benchbase
benchbase copied to clipboard
Tpch load retries
In our testing with TPCH we found that for large scale factors the loader wasn't fault tolerant and would have to be restarted after any connectivity failure. In this PR we implement a retry mechanism where the loader retries batches a few times after failure. This was very helpful in cases where we had very high scale factor and there was intermittent connectivity. It's quite a big PR so feedback on how I can package it better is greatly appreciated.
Thanks for sending this. We will take a look at it later this week.
@dbanda Any update?