How does it work ?

Kalc kubernetes simulator checks if a failure condition is reachable within your current cluster setup. Entirely model-based, it does not disrupt cluster operation and produces a report with full root cause event chain.

Step 1: Send your k8s resources manifest

Do `kubectl get -o=yaml` on your resources with a bash one-liner. Upload either programmatically or manually. Mix in your changes or additional `kubectl` commands.

Step 2: Get predicted failures report

Understand what will happen if nodes get locked up, whether all the scaling will work, instantly know if the proposed change does not evict pods, break labels or starve resources.

Step 3: Mitigate failures waiting to happen

Read the YAML report manually or programmatically to identify root cause of issues. Add more constraints to the cluster. Repeat.

Step 4: Monitor and visualize

See your current risk trends on a chart. Set alerts. Detect broken updates and stop the pipeline before failure happens. Get more insights automatically with every new update of Kalc kubernetes model.

See Full Process

The problem

The rapid adoption of Kubernetes has led to an increase in outages covering the entire company operations.

This problem is so severe that many companies have publicly shared their stories: Target, JetStack, Civis Analytics, BlueMatador, Grafana, Moonlight, DigitalOcean, MonzoBank, Zalando, Pivotal, etc. The existing solutions are clearly not helping enough to minimize failures.

Our solution

We approached this problem by replicating Kubernetes' behavior in AI. Now we can train our Kubernetes AI with the most common failure scenarios and let developers test their configs against these scenarios.

This leads to minimized outages, increased deployment pipeline stability and visibility improvements, thanks to our new risk metrics in your dashboard.

Kalc vs alternatives

Chaos Engineering

Kalc is a less disruptive alternative to chaos engineering. Kalc replicates the current cluster environment in AI and provides autonomous checks and config validations without chaos in your test or in the production environment.

Policy-based control

Rather than limiting developer changes, Kalc AI can test the effect of each change on the AI-replicated Kubernetes environment and help you maximize security and efficiency.

Manually keeping up

The complexity of home-grown validation tools in the microservices space has gone way beyond any human intellectual capacity. Our intelligent risk mitigation solution can keep every version/compatibility point in mind for everyone.

After-the-fact analysis

The production engineering team is usually assigned to after-the-fact outage analyses and debugging to design prohibitive policies all over the cluster that inevitably lead to slowing down the process and reducing the benefits of a new architecture.

Wait for a failure to happen

Many companies with early-stage Kubernetes deployments have not yet adopted failure prevention strategies. Regardless of your stage, please do not hesitate to contact us so we can help you get started.


Our AI-first approach covers substantially more risk mitigation scenarios than possible to define manually.

Protect your cluster from evictions, freezes, and failures to perform scaling due to missing constraints/nodes that have requested resources.

Protect your cluster from hitting resource limits. Including setting incorrect constraints. Our model of RAM/CPU resources covers all possible variations.

Protect your cluster from cron jobs with no/incorrect concurrency policies, combined with all of the eviction/freeze scenarios with the above.

Protect your cluster from idling pods on nodes with a history of consuming lots of RAM or when anti-affinity is not defined. We have a full model of restart policies and real resource consumption behaviors.

Protect your cluster from incorrectly typed labels for services. Your request destination is always safe.

What People Say About Us

"The capabilities of the AI planning in Enterprise is unmatched, and the guys at Kalc have proven that automation with commercial AI planning is unavoidable."

Mike D. Kail
CTO at Everest, former Yahoo CIO

"This could be a "must-have" feature for every Kubernetes installation."

Former President and CEO at Lucent Technologies and ex EVP at Juniper Networks

Kalc can now automate Cloud policies, Kubernetes validation, and deploy features that are up to 10 times faster. Very impressed.

Michael V Dvorkin
Distinguished Engineer at Cisco

Minimize Cloud Native risks with Kalc

Thank you! Your submission has been received!
Oops! Something went wrong while submitting the form.