After spending time in Budapest and Barcelona (see my recap posts here and here to learn more) my next trip brought me to Toronto, Canada. This blog is a short recap and reflection of 4 days in a city that just came out of a very long winter and that is getting ready to host a couple of games of the upcoming Fifa World Cup. And while the kickoff of the world’s largest sports tournament is still a couple of weeks away we did a “Kick Off” for the global Dynatrace Innovate Roadshow event series.
Dynatrace Innovate Roadshow is a great networking event where together, we explore what’s now and shape what’s next in observability, security, AI, and beyond. I was invited to share the stage with my teammate Adriana Villela, demoing how Dynatrace is redefining Cloud Platform Operations. To bring it to live we did a little role play – like what we are doing in Adriana’s “Did you know that Dynatrace can do that with OpenTelemetry” YouTube series.
Redefining Cloud Platform Observability and Autonomous Operations
Adriana – who has spent 16 years as a Java Developer in Financial Services – walked me through the unsustainable and inefficient way to resolve complex issues at scale in modern cloud native architectures when you don’t have unified observability. Having to switch between environment specific tools, coordinating with different teams, connecting data manually and trying to find root cause and solution was always a painful process for her:
I then walked her through the new Dynatrace Clouds App with its unified approach to monitor services, logs, metrics, traces, configuration in a single place. If you haven’t tried it check out the Clouds App on the Dynatrace Playground:
Getting more insights, making it easy to analyze it in a single spot and having default best-practice alerting is however only one part to making the life of Cloud Platform Operation Teams easier.
In our demo we showed how a Dynatrace Intelligence detected Incident triggers an Agentic Workflow that facilitates a closed loop GitOps driven problem remediation, involving several specialized agents that looks like this:
To recap this best practice agentic workflow: It gets triggered when Dynatrace Intelligence detects an anomaly (technical issue, business impact, security vulnerability …). Then:
- Dynatrace runs the root cause agent and opens a GitHub issue with all details
- Azure SRE Agent captures additional environment specific fix recommendations
- CoPilot to prepare a Pull Request based on all findings to fix the issue
- Human Operator validates and approves the Pull Request
- Automated creation of Postmortem Notebook to improve future incident responses
The benefit of this process is that all relevant information, and all needed tools and agents are collaborating through a Git issue whose lifecycle is defined by the GitOps process. Many of the otherwise manual steps are automated through agents while the human is there to supervise, validate and approve!
If you now think: “Andi, that sure looks great in a demo, but I doubt this works in real life!”
Well: Just before our presentation we heard from other Dynatrace users how they have integrated Dynatrace into their agentic workflows. A great example was presented by Telus, Canadas largest telecom provider. Dana Harrison and Kulvir Gahunia got on stage and presented IRIS – their Agentic System that automates many of the operational and SRE tasks for their engineering team. It seamlessly integrates into their software development workflow, provides self-service insights into observability and is accessible from the tools that their engineers use for their daily work such as Slack or VSCode. They have also pioneered the integration of observability into their coding agents via the Dynatrace MCP and now dtctl.
More users presented their stories such as the power of Real-Time Business Observability at Interac, Modern SRE at Intact and insights into how observability transforms day2day operations at Enbridge or iA Financial Group
To wrap things up: Meeting users, get hands-on experience
The day after Innovate the local Dynatrace team also offered a Hands-On Training opportunity. While I spent most of the day at KCD Toronto (a separate blog will follow), I had an hour over lunch to join the Dynatrace community that got trained in the latest use cases such as Business Observability, Autonomous Operations or Turning Data into Insights.
I also took the opportunity to hand out another copy of my latest “Observability in the AI-Native Era” book to Srilakshmi, Platform & Cloud Architect. Hope you enjoy it!
Also a thanks to our local account team who organized a handful of 1on1 sessions with some of the SRE and Platform Engineering teams to discuss how to use observability to increase resiliency of modern cloud native platforms!
From Toronto to Singapore
While there are many Dynatrace Innovate Roadshow stops in APAC, EMEA, LATAM and NORAM my next Innovate stop will be in Singapore. On July 22nd I will connect, share and innovate together with our community in that region. I will also have additional stops in Indonesia, Thailand and Malaysia to meet with our local users to deliver workshops. If you live in those regions make sure to reach out to your local Dynatrace team to sign up for all the activities
