Ensar Basri Kahveci

overly distributed

Recent posts

Mar 1, 2018
Processing Guarantees in Hazelcast Jet I published this post at Hazelcast blog and put a copy here. Stream processing is a paradigm for on-the-fly processing of unbounded data feeds. We have been witnessing that stream processing engines (SPEs) get more attention every day in the era of fast data and become a fundamental component of data processing pipelines. They usually run in distributed settings to be able to cope with the flood of incoming data entries.…
Dec 6, 2017
Hazelcast Jet 0.5: Fault Tolerant Stream Processing Made Easy I published this post at Hazelcast blog and put a copy here. Stream processing is an emerging computational paradigm for on-the-fly processing of live data feeds, targeting low latency and high throughput. Streaming applications are usually deployed on multiple servers to achieve these requirements. Since even a single failure may lead to incorrect results or long interruptions in result delivery, fault tolerance is of paramount importance in such long-running distributed applications.…
Jan 22, 2016
A simple unstructured P2P network implementation in Java Hello everyone, I have developed an unstructured P2P network using Java and Netty recently. You can check its source code on Github. Any contribution is welcomed.…
Jun 3, 2015
K-Means Clustering with 80 lines of Scala I just implemented the K-Means clustering algorithm with 80 lines of Scala. read and createRandomCentroids methods are quite dummy and can be written much better. You can access to the gist via https://gist.github.com/metanet/a385d42fd2cab9f3d20e …
Oct 12, 2014
cd to the directories you "frequently and recently" use pretty easily with Z Z is a command line tool that helps you to cd to directories you use with just a few key stores. It manages this by tracking the directories you visit, and weighting them by how frequently and how recently they have been visited. I have been using it happily for a long time, and now I think that I can recommend it to other people. It uses a term frecent for ranking directories.…