An online conference: "Microsoft Online Cloud Conference: the TechDays team goes online" has had a date change. In my first post about the conference, I said it was running on the 20th September. Well, the registration site was only created this morning and so to give people enough notice of the registration, the date has been changed. It is now running on the 8th October.
Here are the registration details:
Event ID: 1032459728
Registration page here.
Here are a few details on top of what it says on the registration site's blurb. We've said it's a UK conference but that's kind of irrelevant seeing as it's online. Plus - we have presenters from Holland and Germany...
Track 1: Cirrus – the high level stuff
09:30 Welcome and Intro to Windows Azure: Steve Plank
10:30 Lap around Windows Azure: Simon Davies
11:30 Lap around SQL Azure: Keith Burns
13:00 Lap around App Fab: David Gristwood
14:00 Windows Azure: the commercial details: Simon Karn
15:00 Q&A Panel: All presenters
Track 2: Altocumulus – the mid level stuff (cast studies)
10:30 Case Study: TheWorldCup.com - Brian Norman, Earthware
11:30 Case Study: eTV – Dan Scarfe, DotNet Solutions
13:00 Case Study: Mobile Ventures Kenya – Mark Hirst, ICS
14:00 Case Study: : IM Group Experiences – Jeremy Neal, IM Group
15:00 Q&A panel: All presenters
Track 3: Stratocumulus – the low level stuff (deep tech)
10:30 Taking care of a cloud environment: Maarten Balliauw, Realdolmen
11:30 Windows Azure Guidance Project: Dominic Betts, ContentMaster
13:00 Azure Table Service – getting creative with Microsoft’s NoSQL datastore: Mark Rendle, DotNet Solutions
14:00 Release the Hounds: Josh Twist, Microsoft
A welcome to all attendees, online-housekeeping information and quick overview of the agenda.
Then follows a general introduction to what Windows Azure actually is, what the components are and some of the theory behind how it works and operates. This will involve brief overviews of the data-centres, scalability, storage and the database - SQL Azure. Also an overview on how to create an account, how to get free compute hours, how to set up a Windows Azure environment and how to package an application up and get it running. Importantly also – how to remove it and tear down the environment to stop running up a large bill!
Simon gets in to more detail about developing for the Azure platform and its specific peculiarities: using the service and management APIs, architecting for scale, using Windows Azure Storage. He’ll also talk about the Visual Studio development environment for Windows Azure – how to get the best from it, when to use it and when to deploy the application to the cloud for some specific scenarios such as testing for performance.
Keith talks about the architecture of SQL Azure and the changes that had to be made to SQL Server for it to run in a cloud operating system. He details the way your data is handled and how to look after it and secure it. He then talks about how you can connect your cloud apps to this database service but also how you can connect your on-premise apps to the service as well. He ends by showing some demos of how to manage the data – even using familiar tools such as those used for managing an on-premise SQL Server.
“The forgotten ones” – 2 key services – the Access Control Service and the Service Bus. Both of them are used most often to connect on-premise infrastructure and applications to the cloud. He talks about the peculiarities of getting messages into and out of the service bus. How you’d set it up and code for it. And then moves on to how identity elements can be brought together and even shows how a local on-premise Active Directory, when used with the Access Control Service can give a single-sign-on experience – just like you get with say IIS or Exchange.
How you pay and exactly what you’re paying for, how the billing works and other commercial details are often overlooked because the savings can be massive. But in small experimental deployments it’s easy to forget these details and end up with large bills you weren’t expecting. Simon details all the commercial elements of Windows and SQL Azure so you need never get caught out. He also introduces details for signing up to free Windows Azure services and gives details on how to maximise those services.
eTV are one of the world’s leading independent digital media companies, creating value for brands and broadcasters in a world where old media models are becoming obsolete. In the UK they run TV channels including Thomas Cook TV, Gala TV and produce shows such as MasterChef for the BBC. They are well known for their technical and creative leadership. eTV were looking to create a platform to deliver social games for websites such as Facebook and MySpace. This would be used for both for their own customers, but also made available for other companies to use. They were keen to use a platform which could rapidly scale up and down to support these games and one based on a global network of datacentres. They turned to Dot Net Solutions for help. Together the two companies built the Playful Planets platform on top of Windows Azure. In this session we explore why Azure was the right choice for them and explore the finished result
Have you ever had a project that needed to be released yesterday but requires scalable website hosting, video streaming and global delivery? What about one that is only needed for two weeks of intense activity and will never be used again? If the answer is yes to any of these then this session will give you an overview of how we managed to launch a Silverlight powered, HD streaming video map application in just three weeks, using Windows Azure, ready to be launched at the opening of the football world cup.
IM Group have a wealth of experience in cloud projects. Jeremy Neal shares this with you – both the commercial as well as the technical experiences – because the cloud is not the answer to every problem. Sometimes it fits from a technical perspective but doesn’t make sense commercially. Sometimes it’s the other way round – Listen to Jeremy talk about those projects that did and didn’t fit and why.
Uploading and initializing a worker role in Windows Azure can take some time – maybe as much as 30 minutes. There are scenarios where a business may need to quickly start multiple nodes’ processing by triggering an event (or ‘releasing the hounds’). In a scalable architecture with many worker roles this is tricky as these worker roles are hiding behind a NAT and aren’t directly addressable. In this session we’ll look at two possible approaches
· Polling – a simple design where the processing nodes poll blob storage to look for an indication of state and whether they should begin processing.
· Event Based – a more complex implementation that uses the AppFabric ServiceBus to kick the nodes into action
This session compares the pros and cons of each approach.
Whilst solving these problems directly, I also feel that these patterns would educate the attendees in some best practice techniques when faced with the problems exposed by elastic-scalable architectures.
No, this session is not about greener IT. It does cover the environment your application will live in once deployed to Windows Azure: learn about using the RoleEnvironment and diagnostics provided by Windows Azure. Communication between roles, logging and diagnostics are just some of the possibilities for things you can do if you know how the Windows Azure environment works. And who knows, maybe we can even auto-scale our application...
Microsoft’s Azure Table Service provides a low-cost solution for storing and searching structured data in “The Cloud”. Plus, it’s one of these cool new NoSQL data stores that everyone’s talking about. But it’s very, very different from SQL Server and other relational databases, so is it the right solution for your project?
In this session we’ll look at how the Azure Table Service works and how to use it. We’ll look briefly at the high-level Data Services SDK, talk about its limitations, and then quickly move on to the REST API and how to use it to improve performance and reduce costs. We’ll make-up some pretend real-world problems and solve them in new and interesting ways. Code will be written. We’ll de-normalize data (for fun and profit). We’ll talk about how certain social networking sites can deal with huge volumes of data so quickly, and why it sometimes goes wrong.
We’ll also cover some of the very useful features of relational databases the Azure Table Service doesn’t provide, and whether they can be reproduced in other ways. Acronyms such as ACID, BASE and CAP will be tossed around with gleeful abandon. And we’ll discuss the relative costs of Azure Storage Services (including Blob, Queue and Drive) compared to SQL Azure, and ways to appease the bean-counters.
The Microsoft Patterns and Practices group have, to date, released two significant sets of guidance for Windows Azure: "Moving Applications to the Cloud" and "Developing Applications for the Cloud." This session will provide an overview of this guidance, including a description of the scenarios used, demos of the sample applications, and a discussion of a selection of the best practices advocated in the guides.
It's free to attend and you don't have to attend every session. Well that'd be difficult with three tracks running in parallel anyway. I can have three tracks on my screen but three people talking at the same time?
Please join us for a fun day.
My marketing buddies have donated some swag too - you only have a chance to get it if you attend!