Introduction
Today we are excited to announce a new category of log solution. LogAI is purpose-built for Network Operations, and purpose-built for real-time anomaly detection and automation. This is a totally new experience of log data, purpose-built for real-time Network AIOps of log data.
Many of today’s most pervasive log data solutions are focused on collecting as much data as possible, from as many log sources as possible, storing it for long periods of time, and providing search and query capabilities. Augtera Network’s LogAI is complimentary to these solutions and can even send them actionable insights.
LogAI has a different design-point with different outcomes. The goal of LogAI is to process log messages at the velocity they are received, immediately detect anomalies, and take customer-specified action including operation alerts, trouble-ticket creation, or auto-mitigation/remediation. This is a new class of log data solution, and a new experience for Network Operations teams.
In April of this year, Augtera announced our industry-first Natural Language Processing based Zero-Day anomalies for Syslog. Today we are announcing the full log data solution. In addition, we are announcing that we have proven our technology works for more than just Syslog. While our solutions remain Network Operations focused, we believe our technology is applicable to any log data.
Flexible Ingestion
Since our April announcement we have developed and used in customer networks, the ability to process more than Syslog. We now support JSON formatted log data, and JSON with Kafka. While Syslog is a pervasive approach to log data, Augtera customers have found the need to normalize and structure a variety of data sources into their own JSON-formatted logs. LogAI supports Network Operations teams who are doing this. We anticipate adding additional log formats and transports as needed.
Collective Learning
Many Network Operations teams have developed their own log processing rules, often using an approach like regex. There have been several challenges. Sometimes a rule cannot be implemented until a separate IT development team makes the change, the rate of learning log signatures can be slow, and managing the rules can also be cumbersome. LogAI transforms the experience of detecting and managing log signatures – the known knowns.
In response to the detection of log signatures through “classifiers”, the customer can specify numerous actions, including:
- Sending the message to the Network AI/ML pipeline
- Formatting some aspect of the log message – for example, changing the severity, adding meta data, adding IP information
- Publishing the message to a Kafka topic
LogAI manages a central repository of log signatures, for example anomalies, which is distributed to each LogAI customer, thereby accelerating the rate at which individual customers accumulate log signatures – we call this “Collective Learning”. In addition, when a customer determines a log signature, it can be applied immediately, without having to wait for any software development.
Zero Day Anomalies
The LogAI vision is for customers to detect incidents as soon as they happen, and optimally, prevent them from ever happening in the first place. Collective learning enables customers to have access to an ever-expanding number of log signatures. Complimentary to collective learning is the detection of rare and new log messages that often precede outages. Detecting these messages enables customers to prevent future incidents. This capability is based on Augtera’s industry-first high-performance, high-efficiency, purpose-built Natural Language Processing (NLP) implementation. Rare / new log message detection is now available for all supported log formats including Syslog, JSON, JSON over Kafka, and logs ingested through an Augtera API.
Conclusion
There are other capabilities that are part of the LogAI solution including message rate burst detection, metric extraction and algorithms, structured query, and integrations with Slack and ServiceNow. To learn more about all LogAI’s capabilities, go to https://augtera.com/logai
Today’s log data solutions serve a purpose. However, Network Operations teams also need the ability to mine the relevant needles in the rich log haystack, and take immediate action, including preventing future incidents. This is what LogAI was designed for. It is a new log experience.
To schedule an Augtera engineer consultation or demonstration: https://augtera.com/contact-us
