Cache Cleaning and Restart Schedule

We have moved from an i-series to a windows based environment with SQl db and clustered Logic and DB servers. Previously, we schedule the i-series to IPL every GO SAVE (once a week) and the current Windows system does not require a reboot after backup.

I have two questions really.
1. For the JAS server, should there be a cache cleaning schedule put in place and what is the recommended interval for cleaning?

2. When and how should the Windows servers be rebooted? Does anyone have any scripts already setup to reboot a cluster environment?

Thanks in advance.
-Elans

Well right after I posted this I came up a genius idea like I was struck by lighting. Here is what I have setup.

Our setup is as follows for E1.
1. DB server
2. Logic server
3. Cluster environment with DB and logic in a windows cluster
4. Jas server
5. Deployment Server

I setup scripts, just basic batch files scheduled to run starting at 4am, and the last running at 5am.

1. script to stop OAS on jas
2. script to stop JDE_Logic group on logic server
3. script to stop JDE_DB group on DB server
4. script to shutdown in this order, DB, logic, then JAS
5. scripts to startup services in this order, DB, Logic, then OAS
6. script to test the webpage after OAS has been restarted.

This probably is a little confusing without prior knowledge of a windows cluster environment and Cluster groups/resources so if anyone has any questions post them here and I will gladly answer.

Thanks.

Wouldn't you want your shutdown order to be JAS, logic and then DB?