Observability: Key to Faster Deployments
Key Points
- Deployment speed is a key metric of success, so organizations should measure the frequency of updates over a time period rather than the days between releases.
- The DevOps workflow consists of eight cyclical steps (plan, code, build, test, release, deploy, operate, monitor), and any slowdown in a single step throttles the entire pipeline.
- Observability extends traditional monitoring by turning raw visibility into actionable insight and tying together infrastructure and application data, making it essential for all stakeholders.
- Automating observability—and eventually all eight DevOps steps—prevents bottlenecks, improves software quality, and boosts customer satisfaction.
- Real‑world results show dramatic acceleration, such as a bank cutting its product‑to‑market cycle from 10‑12 months to just two weeks after adopting automated DevOps and observability practices.
Full Transcript
# Observability: Key to Faster Deployments **Source:** [https://www.youtube.com/watch?v=x6UCODfHerw](https://www.youtube.com/watch?v=x6UCODfHerw) **Duration:** 00:03:11 ## Summary - Deployment speed is a key metric of success, so organizations should measure the frequency of updates over a time period rather than the days between releases. - The DevOps workflow consists of eight cyclical steps (plan, code, build, test, release, deploy, operate, monitor), and any slowdown in a single step throttles the entire pipeline. - Observability extends traditional monitoring by turning raw visibility into actionable insight and tying together infrastructure and application data, making it essential for all stakeholders. - Automating observability—and eventually all eight DevOps steps—prevents bottlenecks, improves software quality, and boosts customer satisfaction. - Real‑world results show dramatic acceleration, such as a bank cutting its product‑to‑market cycle from 10‑12 months to just two weeks after adopting automated DevOps and observability practices. ## Sections - [00:00:00](https://www.youtube.com/watch?v=x6UCODfHerw&t=0s) **From Deployment Frequency to Observability** - Chris Farrell argues that success should be measured by update frequency rather than days between releases, outlines the eight‑step DevOps cycle, and urges shifting from traditional monitoring to observability to accelerate application delivery. ## Full Transcript
[Music]
when your application is your business
speed of deployment ends up being a
measure of success because deployment
frequency ends up meaning update
frequency an update frequency means that
you can react to markets react to needs
react to your users it is an absolute
proxy for the quality of your
applications
I'm Chris Farrell and I'm the vice
president of value services at IBM
you have to flip how you talk about
deployments instead of talking about the
number of days between a deployment you
should be talking about the number of
updates per time period the shorter the
time period the more you're moving up
performance of speed the devops process
is a cyclical set of eight steps
planning coding building testing and
then release deploy operate and monitor
when you have one of the eight steps
that slows things down that pipeline
slows down what if you were to design
your it devops process for a new company
what would you automate to make better
predictions and accelerate application
delivery I would focus on the last step
I would get out of the traditional
monitoring space and move into
observability
[Music]
there's gaps that have occurred within
monitoring monitoring tools provide
visibility
observability platforms take the
visibility that you have and deliver an
understanding of what it means
the other thing that I like about
observability is that it ties
infrastructure and applications together
observability is not just built for
itops to get visibility of applications
and then solve problems it's actually
built to help everyone that has a stake
in applications see data that they need
to see to understand what they need to
do to change it the one piece of advice
I give is get to observability as soon
as you can some point in the future your
classic monitoring solution is going to
have a catastrophic failure and how
catastrophic it is to your business just
depends on when and where it happens
[Music]
the need to go to observability is
absolute
and it's got to be automated because if
you don't automate it you slow down when
organizations start to automate all
eight steps of the process they can
start to see higher quality and better
customer SAT but my favorite has to be
speeding things up
one example that I saw specifically was
a bank it would take them approximately
10 to 12 months to have an idea of
product before it was live
once they got their new devops processes
they changed that time frame from 50
weeks to two weeks
absolute direct results of success in
the marketplace