Radoslav Petrov

Results 23 comments of Radoslav Petrov

@nbrady-techempower @bhauer https://github.com/TechEmpower/FrameworkBenchmarks/issues/4092#issuecomment-551295713 The problem with the TCP/IP connections is still there and there is no comment from your side. I've found a possible reproduction steps. It's just a configuration...

1. About `Hibernate`: AFAIR most of the Java projects that are using `Hibernate` were updated to use the `JPA` interfaces. If the project is on Hibernate version 5.4.1 this should...

@nbrady-techempower My understanding is that [`innodb_deadlock_detect`](https://dev.mysql.com/doc/refman/8.0/en/innodb-parameters.html#sysvar_innodb_deadlock_detect) should be `OFF`. Also [`innodb_lock_wait_timeout`](https://dev.mysql.com/doc/refman/8.0/en/innodb-parameters.html#sysvar_innodb_lock_wait_timeout) should have the same value as PostgreSQL. For the current very short burst for benchmarking IMO the range should...

@nbrady-techempower @jderusse I stand corrected. I've updated the post.

@nbrady-techempower @jcheron I've get an error when I run `vagrant@TFB-all:~$ tfb --mode verify --test dropwizard-postgres --concurrency-levels 128`. ``` Verifying test fortune for dropwizard-postgres caused an exception: can't multiply sequence by...

@jcheron @nbrady-techempower Is it possible to add some form of delay before the verification queries are executed. For example some command-line option with default '0 seconds'. I'm having a hard...

With `-mode debug` and manual execution of the extracted-from-the-code reporting query (sort of) I don't get discrepancies.

@nbrady-techempower I think we have misunderstanding - the change about `concurrency-levels` is OK and working. My request is about adding some sort of configurable delay before execution of the verification...

Another issue reminded me about these two: https://github.com/TechEmpower/FrameworkBenchmarks/issues/1220 and https://github.com/TechEmpower/FrameworkBenchmarks/issues/1227 Basically reporting latencies when variable loads (different RPS) are used is misleading.

@nbrady-techempower Idea/suggestion for extending the toolset validation. Get and log the basic stats for the number of executed statements. The related info could be found here: - PostgreSQL https://www.postgresql.org/docs/10/pgstatstatements.html -...