Chronicle Map VS HikariCP

Chronicle-Map: Chronicle Map is an in-memory, key-value store, designed for low-latency, and/or multi-process applications.

https://github.com/OpenHFT/Chronicle-Map - Maybe a better offheap map

I've wrangled data sets in the ~600gb range using nothing but plain old Java and a few beefy boxes. This can all be kept in memory, but you have to go off-heap. You can use Chronicle Map and Chronicle Values to model this data and work with it off-heap in a way that's still very clean and object oriented. 128gb of RAM is cheap these days, whether you're in the cloud or not.

I have used hikari and exposed to do this in the past with postgres, although other dialects are supported.

I found one suggestion from the author of HikariCP on how to address this, which I implemented and it worked. However, there are additional classes involved, and it feels a little clunky and hard to follow.

Whatever you choose, presuming your app is database heavy, I highly recommend spending time on DB schema design to make all queries as short as possible, avoid relying on transactions, and keep your connection pool tiny. For reference: https://github.com/brettwooldridge/HikariCP/wiki/About-Pool-Sizing

You reminded me of the HikariCP library and it’s documentation: Clear and simple with references to other libraries trying to accomplish the same thing. It is not in the closure space though.

https://github.com/brettwooldridge/HikariCP

There's a reason pretty much any system will use a connection pool (Hikari is the de facto standard). Yours should to.

HikariCP to pool and conserve relational database connections, which are a scarce and valuable commodity

HikariCP is a popular JDBC connection pool, battle-tested and exhibiting good performance. A Clojure wrapper does the things discussed in this post here. However, our project uses HikariCP directly.

YBDemo is a simple Java program that creates an HikariCP connection pool from the hikari.properties file in the current directory, and takes SQL statements to execute as lines from stdin. There's no multi-line statement: each line is a thread, executing the statement in a loop. The goal is to make it easy to run a demo with concurrent threads by providing the set of queries in a simple way, interactive or though file redirection. Only the first column of the first row is displayed, I use row_to_json() or json_agg() to format a larger result into one value. The thread stops if no row is returned, I use RETURNING to get a row from DML.

Connection pools are configurable, and different implementations have different parameters that you can adjust. For example, you might want to configure the maximum number of connection objects that the pool maintains or the maximum lifetime of a connection in the pool. Refer to the pool implementation documentation for more information.

In Spring Boot application that use blocking connection to DB such as JDBC connection pooling is usually handled by a popular library called HikariCP. Luckily, Spring Data R2DBC already includes connection pooling option that we can use just by enabling it from our properties.