Tag Archives: developers

Cloud Foundry has gone Pivotal – so what’s new?

A few weeks ago I was privileged to be at the launch of Pivotal – a new organisation formed by VMware, EMC, and with investment from GE. You can read all about our new company at GoPivotal.com.

I am Pivotal

I am Pivotal

What does that mean for me, and for my role on the Cloud Foundry team? What is happening with Cloud Foundry right now? What about the Cloud Foundry community?

Well, as my über-boss James Watters recently wrote – we are a central part of the Pivotal business.

Our mission is to become the most popular platform for new applications and the services used to build them across public and private clouds.

That’s a pretty compelling mission statement, and I’d personally even add that we want to be the “best” platform, as much as “most popular”. One of the main reasons I wanted to spend a couple of weeks at the Pivotal office in San Francisco was really to immerse myself in the team and in the culture of Pivotal Labs, as well as to be at the launch event, and to get a strong handle on what is happening with Cloud Foundry, version 2…

Wait, what? Version 2?

In the middle of last year, the Cloud Foundry team started some major work to improve many of the features offered by the platform. Back then, it was written about on the Cloud Foundry blog. We initially started to refer to “ng” components like the Cloud Controller (“cc-ng”), and that’s what we now mean when we refer to “v2”. At the start of the year we published a roadmap which laid out a lot more detail in terms of what is coming. There’s some really great stuff in there – many bugs squashed; a new, high performance router; support for developers to collaborate on apps, via concepts of organisations and spaces; new containerisation via Warden; custom domains (yes, finally!); and most importantly, support for buildpacks. Buildpacks will bring a major change to our platform, replacing the former concepts of runtimes and frameworks (say, Java with Spring) with the ability to drop in whatever runtime or container you may choose, instantly making the platform more customisable. We’re pleased that the folks over at Heroku have allowed us to inherit the buildpack concept and having played with the new platform, I believe this gives us a really cool and solid way to support apps.

Deploying #cloudfoundry v2 on Amazon

Deploying Cloud Foundry v2 on Amazon

While I was in San Francisco, I used BOSH to deploy my own new Cloud Foundry v2 instance to Amazon EC2 (and also attended the AWS Summit, which was a bonus!). Right now the team is working on migrating our  hosted cloudfoundry.com platform to EC2, and when we officially boot up v2 for the public, it will be running right there. This is not new news – both James, and our CEO Paul Maritz, have repeatedly spoken about AWS.  The point of Cloud Foundry has always been that it is a platform that is Infrastructure-as-a-Service agnostic, even when it was started by VMware, and I’m seeing increasing interest from folks want to run it on OpenStack, AWS, and other infrastructures as well as vSphere (by the way, did you read about how Comic Relief 2013 ran on Cloud Foundry on vSphere and AWS? so cool!). There is no lock-in here – write once, deploy to cloudfoundry.com, to a partner running a compatible Cloud Foundry-based instance, or to your own private cloud on your on infrastructure, as you wish. The Open Source nature of the project is exactly why I jumped on board with the team a little over a year ago.

Talking of the update to cloudfoundry.com: it is also worth mentioning that when the beta period comes to a close we will have pricing plans, a nice web console for user, organisation and application management, and the start of a marketplace for partners to plug-in their own services for developers. I can’t give more details in this post, watch the official channels for news!

I felt very strongly that I wanted to write about version 2. It is a very big step in evolving the Cloud Foundry architecture, and I believe that it is important for the broader  community to understand that it is a significant change. If you are running an app on cloudfoundry.com today, we’ll shortly contact you with information about migration to the new platform, as some changes will be needed to adapt to the fact that runtimes and frameworks are now buildpacks, there will be some changes to the way services work, and you will need our new ‘cf’ gem to deploy to the new service. We have already “paused” new signups on the current platform. If you look at the new documentation, you will find that it now focuses on version 2 – we apologise for any confusion during the transition process.

We’ve been talking with ecosystem partners about version 2 as well. For instance, our friends at Tier 3 recently blogged about Iron Foundry plans, and I had the pleasure of meeting with Stackato folks in person in San Francisco recently. If you are working with your own Cloud Foundry instance privately (we know that many organisations are!) I strongly urge you to talk to us via the vcap-dev mailing list to learn how you can start to take advantage of what the new platform brings.

