New Relic And Hortonworks See Strong Public Debuts
techcrunch.com
techcrunch.com
I work for AppNeta, and while we directly compete against New Relic, I love what they've done for the market. Their competitors were these clunky, on-premise, hard-to-use, didn't support Ruby / Python / PHP / node.js apps, and they've built a beautiful service that's legitimately easy to use. A lot of the data is available from CA or Dynatrace, but it's so much harder to set up and use.
Monitoring so commonly focuses on server metrics, or outside-in uptime, or logs, and I really do think that monitoring transactions end-to-end is an important part of the picture. Day to day, New Relic is absolutely the competition, but in the big picture, there's a lot of big companies using technology that's 15 years old because somebody like New Relic or AppNeta hadn't come along yet.
Glad to see that the market likes them as well. There's awesome stuff happening in Application Performance Monitoring (APM), and if you're not already using something like this, I strongly recommend you look at AppNeta, New Relic, ... and not really anybody else, because the rest haven't really figured it out yet.
Looks like you guys have some nice stuff, but do you mind if I give a new customer first impression?
Very confusing pricing.
First off, right off the bat. There is no central price page and I had to go hunt for what you charge.
Then when I did find Enterprise pricing for TraceView, it has some contradictory language. It says $119/host/month. Then in the bullets below it says
unlimited users
unlimited applications
unlimited hosts
Huh? Is it per host or is it per month?? That makes no sense to me.
Part of the problem is that we have a couple different products, and they're not all directly comparable with New Relic. Hence the lack of a central pricing page.
TraceView is our offer that's 1:1, as you found, and it's priced monthly per host. The "unlimited" is meant to be "you can buy as many as you want", vs. the "Startup" plan, where you can have a max of 5 hosts. Does that make more sense? (I'd be more than happy to change it to something that makes more sense!)
IMHO, keep the app and 'hours of retention' limit in. Remove the hosts and user limit (I hate user limits; don't you want more people exposed to your app?!?)
That said, a host collects a certain amount of data, so it's not really tenable for us to sell "site licenses" at a fair price. If you're super-bursty, we're happy to charge for the average, AWS-style, but that model tends to upset the finance folks, so we default to the "you'll clearly pay this" model first.
I beg to differ there. ;) I work on Skylight (https://www.skylight.io) which is definitely doing some interesting stuff in the APM space.
By focusing on just Rails, we are able to provide features that other services that try to be a jack-of-all-trades frequently miss. For example, we just rolled out a feature that helps you monitor memory usage in production with extremely low overhead[1], which I don't think is something any of our competitors offer.
But yeah, definitely agree that New Relic has been a market leader here. They've validated that people are willing to spend real money on improving performance, something that wasn't obvious just a few years ago.
Focusing on Rails lets us build a really awesome end-to-end solution that works out of the box. Building an intuitive, explorable interface means constantly considering the application's context, and I can attest to the fact that every time we do UI, nitty-gritty details about how our customers build their applications can often become the dominant consideration. When I use general-purpose solutions that are designed for "all languages and platforms", I am often very frustrated by the resulting lowest-common-denominator APIs and UIs.
We may eventually build versions of Skylight for other platforms, but if we do, we will apply the same level of care and design for other platforms as we have done for Rails. Among other things, that will mean having team members who have written large applications in those platforms, and letting those team members drive the APIs and design for those platforms.
http://appneta.com/products/traceview is what you want -- scroll down :)
If you want to chat, shoot me an email: TR @ appneta. I'm curious to hear if there's anything different we'd want to look at when working on HHVM.
You probably found the right one, but better link: https://support.appneta.com/cloud/traceview/instrumentation/...
I've met Lew Cirne before (their CEO), and he's not only an incredibly nice, down-to-earth guy, but an awesome programmer as well.
It was really inspirational (and fun!) to go to the NewRelic Futurestack conference last year -- everyone I bumped into was super awesome, friendly, and really loved their jobs.
For the past 4-years (ish), I've been using NewRelic in 100% of my personal and professional projects, and couldn't be happier. Really love everything they're doing -- it's a solid product, and a solid team.
On the other hand, in the current startup economy the pickaxe bubble is almost as big as the gold mining bubble, but I think it still probably applies more than it doesn't.
Disclaimer: I worked for SUNW "selling pickaxes" before, during and after the Dot Com Bubble, I told myself this lie -- and vowed to never tell it to myself again.
I wonder if there's a difference based on how many of those companies were selling mostly actual hardware infrastructure, which has just gotten cheaper and more commoditized over time, versus those which had actual differentiated software.
Both still got hit hard of course, a bubble's a bubble, but I wonder who did substantially worse?
VA Linux (now Geeknet) was near a $10 billion market cap [2] in trading immediately following its IPO. After several pivots and renames, it's market cap now is just $50 million.
RedHat went public at about a $1 billion valuation (raising $100 million), hit about $3 billion in trading after the IPO open, and has steadily grown to about a $10 billion cap over the last 15 years.
Meanwhile, today Hortonworks ends the day with a cap over $1 billion, and a couple of competitors in the wings that may also be doing IPOs at some point.
It's not that Hadoop and Linux are even comparable, but it's really hard to see from a 2014 vantage point that the Hadoop market will be as disruptive over the next 15 years as it would have been to see Linux disrupting in 1999. So Hortonworks and its ilk fall more in the space of advanced enterprise software--a more comparable relative temporary winner from that era was BEA Systems (which also had a spectacular rise and dramatic fall--amplified because they were a company that was disrupted by free software, and suffered due to unsustainable expectations). [3]
[1] http://www.cnet.com/news/10-years-gone-the-va-linux-systems-...
[2] http://www.theregister.co.uk/1999/12/10/ipo_makes_va_linux_i...
[3] http://online.barrons.com/articles/SB998410322172345858?tesl...
So, yeah, pickaxes are great. "Netflix for pickaxes" is more strongly tied to the market's long-term future.
The Mitt Romney comment is rather ignorant. There are many, many major companies that were rescued by Bain. A simple minded view of that industry is to do a discredit to the companies that thrived because of Bain. Staples, Guitar Center, Dominoes Pizza, the list goes on. If it weren't for Mitt Romney, there's be dozens of companies that would have simply gone bankrupt. Business isn't around to be a public service, it's sole purpose is to make a profit; with those profits, they are able to provide people with jobs. If Apple didn't make a profit and closed down, imagine how many people in our community would be out of a job. Profit provides an incentive to operate. I don't make a living writing computer code for charity; our livihoods depend on profit, like it or not.