Skip to main content

Glasnostic Documentation Home

Glasnostic UI Cloud native application landscapes are a big problem for companies that want to innovate, because as their production environments grow relentlessly, so too do their operational challenges.

At the same time, the proliferation of connected services means interactions on the network have become the unifying plane for operational control.

Glasnostic builds on this trend to help companies operate with excellence and focus on innovating. It is a dedicated control layer that makes interactions in cloud native application landscapes visible and controllable, so engineers can see what is going on, maximize reliability and enforce security.

Glasnostic is agentless, inserts cleanly into any existing environment—from mainframes to serverless—and integrates seamlessly with your existing tools.

Why Glasnostic?

Production is hard.

Companies are running more cloud native applications in more places, and these applications depend on more services, than ever before. This makes cloud environments unpredictable, and this unpredictability can’t be "engineered away."

You deploy applications every day, yet something else breaks seemingly every other time. You go back and fix some code, but by the time the fix is deployed, the situation has changed. You spend considerable time manually instrumenting all services, but you can’t see the forest for the trees. And configuration drift makes it impossible to give basic security guarantees.

This inherent unpredictability of cloud native application landscapes must be controlled at runtime.

Who should use Glasnostic?

Glasnostic is designed for DevOps/DevSecOps, site reliability engineering, platform and security teams that want to move beyond fighting fires and reacting in an ad-hoc and bespoke manner that is tailored to specific stack and control abilities.

Why you need Glasnostic

Know what’s going on in real-time

  • See which systems are currently active, how they are interacting and how interaction characteristics change under load.
  • Understand the baseline behavior of your app.
  • Get capacity insights to scale better.
  • Detect anomalies and know when and where to act.

Become more resilient

  • Reduce deployment risk by querying the environment, steadying interactions pre and post deployment or quarantining changes.
  • Apply effective control primitives to rapidly detect and respond to degradations. Minimize mean time to recovery and the severity of degradations.
  • Automatically tune resilience and optimize for service-level objectives. Recommend actions when needed.

Assure essential security

  • Rapidly detect and respond to suspicious interactions.
  • Auto-microsegment dynamic environments.
  • Enforce universal encryption and authentication.

Glasnostic components

Glasnostic consists of a data plane and a control plane component.

The data plane component implements the control layer. It is what’s known in the networking community as a "bump in the wire" — a transparent processing stage that is responsible for measuring and controlling service-to-service interactions.

The control plane collects metrics from the control layer, processes them to generate operational insights, and relays operational control information back to the control layer.

What Glasnostic is not

Glasnostic provides businesses with the real-time operational control they need so they can focus on building innovative services. Glasnostic is not an application performance management, monitoring or observability solution, but it does play well with these tools.

Glasnostic does not specialize in providing deep transactional or "high-cardinality/high-dimensionality" metrics. It instead focuses on the key “golden” signals that characterize service-to-service interactions.

Although Glasnostic shares some similarities with service meshes, notably the existence of a control and a data plane component, it is also not a service mesh. Unlike service meshes, Glasnostic does not establish connectivity between services but instead controls the characteristics of interactions between arbitrary sets of service endpoints once connectivity is established.

Finally, Glasnostic is not a chaos engineering platform. Chaos engineering platforms are built to inject failure into an architecture to learn from the outcome and subsequently improve the resilience of the system — something organizations like to shy away from in production. Glasnostic instead focuses on injecting resilience into production environments.