Hi, We have a problem with the usage of the RAM memory on our Appian PROD s

@Natasha, as Myles indicated, the load metric has nothing to do with engine memory utilization. You simply are outgrowing your hardware with the additional data that you are adding - mostly process data. In the very short term, you could split up JBoss and web server into another VM but you will still be running at 75% capacity which is pretty high and you will need to add capacity. This is also a good moment to review your archiving settings, which may give you more mileage out of your existing hardware. Run the sizing script on all your exec engines, Appian PS can help you analyze the data and provide recommendations. Also, from your indication, it seems like you still have auto checkpointing running in production which is NOT recommended. For your HDD issues, I would recommend lowering the amount of kdbs you are keeping on disk, since they are all cumulative. Check the size of your process archives and logs, you may be able to clean those too.

First of all about this suggested reconfiguration:

cleanup script for logs and reduce the number of logs kept in the <APPIAN_HOME>\\ear\\suite.ear\\resources\\appian_log4j.properties

what we have is following:

D:\\appian\\server\\_scripts\\cleanup.bat logs -delete -keep 7

and the number of logs is set by default.Also as far as I see the D:/appian/logs folder is not holding a big part of the hardisk space ~1.57 GB and it does not reflect on the usage of the ram memory right?

For the second Ronald question, yup we are running everything on the same server (Appian and Jboss).

One more thing:
Have in mind our hardware (16GB RAM ) and MAX_EXEC_ENGINE_LOAD_METRIC parameter set to 960 is that customization is appian best practice or should we decrease it ?

About this point:
Also, from your indication, it seems like you still have auto checkpointing running in production which is NOT recommended.
Should we disable it?

Please leave the Load Metric property as-is, any value above 120 is fine.
As for checkpointing, please check this page forum.appian.com/.../Configuring_Application_Checkpointing
By setting up scheduled checkpoints, you have more control when the checkpoints run and when the kdbs are not available.
However, I would like to reiterate the point that this is all for nothing if you do not increase your hardware and look at your memory profile.

Auto checkpointing, yes it should be disabled.
Logs won't affect memory. Its just good to have in place.

Is it possible to split JBoss and Appian to two different servers?

About mention reconfiguration and disabling schedule script for check-point I see that it is available for appian 7.2 and we are still on appian 6.6.0.But we will have that in mind when we will migrate to appian 7.0.As far as I understood for now we will need to disable that schedule task ,right?

If you run jboss AS in production, you will be on an unsupported configuration.

On 7.0 it is working with 5.1.0 eap not with as,so that is supported?

Also if we split jboss and appian on two different servers do we need additional licence file and it is free or not?