The Schwartz Report

Blog archive

Microsoft Boosts Management Automation Features of Azure Hybrid Cloud Platform

Microsoft kicked off what looks to be its final TechEd conference with the launch of new services designed to simplify the deployment, security and management of apps running in its cloud infrastructure. In the opening keynote presentation at TechEd, taking place in Barcelona, officials emphasized new capabilities that enable automation and the ability to better monitor the performance of specific nodes.

A new feature called Azure Operational Insights will tie the cloud service and Azure HDInsight with Microsoft's System Center management platform. HDInsight, the Apache Hadoop-based Big Data analytics service, will monitor and analyze machine data from cloud environments to determine where IT pros need to reallocate capacity.

Azure Operational Insights, which will be available in preview mode next month (a limited preview is currently available), initially will address four key functions: log management, change tracking, capacity planning and update assessment. It uses the Microsoft Monitoring Agent, which incorporates an application performance monitor for .NET apps and the IntelliTrace Collector in Microsoft's Visual Studio development tooling, to collect complete application-profiling traces. Microsoft offers the Monitoring Agent as a standalone tool or as a plugin to System Center Operations Manager.

Dave Mountain, vice president of marketing at BlueStripe Software, was impressed with the amount of information it gathers and the way it's presented. "If you look at it, this is a tool for plugging together management data and displaying it clearly," Mountain said. "The interface is very slick, there's a lot of customization and it's tile-based."

On the heels of last week's announcement that it will support the more-robust G-series of virtual machines, which boast up to 32 CPU cores of compute based on Intel's newest Xeon processors, 45GB of RAM and 6.5TB of local SSD storage, Microsoft debuted Azure Batch, which officials say  is designed to let customers use Azure for jobs that require "massive" scale out. The preview is available now.

Azure Batch is based on the job scheduling engine used by Microsoft internally to manage the encoding of Azure Media Services and for testing the Azure infrastructure itself, said Scott Guthrie, Microsoft's executive VP for cloud and enterprise, in a blog post today.

"This new platform service provides 'job scheduling as a service' with auto-scaling of compute resources, making it easy to run large-scale parallel and high performance computing (HPC) work in Azure," Guthrie said. "You submit jobs, we start the VMs, run your tasks, handle any failures, and then shut things down as work completes."

The new Azure Batch SDK is based on the application framework from GreenButton, a New Zealand-based company that Microsoft acquired in May, Guthrie noted. "The Azure Batch SDK makes it easy to cloud-enable parallel, cluster and HPC applications by describing jobs with the required resources, data and one or more compute tasks," he said. "With job scheduling as a service, Azure developers can focus on using batch computing in their applications and delivering services without needing to build and manage a work queue, scaling resources up and down efficiently, dispatching tasks, and handling failures."

Microsoft also said it has made its Azure Automation service generally available. The tool is designed to automate repetitive cloud management tasks that are time consuming and prone to error, the company said.  It's designed to use existing PowerShell workflows or IT pros can deploy their own.

Also now generally available is WebJobs, the component of Microsoft Azure Websites designed to simplify the running of programs, services or background tasks on a Web site, according a blog post by Product Marketing Manager Vibhor Kapoor, in a post today on the Microsoft Azure blog.

"WebJobs inherits all the goodness of Azure Websites -- deployment options, remote debugging capabilities, load balancing and auto-scaling," Kapoor noted. "Jobs can run in one instance, or in all of them. With WebJobs all the building blocks are there to build something amazing or, small background jobs to perform maintenance for a Web site."


Posted by Jeffrey Schwartz on 10/28/2014 at 10:59 AM


comments powered by Disqus

Subscribe on YouTube