Dynatrace musters cross-cluster application fluster buster
Dynatrace is a company that specialises in the monitoring and control of enterprise-scale application performance, the underlying infrastructure layer and the experience of users.
It does this at scale, running what it calls a Dynatrace cluster as a coalesced instance of software intelligence designed to provide application control features such as tracing, analytics and management.
Question: So what’s better than a cluster?
Answer: Whether it’s the chocolate-coated nutty kind or the application observability kind, it’s a double cluster.
Dynatrace has now doubled the capacity of a Dynatrace cluster, now scaling to 50k hosts while maintaining system performance.
Cluster luck!
In addition, Dynatrace now supports the clustering of clusters, including cross-cluster distributed tracing, analytics and management to deliver AI-powered observability, automation and intelligence for customers operating large multi-cloud environments.
It’s true to say that web-scale environments were (comparatively speaking) a rarity a few years ago, but they are becoming commonplace as enterprises shift from static, on-premises datacentres to more dynamic multi-cloud architectures with highly distributed microservices workloads.
Dynatrace suggests (because this is the market it operates in) that companies in industries from financial services, healthcare, eCommerce etc. are growing their environments beyond their current monitoring systems’ ability to keep up.
In addition, growth in complexity is outpacing their teams’ ability to identify and understand anomalies and correct performance and availability issues in a timely fashion.
Scale-out architecture
Dynatrace claims to be the ‘only solution’ that has the automation, intelligence and scale-out architecture needed to deliver the observability and answers that today’s enterprise clouds require.
“Driven by the shift of their datacentres to the cloud and growing cloud-native workloads, it’s not hard to imagine hundreds, even thousands of web-scale enterprise clouds in the not too distant future,” said Steve Tack, SVP of product management at Dynatrace. “We reinvented our platform several years ago to provide high-fidelity observability, smart automation and real-time intelligence. We continue to push the boundaries on scalability and robustness.”
Features include automated discovery and instrumentation through single agent instrumentation that automatically and continuously discovers all microservices, components and processes across the full cloud stack – networks, infrastructure, applications and users – and continuously maps dependencies in real-time.
There is also high fidelity distributed tracing and cross-cluster analytics through a single management dashboard regardless of cluster location.
The Dynatrace explainable AI engine, Davis™ processes billions of dependencies in real-time, delivering the ability to ‘go beyond’ (so says the marketing info) metrics, logs and traces to provide precise answers to issues at scale.
Finally, there is also role-based governance for global teams with a branded Management Zones tools. In this way, Dynatrace says it enables fine-grained access across applications and zones for secure, distributed management of shared cloud environments by multiple teams.
No chocolate clusters were hurt (or consumed) in the production of this story.