Agenda
6.30 pm - Registration, doors, and bar open
7.00 pm - The journey of operation from a gigabyte to hundreds terabyte of data on Logging Elasticsearch cluster by Nutchanon Leelapornudom, Database Technical Lead, Agoda
7.30 pm - Machine Learning for Anomaly Detection, Time Series Modeling by Tom Grabowski, Machine Learning Principal, Elastic
8.00 pm - Q&A, Mingling
Talk 1
The journey of operation from a gigabyte to hundreds of terabytes of data on Logging Elasticsearch cluster.
Speaker: Nutchanon Leelapornudom, Database Technical Lead, Agoda
Abstract
This session will show you how our team manages and operate Logging Elasticsearch cluster from a few gigabytes to terabytes of data. I will talk about our Elasticsearch Architecture, which covers the area of availability, scalability, performance, JVM, and interested Elasticsearch features (i.e. Dedicated node, Shard allocation, Task management, Index priority) since small local cluster to large global cluster. I will share how we use 3rd party tools for provision, managing, routine scheduling, load balancing, and some in-house tools that our team develops for efficient query distribution, in-deep cluster monitoring and synchronize index mapping. I will also share some issues that I have experiences during scaling the cluster.
Talk 2
Machine Learning for Anomaly Detection, Time Series Modeling, and More
Speaker: Tom Grabowski, Machine Learning Principal, Elastic
Abstract
You don't have to be a data scientist to use Elastic machine learning features to build and operationalize real-time data models. Learn how they're integrated into the Elastic Stack, see how to use time series modeling for anomaly detection and forecasting, and get a firsthand look at new and upcoming features that will extend your use of machine learning to new use cases.