Yes its saved in DB. The reason to do it is to have this file available to agents running on all nodes. Due to security we cannot have shared folders between nodes. Please consider a node as 1 physical node, not a virtual node.

I have attached the document which explains what I am doing...I have couple of other performance questions revolving this POC. 

Task 1 : Make Excel stream parsing using OOTB.

Task 2 : If excel has 40,000 rows it would be better I load first 500 rows, process them and load next 500. This would reduce the memory usage and ensures it works fine on multiple file processings.

This post is for task 1. Once we figure out the solution for task 1 next work will be on task 2. We have third party libraries like Apache POI which can do both task 1 and task 2. Instead of adding these libraries and build a custom solution in Pega I am thinking to expose Apache POI solution as an integration point for Pega.

Before that I am researching if there is any feasible option in Pega. If Task 1 works, there is a chance to request users to break down the large excel file as an alternate solution for Task2. 

Hi John,

Attached file shows the code.

Regards,

Naveen