Join this virtual meetup on Zoom: https://dell.zoom.us/j/99542967582?pwd=VUF1bUY0Ym8yWnJHUm5yR280Y1pjUT09

Key Value Tables is an experimental feature of Pravega that provides key-value stores (families or collections of key-value pairs) atop segments. An overview of Key Value Tables will be given with a goal of soliciting public review and outside feedback as the direction of phase 2 of the beta is being considered.

During the 0.9.0 development cycle, a Performance Evaluation of Pravega was undertaken, resulting in reproducible experiments comparing against Kafka and Pulsar, as published in two recent blog posts. This performance evaluation led to certain performance improvements and an overall characterization of Pravega's performance properties.

Part 1, When speeding makes sense — https://blog.pravega.io/2020/10/01/when-speeding-makes-sense-fast-consistent-durable-and-scalable-streaming-data-with-pravega/

Part 2, When Speed meets Parallelism — https://blog.pravega.io/2021/03/10/when-speed-meets-parallelism-pravega-performance-under-parallel-streaming-workloads/

The Pravega connector in Akka's Alpakka integration library is getting native support for Key Value Tables. In this presentation about Pravega Akka connector, we will learn about the connector in general and the recent work to adopt the new KVT APIs.

Akka PR: https://github.com/akka/alpakka/pull/2566