May Picks

Sat, Jun 11, 2016

“Random picks from all over the Internet”.

English

Articles

• Apache Showdown: Flink vs. Spark If you’re looking into stream processing solutions, you may want to check this out. It gives a nice overview of differences between Apache Spark and Flink.

• Notes on practical machine learning ru-RU How to get good results when using machine learning and the steps required. Lots of very good points.

• Disadvantages of purely functional programming Notice the word “purely”. I tend to agree with the author on 1, 2, 4, 8 and 9.

• No shard left behind: dynamic work rebalancing in Google Cloud Dataflow If you’re working with one of the MapReduce frameworks, it might be interesting for you to see how Google Dataflow implemented dynamic work rebalancing as a way of dealing with stragglers. It reminds me work-stealing in Erlang.

На русском

Статьи

• О приоритизации

  1. Почему важно отказываться от некоторых задач
  2. Почему важно делать некоторые задачи не на 100%

• Размышления о Питоне Clojure захватывает умы питонистов! Плюсы / минусы языка.