Louis FruleuxinTeads EngineeringRunning Spark Pipelines on EMR Using Spots InstancesA compilation of good practices and lessons learned in a production environment8 min read·Feb 22, 2022----
Louis FruleuxinTeads EngineeringInvestigating a network issue encountered with Kafka on AWSTL;DR we ended up tuning the ARP cache on our EC2 instances6 min read·Sep 16, 2021----
Louis FruleuxinTeads EngineeringUpdating to Spark 3.0 in productionBreaking changes and expected improvements: a production point of view10 min read·Dec 3, 2020----
Louis FruleuxinTowards Data ScienceCompact prediction treeA Lossless Model for Accurate Sequence Prediction over a finite alphabet7 min read·Oct 29, 2020--3--3