Alibaba Tech Meetup #1

JVM Health Rhythm by Alexey Ragozin 
Java Virtual Machine is a very sophisticated system. Usually, it just
works and you can care less about all the magic behind the scene.
Though, if something goes wrong, understanding internal mechanics of
JVM would help you to find a root cause.

There is no lack in Java profilers both commercial and open source.
There are bunch of metrics typical Java profiler can show. But how to
distinguish healthy metrics from a symptom of the hidden problem?

We would walk through common metrics exposed by JVM and uncover what
patterns you should expect and what problem symptoms can manifest
themselves through these metrics.

Data Pipelines. From zero to cloud scale by Artem Pyanykh
Data is one of the most important assets that can give a competitive advantage to your business. Back then you would probably have to build your own Hadoop cluster which requires big investments and incurs significant maintenance costs. Luckily things have changed and now building a data processing pipeline can actually be quick, easy and fun thanks to awesome open-source libraries and cloud infrastructure!

In this talk, Artem will explain what are the main components of a data processing pipeline, show how to manage ETL tasks with an open-source tool called Luigi, discuss potential issues with resiliency and scalability of the solution, and how cloud services like Google Cloud Storage, Google BigQuery, and Google Dataflow can remedy them. He’ll conclude with an overview of operational costs for the system, both in terms of development and maintenance effort and service bills.