why I am getting the variation on process execution time for the same file and the same process
Process Execution 1 -> 0.45 sec
Process Execution 2 -- > 2:11sec
Do you run the process on a local Atom or on an Atom cloud? Can you share the code of the data process shape "Group by user ID?
The process is running on a local atom.I have used split document - Profile - User ID in data shape.Is there any other ways to improve split document By Profile - Field Name.In map , I am transforming data from one profile - to another profile without any modification.
The split function of Boomi is very simple. I'm not aware of any undocumented features. Anyway, it's difficult to judge this issue remotely. Looking at the screenshots, you say that splitting the same file takes 0.6 vs 50 secs using the same process? And if you run it 10 times, how do the numbers then look?
There are many areas which can be looked into but it all depends on how you are executing these processes. Few areas to look into are as follows
1. What is the memory being used by Atom.exe in each run?. Worth looking at the Resource Monitor
2. Are you running this process multiple times in parallel?.
3. What is the load on your local atom for each run?. I mean number of other processes running.
4. Running on Low Latency mode might help increase the performance of process but then you will have to look at the timing in the log directory.
5. Garbage collection properties can be revisited if required.
Retrieving data ...