Gonzalo Ortiz Jaureguizar

Same happen with Strings, which are just binaries that are not being [marked as Strings](https://github.com/apache/parquet-format/blob/master/LogicalTypes.md#string). Apache Pinot is quite strict about that and when importing data from parquet (which is...

Also, right now brokers temporally store the data from servers in a hash map, which implies to hash is twice (once when it is being read from the server and...

I would recommend to give a try to [Chronicle bytes](https://github.com/OpenHFT/Chronicle-Bytes). We could use it to replace larray (and therefore be able to run with Java 17) and also to be...

I've just made an analysis of the docker image. It seems that the three bigger layers are: - 344MBs of base image (jdk-slim), which is highly reusable. That means that...

BTW, I had very good experience with Jib in the past. Given the complexities of Apache Pinot building process, I don't think we can directly use Jib, but we can...

I've added some changes to also optimize the `order by sorted_col DESC (, more columns)*` case. The results are even better in this case because the priority queue is quite...

I've fixed one bug detected by the integration tests (it is an error to create a PriorityQueue with 0 elements) but I wasn't able to fix the second one because...

I think I finally found why the test was failing. It was due to a `cmp >= 0` assertion that should be `cmp

> assert can be enabled by setting a JVM parameter: -ea Yeah, that is the behavior of assert. The problem is: In several parts of Pinot there are try-catch sections...

> For descending order, ideally we should implement an iterator which can read rows in reverse order That was my first approach and for sure it would be better. But...