If I run against a Postgres backend, that is. And what are there variables that influence memory usage the most? Like large DAR packages, ACS size, etc.
2 Likes
Lately we’ve been testing heavily using a very complex DAML model (with a DAR artifact in the tens of MBs) trying to push throughput and we manage to run this with a 4GB heap. The ACS is kept on the database, do you don’t have to worry about it. The biggest factor to consider are likely the biggest transaction you may want to process and the size of your DAR (noting that it’s not kept in memory as-is but it’s translated into a format which is optimized for interpretation).