On-premises edition now available

Learn More

Zebrium is no ordinary log manager!

It uses machine learning to find root cause


While Zebrium provides all the log management features you'd expect (aggregation, search, filtering, etc.), what sets it apart is its use of machine learning. Automatically see root cause without hunting, proactively detect new failure modes without building rules and enjoy fully structured logs without defining parsing expressions.

Use Zebrium as your primary log manager or to augment log mangers like the Elastic Stack.

All the features you'd expect

Broad platform support
Kubernetes, Linux, Docker/ECS, syslog and CloudWatch. Most other platforms via Logstash (see integrations).

Automatic collection and storage in a scale-out MPP relational column store with cloud-scale scalability and query performance.

By logtype, severity, event type, labels (host, container, app, etc.) and any user defined expression. Views can be saved and reloaded.

Regex Search
Search and filtering supports full PCRE2 regex syntax.

Role-Based-Access-Control (RBAC)
Users are assigned roles and can be granted access to one or more deployment groups (e.g. test and production).




log management features you would expect

one click charting p

And ones that will surprise!

ML-based structuring
Machine learning is used to automatically structure log lines of any format without requiring manual parsing expressions.

Visualize and navigate user-defined sets of events. Stacking maps allows you to see event correlations.

One click charting
Since all event variables are automatically extracted, chart any string, numeric, IP address, etc. with just a click. 

Search for answers
If you see a cryptic log event, automatically see Google or Stack Overflow search results.

Share your entire session, including filters, maps and current position for easy collaboration.

ML-Powered Root Cause Analysis

Proactive root cause reports
When looking for root cause, instead of hunting through logs, immediately see root cause by selecting the relevant proactive root cause report.

On-demand root cause scan
The "scan for root cause" button lets you perform on-demand scans for root cause around a specified time.

Core and related events
Root cause reports initially contain a core set of log events and correlated metric anomalies that describe what happened. If you require more detail, the ML can also pull in additional correlated errors and anomalies.

Interactive navigation
Root cause reports can be viewed on their own, or in the context of surrounding aggregated log lines. 

rca drill down

create ml alert rule

ML-Powered rules and alerts

ML-generated rules
When our ML detects a problem, you can choose to be alerted on future occurrences without building rules. Great for catching new/rare failure modes.

User-defined rules
You can easily define a set of log events and conditions, and be alerted whenever they occur. 

ML augmentation
User-defined alerts often only detect symptoms. Zebrium can augments these alerts by using machine learning to uncover details of root cause.

Get started in < 5 minutes

Just install one of our lightweight log collectors or fork a copy of your logs using Logstash. Zebrium works with any app - no parsers, code changes, rules or config needed.

get started in less than 5 mins

Experience Zebrium for yourself

Try with a sample microservices app

Our simple guide and video shows you how to setup a minikube Kubernetes cluster, install a cloud-native app, break the app and then see Zebrium detect root cause. The entire process takes 20-30 minutes.

FREE SIGN-UP        




Try with your own data

It takes less than 5 minutes to sign-up for an account and install our log and metrics collector. The rest is automatic. For best results, you will need to experience a real or induced failure in your environment.