Since Java 11, JDK Flight Recorder (JFR) is getting more and more popular among developers. There are two reasons behind this fact, first, JFR got open-sourced and included into OpenJDK - previously it was part of a JRockit JVM and commercial version of Oracle JDK (see JEP 328 for more details.). The second fact is that it is a tool for collecting diagnostic and profiling data about running Java application - because it is so deeply embedded to JVM it provides extremely low overhead (about 1% out of the box).

AWS Lambda support many build in runtimes. But we also have a possibility to create custom runtimes to run functions in a language that is not yet supported by Lambda service or when we need to do some extra optimization (i.e. reduce startup time or memory footprint of functions build for JVM). Today we will learn how to build a very simple runtime focusing mainly on steps required to do so.

Looking on some of the AWS efforts around RDS service it won’t be an incorrect statement to say that they want to bring the world of relational databases closer to Lambda functions. First, Aurora Serverless was announced then less than a year later this revolutionary service was extended by Data API where interaction with DB is done through API endpoint. But this is not an end, the next year has come and we got RDS proxy whose main task is to help improve applications scalability by managing DB connections.

If you are wondering how could you set up your build pipeline to do a deployment of your application to Kubernetes cluster then you can find this post interesting.

It’s almost end of 2019 but there is still a lot of application which depends on Servlet specification - this statement is base on my personal experience where I still see the majority of a new application being built using this technology. In this blog post, I will focus on a single use case where we will retrieve HTTP request data from outside of calling thread in Spring Framework.