Upgrade to Pro — share decks privately, control downloads, hide ads and more …

Important Cloud Metrics To Monitor.

Important Cloud Metrics To Monitor.

If you've got no plans to monitor your cloud infrastructure then there's no need to go for cloud in the first place. Monitoring is a measure of how well internal states of a system can be inferred by the knowledge of it’s external outputs.

Avatar for Amos Chidiebube

Amos Chidiebube

August 24, 2017

More Decks by Amos Chidiebube

Other Decks in Programming

Transcript

  1. Speaker Bio ▪ Lazy DevOps Engineer at Shopp!ng. ▪ Lifelong

    Learner and teacher. ▪ Technical Writer. ▪ Aspiring philanthropist J.
  2. Disclaimer This talk neither endorses nor condemn any specific products,

    open source or otherwise. All opinions expressed are in the context of my experience so far.
  3. What is Observability ? In control theory, observability is a

    measure of how well internal states of a system can be inferred by the knowledge of it’s external outputs.
  4. How was monitoring and alerting done in the early 2000

    say 2006 ? When…… ▪ No cloud computing – AWS had only just launched – Google App engine and Microsoft Azure didn’t exist yet ▪ Few cloud applications – GMail was still in limited public beta – Google Docs was still in limited testing – Facebook opened for public access in September of that year ▪ No smartphones – Apple had not even unveiled the iPhone yet – Android didn’t become available until 2008
  5. “Our first alerting system was an engineer, who looked at

    graphs all the time; day and night; weekends. And he would start calling up people, when something looked wrong.” Emil Fredriksson, Operations director at Spotify (Slightly paragraphed)
  6. Why do we need to measure and monitor ? Because

    not only is the old saying “If you can’t measure it, you can’t improve it” true, but visibility into your metrics allows you to identify WHERE you can make the easiest and most impactful improvements.
  7. Can you share with us how you manage logs and

    monitoring from all these services ?
  8. 3. MONITOR END USER EXPERIENCE Do you think I care

    if you used Salt or Chef for Configuration Management ??
  9. 5. 4XX and 5XX Errors You should have systems that

    doesn’t just interpret these errors but points you to where they can be fixed.
  10. 6. SCALABILITY How many of us have had our servers

    crash due to a hike in the number of users ?
  11. 8. DATABASE Queries! Queries!! Queries!!! A database’s work is running

    queries, so your first monitoring priority should be making sure that your database is executing queries as expected.
  12. The Future of Alerts and Monitoring AI automates all the

    “heavy lifting”—all the discovery and analysis that take teams of experts hours or days, AI does in milliseconds—to proactively identify problems and pinpoint the underlying root cause. ▪ Self-healing troubleshooting. ▪ Accelerated time to delivery. ▪ Future-proof scalability. https://www.dynatrace.com/platform/artificial-intelligence/ The future is now…..
  13. Key Takeaways ▪ Almost all system ends as legacy.. You

    probably haven’t installed your last monitoring system. ▪ A technically superior solution is worthless, if your users don’t understand it. ▪ Avoid hype driven development. Be slow to jump into the next new and shiny thing in town… ▪ Monitoring is suppose to be reliable, but not necessarily accurate. ▪ Monitoring is also cost intensive. Take it one step at a time. ▪ Build your systems like it’s serving users all over the world.