Machine log roller Sumo Logic expands to swallow AWS data-watching, DevOps performance
I always feel like ... somebody's watching me (and they're doing it continuously)
Log management and analytics biz Sumo Logic has tweaked its technology platform to help sift through applications and infrastructure that run on AWS as well as keep an eye on software development cycles.
Rather like Splunk, Sumo Logic, born in 2010 in Redwood City, California, started out helping manage and analyse telemetry data from physical machines and expanded to event logs in computing environments. It now applies analytics to machine data churned out by IT gear, and pinpoints issues that could crash applications as well as other anomalies. Its tech is based on Apache Lucene, the open-source search engine software library, which is also the basis for Elastic Search, and includes other proprietary and open source technologies.
The new features form part of the Sumo Logic Observability suite, the idea being to help developers and operations people keep track of software during both build and deployment phase as well as lock down the raw observability data in case it contains sensitive info.
Both rely on the vendor's Continuous Intelligence Platform, an analytics system that aggregates anonymised information from service deployments to establish baselines for the performance allowing users to compare their technology stack against infrastructure operated by other organisations.
In September last year, Sumo pushed out wares to help manage containers built with Docker and Kubernetes. Now, it is going a step further, and muscling into the entire cloud environment, starting initially with AWS.
What it calls the Sumo Logic AWS Observability solution uses AWS CloudFormation infrastructure management to find all the users’ accounts and services, pulling that data and configuring it into a single dashboard view, said Ben Newton, Sumo Logic's director of “field evangelism”.
The log roller collates and analyses telemetry data from "popular AWS services like Application Load Balancer, Amazon elastic Cloud Compute (EC2), Amazon Relational Database (RDS), AWS Lambda, Amazon DynamoDB, and Amazon API Gateway" in order to "detect anomalous events" and "enable root cause analysis".
“One of the things that AWS users have to deal with is that they might have dozens of accounts or different development teams, spread across different accounts, in different regions, running different services, and even within AWS you're having to switch around a lot. It's really hard to get a single view,” he said.
Using the Sumo Logic Continuous Intelligence Platform, with a bit of machine learning, the solution also compares the user’s performance with other AWS users, and helps understand the reason for poor performance using a root cause explorer tool.
But the company’s plan is to provide “observability” through the whole cycle: software development to deployment and management.
On the development side, it has taken KPIs developed by the DevOps Research and Assessment (DORA) organisation, co-founded by devops guru Jez Humble, and built them into a performance dashboard. The Sumo tool also supports GitHub, Jenkins and PagerDuty, helping benchmark performance and understand where to improve.
The platform has also been extended to track microservices and transactions in distributed application environments in the cloud.
As with many other vendors, Sumo has found itself starting with AWS because the Bezos giant is everywhere, and that also applies to DevOps, natch: 60 per cent of its customers use the cloud giant’s infrastructure and it has built its own system on AWS.
“One of the reasons why we decided to build our own distributed tracing system was because we need to solve specific problems internally and there just wasn't a technology out there that we could use. All this stuff we’re releasing is stuff we dog-fooded,” Newton said, unnecessarily verbifying the well-worn idiom coined in the 1980s IT industry.
Sumon Logic said it intended to extend the platform to support GCP and Azure in the future. ®