NEW YORK—Application performance monitoring may not be as cool a buzzword as Kubernetes, but APM is just as important for companies making the move to DevOps and cloud-native development.
New Relic, the leader in Software-as-a-Service APM, today outlined several products that deepen the integration of its monitoring tools into Web and mobile applications that make it easier to find coding errors, fix them faster and prevent them from happening in the first place.
The San Francisco company announced New Relic Applied Intelligence, a set of tools that brings to bear big data analytics, machine learning and predictive algorithms to “learn” and predict application behavior.
The new tools include Radar, which helps users see or predict trouble spots. They include NRQL Baseline Alerts which is a tool for drilling into granular New Relic data using the New Relic Query Language that establishes thresholds for performance alerts. In addition New Relic introduced Error Profiles, which helps in understanding, comparing and resolving errors.
Further reading
The company also announced new levels of integration with Microsoft Azure and Amazon Web Services clouds. For Azure, New Relic released the beta of its APM tool for Microsoft’s .NET Core 2.0, which puts the New Relic agents deep inside Azure applications.
For AWS, New Relic adds support for the Amazon Elastic File System, API Gateway, Application Load Balancer, Relational Database Service (RDS), Redshift and Virtual Private Cloud Flow Logs. New Relic now supports 25 AWS services.
New Relic also announced added improvements for users including role-based access controls, new dashboards, and a dashboard API.
Finally, the company demonstrated a preview technology called Distributed Tracing, an implementation of the open source OpenTracing project, which gives snapshots of the health of dependences between all aspects of an application service chain.
For instance, today’s mobile apps may involve services, such as payment services or location-based services that need to be working for an app to function property. Tracing across distributed application components enables administrators to spot trouble quickly.
Instrument everything
The key to all of these products is the New Relic APM Agent, which works in every piece of code in whatever format it has been deployed in (web app, container, microservice, or app function) to enable the collection or "instrumentation" of application performance data and metrics.
"Instrument everything," said New Relic founder and CEO Lew Cirne. "It's not production if it's not instrumented."
Instrumentation is important considering that companies focused on DevOps and the cloud are making thousands of code deliveries into production every year or even hundreds a day, such as New Relic customer AirBnB. The key to success is not the performance monitoring per se, Cirne says. Rather, it’s “continuous delivery every day and going fast at scale” to deliver code that contributes to new business for a company. The monitoring makes that happen.
Other New Relic users include Dunkin Brands, which achieved success in rolling out its new mobile app with on-the-go ordering and payments. The response times are critical in these applications, so Dunkin needed to be sure the app would deliver when it went live.
“The mobile ordering app was going to generate a lot of questions from the business and they expected answers about how the system was working or not, how it was being used and how effective it was,” said Matt Kraft, Architect for Retail and Mobile Technology at Dunkin Brands. “We had to plan for scale and we had to manage expectations.”
Another user, Draft Kings, needs to plan for its weekly “digital moment of truth” when fantasy football leagues kick off every Sunday. Their app needs to be able to withstand the millions of signups that happen in the hours before kickoff, and the frequent “Gronk Spikes” of traffic, when a widely owned player scores and owners check the app for score tallies.
"Gronk Spikes" refer to New England Patriots tight end Rob Gronkowski, nicknamed Gronk, who is a popular National Football League player and a favorite of fantasy league players.
Monitoring tools gives “us a common understanding of the metrics needed to improve our product across our entire code base,” said Mark DiAntonio, Director of Engineering at Draft Kings.
Scot Petersen is a technology analyst at Ziff Brothers Investments, a private investment firm. He has an extensive background in the technology field. Prior to joining Ziff Brothers, Scot was the editorial director, Business Applications & Architecture, at TechTarget. Before that, he was the director, Editorial Operations, at Ziff Davis Enterprise. While at Ziff Davis Media, he was a writer and editor at eWEEK. No investment advice is offered in his blog. All duties are disclaimed. Scot works for a private investment firm, which may at any time invest in companies whose products are discussed in this blog, and no disclosure of securities transactions will be made.
