Rootconf Hyderabad edition

On SRE, systems engineering and distributed systems

Great Expectations : Observability in asynchronous systems using temporal logic

Submitted by Aditya (@adityabharadwaj) on Mar 5, 2019

Section: Full talk of 40 mins duration Technical level: Intermediate Section: Full talk (40 mins) Category: Monitoring and logging Status: Rejected


This talk will focus on building observability across microservices.
The main focus will be on
1. Challenges faced while monitoring flows in asynchronous systems
2. Building on temporal logic constructs to verify system objectives
3. The concept of an expectation applied to detect failures in critical flows


  1. Challenges of monitoring asynchronous systems
  2. What is an expectation ?
    • Parts of an expectation
    • Basics of a state machine
    • Basics of linear temporal logic
    • How it all fits together
  3. Instrumenting the Expectation engine - with a description of components and reason for choice
  4. Variations of the basic expectation in context with real world examples
  5. Success stories where other monitoring systems would be insufficient
  6. Challenges faced and how we overcame them


To get the most out of the talk, experience in the following will help
1. Basic understanding and use of monitoring tools (Any kind).
2. Understanding of event sourcing

Speaker bio

I work at Capillary as a Principal Engineer and am actively involved in scaling our platforms. For over a decade i’ve been working on systems at scale, watching many of them closely as they grow, trying to eliminate every last bit of inefficiency. In the process, i’ve built up a good understanding of systems and pros and cons of various monitoring tools and processes.



Preview video


{{ gettext('Login to leave a comment') }}

{{ gettext('You need to be a participant to comment.') }}

{{ formTitle }}
{{ gettext('Post a comment...') }}
{{ gettext('New comment') }}

{{ errorMsg }}