What else does Pivotal mean for Cloud Foundry? Well – we are more open than ever, and keen to work with the community on pull requests to add features via Github. I’ve just written a  post for the Cloud Foundry blog about how to participate in the Open Source project. In fact, I’ll be talking more about this at the Cloud East conference in Cambridge next Friday May 24. We’re always happy to talk more about how to collaborate.

These are exciting times!

 

Advertisements

My next steps – joining the Cloud Foundry team

I’m very excited to announce that, from April 10th, I will be joining the Developer Relations team for Cloud Foundry at VMware.

This is a thrilling opportunity for me for a number of reasons.

  • from a technology perspective: Cloud Foundry is very, very, very cool. In my opinion, it really comes from a different set of thought processes than the other Platform-as-a-Service offerings out there, which make it unique and compelling.
    • the operating system stuff gets out of the way (why should it matter?), but multiple language runtimes and backend resources are available for easy scaling. Seriously, the first time I walked through the command-line tutorial and scaled a Ruby app to 6 load balanced instances with a single command, I was instantly impressed.
    • it is Open Source. The code is on Github. You can run your own cloud if you like. You can add support for your own languages and frameworks, much as AppFog have done for PHP, Tier 3 and Uhuru have done with .NET in Iron Foundry, and so on. This provides a huge amount of flexibility. Oh, and of course mobile and cloud go hand-in-hand, so last week’s announcement of FeedHenry providing tools to develop HTML5 apps to deploy on Cloud Foundry was really significant, too.
    • you can take your cloud with you using Micro Cloud Foundry – so the development and deployment model remains the same whether you are online or offline. I love this idea.
  • for me, personally: it’s a natural evolution of much of the work I’ve been doing over the past few years – focusing on developer communities and promoting technology adoption, as much as top-down solution selling. As my good friend James Governor is fond of saying and as his colleague Steve O’Grady wrote, developers are the new kingmakers – and with trends like mobile, cloud, and devops, nurturing those communities is more important than ever. You don’t impose technology on a community – you explain it and earn your place and reputation.
  • I’m looking forward to more speaking, more writing, more mentoring, and more online community building. These are things I’ve grown to enjoy (and in the case of the latter, appear to do naturally).
  • I’ve followed Patrick Chanezon, the Senior Director of the team, since he was setting up the developer advocacy programme back at Google – I have a lot of respect for what he’s achieved and the way he operates, so I’m delighted to have the chance to work closely with him. I’m excited to join everyone in the team, of course – I have spoken with most of the group already and I’m really looking forward to learning from their diverse range of experiences and backgrounds.

Between now and April 10th, I have a few things planned including a vacation (!), heading to EclipseCon to talk about MQTT and M2M topics, and some other speaking engagements. After I start the new role, I expect I’ll join in on the Cloud Foundry Open Tour and start to meet folks. I’ll also be on the team for the GOTO conference in Aarhus in October – exciting times ahead!

WebSphere MQ and Ubuntu (and other developer resources)

For some time now, I’ve been using Ubuntu as my desktop operating system. Although I’m yet to be convinced by Unity (it’s getting there, the more I learn the shortcuts and stick with it), I do know that Ubuntu is a hugely-popular platform for developers – and I know that many of my colleagues at IBM who are in development roles choose our internal Linux-based client options (which cover a range of distributions), instead of Windows or OS X.

So, what about developing with or using WebSphere MQ on Ubuntu? Well, the officially-supported platforms for WebSphere MQ V7.0.x don’t include Ubuntu – that’s primarily a combination of the relative popularity of RedHat or SuSE Enterprise platforms in production deployments, time and resource spent on testing, and the fact that it would probably only be practical to test and support it on a Long Term Support release if it ever became supported.

However, it is possible to get WMQ installed and running on Ubuntu without jumping through too many hoops. The primary stumbling block is that the software is packaged in RPM format rather than in Debian/Ubuntu-friendly DEB files. One piece of advice is to avoid any guides that suggest converting the packages using alien… it may seem unusual, but you’re likely to find it far easier to get it working by installing rpm on the system instead. My colleague Rob Convery has posted a couple of very useful blog entries on this subject which I’d recommend if you have a need to get yourself running on Ubuntu – again, bearing in mind that it is not an officially supported platform, and that should you encounter issues then it might be necessary to reproduce them under RHEL or SLES when raising a service call with IBM.

