New Relic Unleashes the NERD to Analyze Raw Performance Data
New Relic (NYSE: NEWR), a cloud-based provider of application monitoring, today announced that it’s opened up its New Relic Database (NERD) to enable a wider swath of customers to get fine-grained analyses of their application performance data.
New Relic built its massively scalable database, which it affectionately calls NERD, in the cloud to enable it to ingest huge amounts of data. Every second, NERD ingests more 2.5 million events generated by its clients’ servers, SANs, routers, networks, applications, smartphones, PCs, and other IT assets.
As part of its service, the company offers basic canned reporting, which aggregates the data in basic ways, such as by host or transaction type. But New Relic realized that some customers need deeper and most granular analytic experience, which is why it introduced the New Relic Insights service About 12,000 New Relic customers have signed up for this service, which allows them to perform advanced analytics on the raw, un-aggregated data.
But many thousands more organizations are using New Relic’s free performance monitoring services New Relic. Up to this point, these customers have had to make do with the canned reports that New Relic offers them.
Today, New Relic announced that even its free customers can begin querying their raw performance data feeds stored in NERD, enabling them to get the same deep understanding as customers who pay for APN. The catch is that New Relic’s free customers can only use the advanced analysis of raw, unaggreated NERD data for only the past 8 days, on a trailing basis. If they need more, they have to pay up.
Analyzing the raw, un-aggregated data is important for customers who need to get a low-level understanding of how their applications are working, and what kind of experience their customers’ customers are getting, says New Relic’s Vice President of Product Management Patrick Lightbody.
“We’ve effectively democratized APM by making it availed to lots of people,” Lightbody tells Datanami. “As businesses have to iterate more and more and rapidly to meet the needs of the market, they need to look at data from every dimension possible, and you can only do that if don’t aggregate the data.”
The phrase “every business is now a software business” may be a cliché, but that doesn’t mean there isn’t a kernel of truth to it. As Lightbody sees it, the acceleration of the digitization of the consumer life means companies must strive to differentiate themselves in the Web and mobile experiences they deliver, or simply fall behind.
“The primary interface these days for a lot of business is through the phone or the Web browser,” he says. “As a result of that, what we’ve found over the years is our customers, especially the ones building really great software, aren’t just building a release and saying we’re done. It’s a constant iterate process, where they’re taking feedback from customers and systems, learning what worked what didn’t, what scaled or what didn’t, and pushing that feedback back into the development process.”
In such agile DevOps environments, having a deep insight into exactly where website visitors are when they abandon their shopping cart, or what part of the Java or .NET application is acting up for customers in Australia, can mean the difference between delivering a compelling user experience and losing a potential customer.
Getting that deepness of data requires not rolling it up into aggregations, where you inevitably lose something in the translation. “That’s why we have this big database called the New Relic Database that’s able to collect all this data without dropping any of it, without any aggregation, where you lose that dimensionality,” Lightbody says. “The database is capable of looking at all the application performance and user interaction data that we’re collecting, from effectively every dimension possible.”
New Relic enables customers to query NERD through its New Relic Query Language (NRQL), a SQL-like language that’s adapted somewhat to trailing time periods, such as the past 24 hours or the past week. (The selection of an eight-day trailing time window for the free version of New Relic Insights was not made arbitrarily, as it allows customers to do week-over-week comparisons of their application performance, with one day’s worth of wiggle room.)
The San Francisco, California-based company introduced several other capabilities today as part of the expansion of its Insights service, including error analytics, geographic perspectives, and a visual data explorer.
New Relic, which went public in late 2014 and developed its own massively distributed database technology, has a market capitalization of $1.7 billion, making it one of the more successful big data service providers.