Also for DB backups, there’s an Azure backup tool for SQL Server. It installs a file system driver and redirects filesystem writes (ie: writing a file like a backup) to a storage account and just leaves a stub in place. When you read from the stub, it redirects the data from the storage account to you.
Regards, Greg Dr Greg Low 1300SQLSQL (1300 775 775) office | +61 419201410 mobile│ +61 3 8676 4913 fax SQL Down Under | Web: www.sqldownunder.com<http://www.sqldownunder.com/> From: [email protected] [mailto:[email protected]] On Behalf Of Greg Keogh Sent: Monday, 15 June 2015 12:09 PM To: ozDotNet Subject: More on 'CloudStream' I discovered by accident this morning that someone has already considered the issue of "a file system in the cloud" and made this announcement a year ago: http://blogs.msdn.com/b/windowsazurestorage/archive/2014/05/12/introducing-microsoft-azure-file-service.aspx I see there's a NuGet<http://www.nuget.org/packages/WindowsAzure.Storage> package for Azure Files but I haven't tried it. As a small technical exercise yesterday I created a class derived from System.IO.Stream which worked on openstack CloudFile objects. It completely works, but I stage the whole object in a MemoryStream for reading and writing, which is a clumsy hack, and there is an irritating delay as the object is initially loaded and on each flush/close. The Azure blob APIs support working with byte ranges, so you could potentially write a smarter more efficient stream, but it's not worth my effort. Greg K
