Skip to main content

Monitoring Distributed Systems in Real-time with Riemann and Cassandra

Patricia Gorla (The Last Pickle)
Average rating: ***..
(3.62, 8 ratings)
Slides:   1-PDF    external link

Cassandra is at the forefront of the distributed database push. Many companies are turning to Cassandra for its low-latency writes and configur

But with each new server comes monitoring responsibility. How do we monitor a cluster with many nodes? What statistics do we want to see at-a-g?

Additionally, these metrics that are being created also need to be stored. How should the data points be kept? What is a useful level of granu

Enter Riemann, a stream-based monitoring system built to watch events from all across the network, in real-time.

Although the talk focuses on Cassandra-specific statistics, these lessons in distributed monitoring can easily apply to other kinds of distribu

This talk will cover:

  • The difficulties with distributed monitoring, and current monitoring patterns.
  • Setting up a monitoring system over a multi-node system.
  • Choosing key statistics and dashboards.
  • Understanding the limitations of monitoring.

At the end, the participant will have an understanding of what is needed to effectively monitor their own distributed system.

Photo of Patricia Gorla

Patricia Gorla

The Last Pickle

Patricia Gorla is an Apache Cassandra Architect at The Last Pickle, a Cassandra consultancy.

She has been involved in all aspects of software development, from server administration to application development, and from data analysis to data storage.

She has worked with companies and governmental entities on all aspects of data migration to non-relational data stores, and training the technical teams on the new architecture.

She helped the US Patent and Trademark Office ingest more than 6 million patent documents and images; architect secure search systems for a large mortgage insurer; and introduce Cassandra to a digital marketing firm’s data pipeline.

Prior to architecting databases Patricia focused on the analysis and visualization of data.

Patricia speaks often at conferences and meetups such as O’Reilly’s StrataConf + Hadoop World, the Datastax Cassandra Summits, and local user groups. She was also voted a DataStax MVP for Apache Cassandra by the community.