We asked readers for questions about Windows Azure to put to Microsoft, and you came up with some great responses. Tim Buntel, product manager for the Windows Azure platform at Microsoft Australia, has all the answers. Read on!
There’s so much happening in the cloud computing area nowadays, and Windows Azure just seems like Microsoft’s version of Amazon Web Services. I notice that Azure is coined as “platform as a service”. Can you walk us through the differences?
Amazon Web Services are generally considered “infrastructure as a service”. Both Infrastructure and platform as a service are public cloud offerings to deliver computing resources. The difference lies in the level of abstraction provided by each.
The Windows Azure platform abstracts most of the operating system level details and allows you to focus on the application itself. It provides services for the kinds of things you need to develop for every type of application, from consumer web to enterprise, such as compute and storage, access control, caching, a CDN, and business intelligence based on SQL Server among others.
Infrastructure as a service provides a virtual computing environment into which you deploy virtual machines that will run your application. These machine images are just like real servers which you need to administer from the operating-system up including user and account management, patches and updates, and so on. With Windows Azure, you build and deploy applications; with IaaS, you build and deploy machine images in which applications can run.
What can Azure do for a small business that Office 365 can’t? We have a small consultancy that uses Sharepoint, Outlook and Lync from Office 365.
I’m really glad to hear that you’re using Office 365! It’s a great way for organisations to use the familiar Microsoft productivity applications that power so many successful businesses without having to spend time and money on physical infrastructure, security and reliability.
What Office 365 is not is an application development platform. Windows Azure allows you to build, host and scale applications in Microsoft datacentres. For example, you might use Azure for a custom database application that helps you manage and report on inventory and sales.
I have an idea for a cloud-based application that involves uploading and storing large images. I'm not a programmer, but all kinds of cloud complexities come to mind in modelling this – bandwidth usage, upload timeouts, optimal use of HTML5 browser local storage, etc. Where can I go to get advice on modelling a practical structure within Azure to best handle this?
Windows Azure offers a Binary Large Object (BLOB) Service which is the simplest way to store text or binary data (like images or other media files) with Windows Azure. There are many helpful resources available to learn more about it. A good start is here, which provides an overview of storing and accessing data in Windows Azure. I’d also recommend taking a look at the tutorial on using the BLOB service. MSDN has heaps of tutorials and other learning content including downloads to free tools to help get you started. Finally, you might want to consider working with a Microsoft partner who can provide expertise to you directly.
My organisation currently hosts a large MS SQL database, and a web app that uses stored procedures on the DB. We offer this as a SaaS model for our clients. The database handles approximately 1,000,000 new records per day. My concern with moving this type of application to the cloud is around speed and scalability. Can you please give some real life examples of very high volume applications that are using Azure, and outline how to determine whether Azure is appropriate for high transaction applications like ours?
There are many case studies available that illustrate how companies have been successful with the Azure platform -- check them out. for all of them. One that comes to mind is online ticket-seller Flavours – they can handle selling 150,000 tickets every 10 seconds!
We also recently announced that Azure will be adding support for Hadoop; a great platform for analyzing unstructured data or large volumes of data – read the blog post for an overview. There’s also a whitepaper available on creating high-performance SQL Azure applications that would be helpful.
How do I use SSIS with the Azure implementation of SQL Server?
For those who aren’t familiar with it, SSIS is SQL Server Integration Services; a component of SQL Server that can connect and transform disparate data sources. SQL Azure does not include SSIS, however, depending on what you’re trying to do, you might find the SQL Azure Data Sync service useful. It allows you to synchronize data across on-premises implementations and Azure. Watch the video at this link for an overview.
In this video, James Coenen-Eyre explains how Azure has transformed his development process.