 

There are other ways to get to use and learn about WMQ too, of course – for example, you could grab one of the IBM Industry Application Platform cloud images to run on the IBM SmartCloud or Amazon EC2 (containing WAS V7, DB2 Express-C 9.7, and WMQ V7.0.1, running on SLES), or you can try a number of the WMQ family products in IBM’s SOA Sandbox, (including WMQ File Transfer Edition, and WMQ Advanced Message Security). You can also check out the MQonTV YouTube channel. Let me know what you think!

What a week for MQTT!

Part of my role as WebSphere Messaging Community Lead involves IBM’s MQ Telemetry Transport protocol. I spend a chunk of my time talking about how MQTT relates to building a Smarter Planet, and explaining how it can be used to build some very cool new applications and solutions.

MQTT logoFolks from IBM and Eurotech may have jointly authored MQTT, but it has been published online with terms enabling royalty-free use and implementation of the protocol. The next stage is to put it forward for standardisation. Last Friday, the call for participation in a standards discussion was published on mqtt.org. It’s open to anyone to join, and given the excitement I’ve personally seen in the developer community, I’m hopeful that we’ll see plenty of interest.

Friday saw even more big news, from an entirely unexpected source. As I stood chatting to people arriving at the OggCamp party that evening, my Twitter alerts and email went crazy with MQTT chatter… Facebook announced that their new Facebook Messenger application (a result of their acquisition of the Beluga team earlier in the year) uses MQTT! I’d been aware of different mobile app developers using MQTT for a while now – in fact we recently highlighted what a great match the protocol is for Android applications, on the mqtt.org blog – but had not known about Facebook’s interest or usage. In their post talking about how Facebook Messenger works, they call out the characteristics that make it a strong protocol for a mobile group messaging application – low bandwidth, low overheads, low power cost… all of the things that have made MQTT successful in sensor networks and solutions, make it ideal for these kind of applications as well.

Well… as I said, a big week, with some exciting news. So it seemed only right that I should give a talk about MQTT and all of these latest developments at OggCamp this past weekend – the event which three years ago, resulted in Roger Light creating his mosquitto broker.

You may recognise the slides as a remix of the talk I gave at LinuxConf in January, but I’ve updated them to highlight the OggCamp dimension and to talk about the recent news. There will be more to come during the coming weeks, so join the chat in channel #mqtt on Freenode IRC, and keep an eye on mqtt.org!

 

IMPACT 2011 – and the Unconference

This week is the last before I hop on a plane a head over to The Venetian in Las Vegas for this year’s IBM IMPACT conference. I’ve been privileged to attend the last few of these customer and user conferences, and I’m always happy to have the opportunity to meet users and techies, as well as to talk to new folks about what is happening in the WebSphere Messaging portfolio.

IMPACT 2011 banner

This year I’ve been given the task of talking about what’s new in WebSphere MQ (hint: if you look at the whole WMQ family, there is a lot going on); and to talk about our Smarter Planet protocol, also known as MQ Telemetry Transport.

As before, we have a social aggregator where you can follow the conference. One particularly exciting addition this year is the idea that we’ll run an unconference within IMPACT itself, and we’ve carved out the Thursday in order to do so. If you’ve never been to an unconference before, the concept is that the organisers put up the venue, framework, and support, and the attendees build up the actual agenda. The call for participation is open now… if you are really a techie or an architect and don’t want to be bored by the business-oriented stuff, you need to be at the unconference. It’s for developers, by developers (actually – that’s YOU).

WebSphere Unconference

The ideas will be posted at Impact 2011 April 11-13 using the wall grid posted onsite, where you can continue to submit and vote on ideas. The Unconference is held on April 14th from morning through afternoon. Sessions will run about 1 hour with several rooms open during each time block. The kick-off is at 8:30AM with a special guest speaker. The format of sessions is up to speaker(s) to decide – go solo, pair up or make a roundtable with a group. Make it your own.

There’s a slightly different way of doing things here, as instead of only having folks nominate themselves to present on the day, we’ve got a forum where you can suggest and vote on topics right now.

If you don’t believe that this thing is going to be “the place to be” on the Thursday of IMPACT, here’s what I actually consider the second best part (apart from the conversations with developers themselves): RedMonk. The guys are going to be all over the Unconference, including a keynote by James Governor, and a breakout session with Stephen O’Grady. Listen to @monkchips:

James Governor: Developers should be excited. The pendulum is swinging back… you are the new kingmakers, so what are you going to do about it?

For more on the unconference, you need to be following Ryan Boyles and Kathleen Holm. Oh, and if you’re coming to IMPACT, and you use Twitter, give Lanyrd a try – it’s a great way of sharing what you’re up to.