Three new goodies from the New Relic engineering team that we like to share:
Many of our customers leave large-screen monitors with New Relic dashboards running in a public place in their offices. We saw this tweet yesterday for example:
“@adamse: @barryntech replaced a bunch of monitoring code and systems with a 55″ monitor and New Relic”
Live charts should appeal to all you dashboard watchers out there. Instead of refreshing the full page every 5 minutes, the charts update themselves every minute, with new performance data.
Almost all of the overview charts are now live charts. If you were thinking about setting up a New Relic dashboard in your office, now you’ll get even quicker updates. And for those of you who like to take bets on what the next refresh will show (you know who you are), things might be a little more exciting.
Query analysis: Explain plans, explained
Explain plans are a fairly advanced tool for web developers. Many developers don’t use them at all, because they are cryptic and can take some time to decipher. Even for more advanced developers, they tend to be a tool of last resort, used to tune SQL statements that resist more straightforward techniques.
A few weeks ago, I was having a conversation with another engineer here at New Relic, talking about SQL formatting, when I had the idea of taking the most relevant information from an explain plan and superimposing it on the actual SQL statement. It would look something like this:
We liked the idea, but the feedback came back that the wavy underlines were too subtle. So we played around with it for a bit, and now include the hints right after the SQL:
The design goals here are fairly typical of New Relic: let us do the heavy lifting, gather the metrics, and show you only what you care about. In the query analysis, we highlight the most common issues with queries, such as full table scans, and queries without an index. But we don’t show you every detail from the explain plan. If you want that, you can look at the raw explain plan by clicking “show details”.
The query analysis explanations are for MySQL only at this point. I hope you find them useful!
Heroku Dynos Tab
For both Java and Ruby Heroku customers, we have a present: a new Dynos tab with the information you could want about your Dynos: Memory usage, the number of dynos running, restarts, and more. Now you can get a better grasp of the details when you’re running at capacity and make informed decisions about upgrading when you need to.
As always, let us know what you think! We hope you love it!