Description
Discover how IBM Cloudant engineers use data to inform development and operations every day! Sources such as Splunk and Graphite are pivotal for meeting our service needs. I will demonstrate a data processing engine based on Pandas and NumPy that can help tame these sources. Then discover how you can use Jupyter notebooks with IBM's Bluemix platform to do more with data.
Snakes on a Cloud: "Data science, meet DevOps"
Running an always-on database service such as IBM Cloudant is a complex task. Accomplishing this requires a mix of development and operations informed by data.
Various data sources such as application logs, system metrics, and project management systems are used every day by Cloudant engineers.
First I'll describe how DevOps works at Cloudant, and how data is used to meet our service needs. Break through the buzzwords and see the reality of a running service.
I'll demonstrate a data processing engine called Forecast, which combines various sources at Cloudant for analysis. Forecast allows engineers to go beyond manual inspection by using Pandas and NumPy to analyze information. We'll see how we can realize new insights from our service metrics with tooling and automation.
Finally we'll examine data using Jupyter Notebooks as provided by IBM Bluemix's data science platform. I'll walk us through visualization of Forecast's output and we'll learn how Bluemix can make data science simpler.