As other TSE mentioned there are lot of internal parameters and Suggested configuration to put in place. However, this is something which need a thorough analysis of logs and task setting . So that one can suggest as per issue or error occurred in your Specific environment.
As we mentioned earlier it require thorough analysis of logs and task setting. if you Still want to discover the options. then Click on the target endpoint (VERTICA) click on advanced Setting and then click on INTERNAL PARAMETER. press SHIFT+1 then you will find list of internal parameters.
As TSE we don't provide much expertise on internal parameters part, and internal parameter are not documented as well. we need to discuss the case or issue before recommendation.
Thank you for the post to the QDI Forums. As noted by @SushilKumar you can add a lower case 'x' as it will show all parameters for any given endpoint along with the extended options for the endpoint eg: Additional Connections Properties as shown below. Hope this is helpful and please share reply.
Under the Target Endpoint Vertica select Advanced and Internal Parameters
>> Are there any implicit parameters to increase the performance of writing to Vertica
Replicate is not 'stupid'. It was designed and has chosen default endpoint configuration values for maximum performance for typical usage. Without knowing specifics about your intended usage, or perhaps a description of what you believe to be insufficient performance, there is no useful advice anyone can give other then 'do your homework, or hire a consultant.'. A list of internal parameters may provide things to try, but you could easily end up shooting yourself in the foot. Measure! Test! Try! Measure some more, Test some more, try some more. Repeat until satisfied or tired.
Now the Replicate code Target Apply engine (which you did not ask about) has a generic design leaning towards near real time Replicate performance. As Vertica is more of a BI (large query) machine vs OLTP machine (like Oracle or SQLsever) you may want to tweak that tuning towards larger memory, longer delays.