Sumo Logic and the Challenge of Full-stack System Management

cloud_in_hand.jpgModern application development and deployment is built around the principles of virtualization, agility, scalability, and manageability.

  • Virtualization, now best addressed through containerization and microservices.
  • Agility helps guide efficient resource utilization, including people.
  • Scalability speaks to matching capacity to demand while embracing the principle of immutable infrastructure.
  • Manageability stresses the importance of securely supporting policy-based orchestration and support across the lifecycle for the above tenets.

When thinking about the above tenets, manageability is by the far the most challenging. This is because the overall increase in infrastructure flexibility and responsiveness (virtualization, agility, and scalability) is forging a new dynamic form of IT, in which new approaches to managing, trouble-shooting, and monitoring systems and applications are now required. Software-defined storage and networking and the availability of hyper-converged systems mean a full-stack approach to system management is now a necessity. This also means that the foundation of full-stack system management requires comprehensive real-time continuous data collection.

Sumo Logic is one of a new breed of system management vendors that is working to change the status quo by leveraging a combination of knowledge and services derived from the evolution of mobile, cloud, social, and big data. The founding principle of Sumo Logic is real-time continuous data collection. Approaching the challenge from that angle is itself a new tactic.

Sumo Logic's data collectors generally link to log data being generated across a wide variety of application, server, storage, and network resources. The sheer volume of information means their approach is only possible with elastic log processing. A SaaS-focused approach gives Sumo Logic comprehensive data describing IT operations several orders of magnitude beyond what is normally available for system management activities.

This offers extensive information from which to monitor, detect, alert, investigate, report, and optimize system behavior. The volume, variety and velocity of the data collected necessitates the use of big data management and analytics techniques, including machine learning. This also opens the door for even more sophisticated analytics involving prediction, optimization, and adaptation.

The SaaS approach based in the public cloud also takes advantage of the scalable infrastructure provided by this environment. Sumo Logic is an example of how big data can be leveraged to manage data and can apply machine learning to perform real-time forensics and establish how anomalies impact system performance. Enhanced context, in the form of a much wider variety and volume of resource data provides a faster and more reliable foundation for enabling full-stack system management.

There is significant knowledge to be gained from monitoring a vast network of data sources, event, and relationships. I recently elaborated on the topic of real-time continuous data collection in a white paper commissioned by Sumo Logic, and you can take a look at that info on their site.

Decision Analytics Report

Topics: Cloud Services & Orchestration