Sergey Nivens - Fotolia
As IT monitoring systems gobble up huge swaths of data, enterprises hope they'll regurgitate more refined hints to help troubleshoot the systems they oversee.
Datadog is the latest vendor to widen its view of IT systems data with this week's acquisition of Logmatic.io, a French log monitoring and analysis company it snapped up for an undisclosed sum. Datadog also joined the full-stack monitoring trend among IT monitoring systems vendors earlier this year with the launch of application performance management features for its platform.
This trend reflects the desire of enterprise IT pros to simplify the ever-increasing complexity of IT environments that now comprise many layers of abstraction and automation from a plethora of different vendors.
"Just about every CIO I've talked to in the last three years says they'd love to rationalize their vendor portfolio," said Stephen Elliot, analyst at IDC in Framingham, Mass. "But there are a lot of politics and subject matter experts beneath them, and for various reasons some organizations have massive tool investments -- it's not easy."
Datadog customers said they prefer a single shop that combines IT monitoring systems rather than cobble together information from many tools.
"Right now our logs are split over all sorts of different systems, but being able to keep them in one place would be a huge boon," said Calvin French-Owen, co-founder of San Francisco-based data analytics company Segment.
In the past, Segment set up log monitoring tools, such as LogDNA and AWS CloudWatch logs, to send IT operations pros alerts on upticks in critical or error log lines that come from different services.
"That has proved immensely valuable as a 'line of first defense' when it comes to issues cropping up," French-Owen said.
Less data sifting, more action the goal for IT monitoring systems
Beyond log analysis, French-Owen said he would like to see Datadog delve further into machine learning and artificial intelligence (AI) to offer more root-cause analysis and automated troubleshooting features. AI is also a popular trend in IT management systems among DevOps vendors, from ServiceNow and Electric Cloud to PagerDuty and Red Hat. Datadog already has some anomaly detection algorithms for both infrastructure metrics and application traces. The addition of log data will further flesh out AI and machine learning capabilities, the company said.
"If they are collecting application logs, they can start tying together the metrics from an application to its output," French-Owen said. "Long term, I see Datadog adding value as the nervous system around what's happening in production."
French-Owen said he wants to see more correlation between various metrics, and some sort of 'auto-detected' root cause analysis generated by machine learning and AI pipelines.
"Eventually, they might even be able to change parts of your infrastructure in response to incidents," French-Owen said. "Have an instance that is running out of disk space? Trigger an action [such as] scaling, paging, [launching] an automatic runbook, or removing an instance, to solve it in real time."
Stephen Elliotanalyst, IDC
Large enterprises also need help to covert IT monitoring systems' data into actionable business insights, IDC's Elliot said. Datadog competitors, such as New Relic, have focused heavily on dashboards and reports for line-of-business personnel.
"Businesses don't care about log management -- they want to know, 'Can you reduce my time to market? Help my net promoter score marketing campaigns and lead generation?'" Elliot said. "It's great from a technical standpoint that we can collect a lot of information, but IT experts have to translate what it means to the business."
IT monitoring systems improve reliability for Brazilian cosmetics company
Admins who seek full-stack monitoring don't have to look very far
A comprehensive monitoring strategy can predict issues in the data center