Comments on: How Do You Measure Up? Introducing the App Speed Index

By: Brent Miller

Brent Miller — Fri, 29 Jun 2012 14:21:48 +0000

Ian: The server-side speed is the time the app takes to handle the request on the server — it’s equivalent to what our agents report on the app overview chart. Front-end time like DOM rendering aren’t included. Those are part of the End-user speed.

Eric: The error rate tracking bug has been fixed. Thanks for reporting it!

By: Ian

Ian — Fri, 29 Jun 2012 10:19:27 +0000

Do you have some more details about what makes up “Server-side speed” is that things like DOM rendering and the actual load time of the page?

By: Ralph Bodenner

Ralph Bodenner — Thu, 28 Jun 2012 00:23:37 +0000

Hi Eric, we should indeed be showing you the Errors tab and the rate graph there. Error traces (with all their lovely stack trace detail) are Pro and up, but Standard gets error rate tracking. We’ll fix that shortly!

I’ll look into the errors posting comments, too. Thanks for the heads-up!

Ralph

By: eric

eric — Wed, 27 Jun 2012 07:47:59 +0000

Hi Ralph, you said that error rate is reported for standard and up, I’m on standard, but the errors report under the diagnostics in the menu is not enabled (only for pro). Are the errors just tracked, but not showed and used only in the speed index?

btw, you have problems with the blog, when I comment 90% of the times it throws an error (connection error), I guess that the guys at pagely should have a look at their errors report

By: Ralph

Ralph — Tue, 26 Jun 2012 20:58:57 +0000

@Ben: Your percentiles look good now. There is a little bug where, if we’re recalculating the daily rankings between midnight and 1am, we end up using the (missing) data for the new day. We’ll fix it.

@eric: Glad you like it! Error rate is reported for Standard and up, but you’re right about Lite. We’ll remove Lite apps from the ranking calculations and regenerate the historical rankings. We don’t plan to add rpm/ppm comparisons, since we think they’re not relevant to the experience your customers are having on your site.

@Ashley: You’re right, API services aren’t covered by the App Speed Index yet. Since the performance expectations for APIs vary more widely than for web apps used by real people, and are much different than for such apps, we didn’t include them in the initial release of the Index. We’ll consider creating separate groups in the future for APIs, and at that point we’d compare just app server response time, error rate, and availability for those groups.

By: New Relic Now Tracks Fastest Apps & Industries, Lets You Compare Your Speed To The Competition

Tue, 26 Jun 2012 20:12:31 +0000

[...] For more on Index and Custom Dashboards, see New Relic’s blog post. [...]

By: New Relic Launches App Speed Index and Custom Dashboards « Joyeur

New Relic Launches App Speed Index and Custom Dashboards « Joyeur — Tue, 26 Jun 2012 18:04:41 +0000

[...] more about the App Speed Index here, or check out this blog post. And don’t forget to check out our living infographic, updated daily to show how the peer groups [...]

By: Ashley

Ashley — Tue, 26 Jun 2012 16:22:43 +0000

What API service includes end user monitoring? I think you need to reconsider your requirements for this feature.

By: eric

eric — Tue, 26 Jun 2012 15:54:54 +0000

other 2 info that would be really useful for a better comparison are the rpm and ppm, will you add them?

By: eric

eric — Tue, 26 Jun 2012 11:56:49 +0000

I already love this feature! One little problem: error rate is useless IMHO, as the errors tracking is only with pro accounts, all the others (standard/free) will get 0 errors, so a pro will almost always be in the right of the chart… with a lot of applications with 0 errors, that’s not realistic