New Relic: Turtles? No. It's cloud infrastructure all the way down

If you build it, they will come and measure it. "They" in this case is New Relic, which offers developers a service for monitoring application performance and, now, computing infrastructure.

On Monday, the company plans to announce general availability of New Relic Infrastructure, for overseeing cloud and hybrid infrastructure.

The service won't actually be available then, but on November 16, in conjunction with the company's FutureStack conference.

The announcement comes with a rebranding of New Relic's product portfolio, hence to be referred to as the New Relic Digital Intelligence Platform.

"A lot of things have changes in the past five or six years," said Bharath Gowda, senior director of product marketing for New Relic, in a phone interview with The Register. "People have moved from managing static servers to cloud servers that are highly dynamic."

This creates a management challenge, said Gowda, because lots of servers magnify the opportunity for errors. Gowda observed that a recent Gartner report attributed 40 per cent of outages to configuration changes.

This is a problem of scale. When people had only a few servers, said Gowda, people could manage them like pets, giving them names and tending to them individually. As IT organizations incorporated virtual machines, it became more like managing cattle. "And in the containerized world, it's like managing bees," he said. "You don't know when they come and go."

Hence the need for tools to keep tabs on the hive.

New Relic Cloud Infrastructure, introduced in August as a private beta, aims to provide more control to IT teams as they try to keep pace with the rigors of agile development practices and continuous integration/continuous deployment pipelines.

Through live monitoring and change tracking, the service lets administrators keep an eye on files, packages, settings, service status, logins, and security groups, among other things.

It provides system health updates every five seconds, for metrics like CPU, memory, and disk utilization. It also allows the creation of customized alerts and alert policies, based on Amazon EC2 tags or data from automation systems or elsewhere.

In addition, the service, integrated with Amazon EC2 and Docker, can locate vulnerable packages and unauthorized services across a large number of hosts.

New Relic is also updating its product dashboards by making both system health metrics and configuration changes available. "With our unified dashboard now we can plot both metrics data and events data in the same interface," said Gowda.

Another new capability is the ability to define the criteria for issuing alerts. By using historical performance for prediction, said Gowda, operations teams can have a better sense of the appropriate threshold for when notifications should be sent. ®