Algorithmic systems monitoring
Continuous and automatic analysis of full stack log data that adapts to your evolving data center.
A new way of analyzing streaming data
Unomaly analyzes data upfront and remembers everything it has seen.
A new way of highlighting and revealing must-see data
Unomaly universally detects and reveals non-normal data without prior knowledge of incidents.
A new way of reducing, storing, and scaling data.
Unomaly efficiently reduces the amount of data to analyze, store, and transmit.
our difference
What is algorithmic monitoring?
Universal data tokenization
All software systems produce log data that explains what it is doing and when it is failing. Unomaly specializes in ingesting and analyzing any full stack, cross-platform raw data in real time, without concerns for format, structure, or volume. This allows Unomaly complete coverage of full stack environments.
Data privacy and integrity
Today, working with data comes with responsibility. Unomaly has been built from the ground up with data privacy and compliance in mind. As it reduces data and only keeps the vitally important, it helps you avoid harboring sensitive data that you later need to clean. It is designed to profile software systems — never people.
Multi-type anomaly detection
Incidents create anomalous data as they develop from cause to impact. Unomaly is designed to detect different types of anomalies in real time - such as new events or changing structures and parameter. By being able to isolate all relevant data, Unomaly can paint a complete picture of every issue.
Rapid progressive learning
The modern data center is constantly changing. Unomaly analyses streaming data in real time and progressively learns normal. It recognizes normal patterns in events, structures, parameters and frequencies and update models as new data arrives. Unomaly automatically stays up to date and relevant.
Data reduction pipeline
When dealing with data transmission, storage and processing quickly becomes a problem in itself. Unomaly is designed to reduce data while analyzing it. By reducing repeating data, it can focus on storing and transmitting data that actively contributes to new information. It protects your infrastructure and team from spikes and the burdens of scale.
Flexible and self-managing
Software systems should not need continuous maintenance. Unomaly was built to work on its own, with very little care needed. It deploys close to where you run your software, automates analysis and effectively reduces data automatically.
our technology
How Unomaly works
Deploys instantly
Download and install Unomaly in 30 minutes on physical hardware, virtual machines, or cloud and hybrid environments.
Start with a single instance or with multiple instances in a distributed deployment. Unomaly is flexible and scaleable with your growing data center.
Receives real-time data streams
Unomaly analyzes all the data you forward to it, regardless of format and structure. For the most part, you simply configure your systems to forward their log data. If needed, you can forward the data with an agent or use one of our integrations for third party aggregators and log management systems.
Continuously learns and profiles systems
Unomaly creates profiles for each system that sends data to it. Depending on your environment, Unomaly can have a baseline of the systems in a matter of days or a couple of weeks. Analysis and learning never stops. As long as the systems continue to send data, Unomaly continues to learn and update the system profiles — adapting its understanding of normal to your evolving data center.
Automatically reveals anomalies as they happen
As incidents, problems, or manipulations happen, systems react by creating new data. Unomaly automatically reveals this data in real time so that you can detect everything from subtle problems to complex outages as they happen.