A presentation at ProgNet in in London, UK by Jessica White
Measuring Impact
Guiding Decisions
Detecting/Resolving Issues
Improving Communication
Speaking the same language, developing a glossary.
Rather than using the word “monitoring” define the lower level purposes and what you are actually wanting to do.
There are a number of types of measurements that can have an initial impact
What goals are your department/company trying to achieve? What impact are you having on these goals. It’s nice to consider KPI’s at feature launch meetings.
Internally and externally. This might be service level agreements (SLA’s). Not meeting contractual agreements can have financial consequences.
Is your system is behaving as expected
-Diagnosis
What is useful from bringing down the time to address issues.
There is no “one size fits all” solution. Use the right tool for the right solution.
Who are the users? How will they interact with this information? Do they have experience of this tool or will they need training?
Who needs to be able to access this information? Do you want to lock down who accesses this information or for how long?
Cost
Does it meet the need
Does this tool do the type of monitoring you want in the way you need (i.e. real-time data)
Does anyone in your team have experience with the tool? Is training required? How much will this edcuation cost (including time)
Everyone can understand and react to the information on a car dashboard, ours should be equally as transparent.
API patterns:
Service patterns: -USE.
Can have these inbuild as part of your code generation.
On-call teams will have a better and easier time.
Cultural investment is the most difficult. Making monitoring a first-class citizen
Knowing your baseline, having an effect, seeing its impact.
Makes backend work more understandable at a business level too!
“We need to monitor the things!” they cry, “We have a tool and it’s going to change our world!” they scream.
“What do you need to measure and why?…” you ask.
Silence follows.
In this talk, Jessica will explore the different aspects of monitoring from log analytics to business metrics. What steps do you take in getting cultural investment for monitoring? How do you determine what you need to measure? What patterns can you follow for this work? How do you keep the investment and enthusiasm for this work going?
This talk is for people of all walks of life in technology. The aim is to provide examples and guidance on what you measure, why you should be excited about it, how to sell this to your business and how to keep these measurements effective and useful.
The following resources were mentioned during the presentation or are useful additional information.
In the "Monitoring Distributed Systems" Chapter of the Google SRE book
The role of autonomy, mastery and purpose in drive and motivation
My own blog
Here’s what was said about this presentation on Twitter.
for free. You
can too.