMiller I don’t think their SLA includes any kind of performance agreement.
One thing that you can try is an Azure Virtual Desktop session. I haven’t seen any numbers but the closer you can get the client to the server the better the performance. Ideally, you want your Azure Desktop sessions to be in the same datacenter as your Epicor instance.
1 Like
We are cloud based and have our VM’s set up in a datacenter(not on site) ( SE Wisconsin), Not sure Epicor would let us part a desktop on their server 
. I will work on collecting ‘normal’ for us and set that as the baseline.
I had mentioned earlier about collecting Trace Logs, working on a C# VS solution to read the trace log and parse out BO,Method,Execution time, time of dat etc.
@DeanMiller With Azure Virtual Desktop your desktops on are in Azure which is where Epicor hosts the servers. The desktop is managed by your Microsoft Azure subscription though and not through Epicor.
You would install the Epicor client on a remote desktop that is running in Azure and see what the performance is like.
1 Like