Hi Surendra,

Pega server's entire startup times is a composition of various under the hood tasks. As far as I know, there are no readily available settings to minimize startup times significantly except for one(final note). However, 1 min seems to very determined :)

I have a vanilla Pega 8.3 running on Tomcat 9 and PostgreSQL server on the same system and it takes about 2 mins approx to finish the JVM startup considering there are no latencies or errors in the below pieces.

Pega startup time depends on the below items mainly:-

1. System Resources (like Memory, CPU) and DB performance & I/O (Numerous queries gets fired starting from initialization)

2. Loads the class definitions and Declarative caches.

3. Initializes Hazelcast cluster (Discovers other nodes using the same DB and establishes membership) 

4. Libraries, Conclusions, VTable caches, preLoad caches. 

5. Initializes Elastic Search (full-text search)

6. Initializes DSM, DataFlow, Stream services (based on node type)

7. Initializes Agents & Listeners. 

A bit of latency in any of the above tasks contributes to little higher startup times. 

I would suggest you to go through the entire startup logs starting from "System Start" until "Web Tier initialization is complete" to understand & fine-tune any of the above tasks. 

As you are done reading through all the points that I've mentioned, you can make use of the below setting to avoid loading declarative cache, Assembly cache preload, Conclusions cache on startup.

Though you get some benefit during the startup times, the user experience and performance will be impacted as the system tries to build all the caches on the first usage. So We don't really suggest it.

This is usually recommended for CLI processes like BIX where Pega Engine initialization is required without loading several caches.

Hope this info helps you and let me know if you any follow-up queries.