Data Driver

Blog archive

Windows 8 Ups the Data Transfer Ante

Talk about driving data: the audience broke into applause at last week’s BUILD conference when some of the new blazing fast data transfer capabilities were demonstrated by Microsoft’s Bryon Surace during a keynote address.

“With Windows Server 8, we can use multiple NICs [network interface controllers] simultaneously to help improve throughput and fault tolerance,” Surace said.

To demonstrate the new speedy data-transfer capabilities, Surace used a server running Hyper-V with two virtual machines, one of which was connected to two disks. One disk was connected using a 1GB Ethernet connection, a setup he described as “very typical, very commonplace in today’s environment.”

The other disk was connected “using multiple high-speed NICs that are leveraging SMB 2.2 multi channel and RDMA [remote direct memory access].” Starting up a SQL load generator and going to a performance monitor, Surace pointed out how the 1GB Ethernet card was transferring data at less than 100MB/sec., which he said was “pretty typical.” The second disk, however, was transferring data at more than 2GB/sec. That’s when the applause broke out.

“Now, previously, these technologies were only available in high-performance computing, but now with Windows Server 8, we're building them for one of the most common roles in Windows, Surace said. He went on to show that the NIC wasn’t saturated, but rather was using only about 15 percent of the available throughput.

“This is a clear indication that we haven't even scratched the surface of what's possible with Windows Server 8,” he said. “And as we move over and take a look at the performance, we're only using about 1 percent of the CPU on the server to be able to push this throughput.”

Surace also demonstrated the simplified storage array management capabilities of Windows Server 8. For this, he used a server connected to 16 SSD hard drives, with no specialized controllers, “just a bunch of disks, or JBOD, directly connected to our server and being managed by Windows.” He noted how the disks were used to create a storage pool for which some space was carved out and represented as a drive on the server. He also showed file shares connected by the “improved SMB 2.2 protocol.”

“So, the key here is you don't need a PhD in storage, Surace said. “You can simply attach just a bunch of disks to Windows and have it all managed and deployed right there.”

The full keynote can be viewed via Microsoft’s Channel 9 video service.

What are the software development ramifications of the new Windows Server 8? Comment here or drop me a line.

Posted by David Ramel on 09/21/2011


comments powered by Disqus

Featured

  • Cloud-Focused .NET Aspire 9.1 Released

    Along with .NET 10 Preview 1, Microsoft released.NET Aspire 9.1, the latest update to its opinionated, cloud-ready stack for building resilient, observable, and configurable cloud-native applications with .NET.

  • Microsoft Ships First .NET 10 Preview

    Microsoft shipped .NET 10 Preview 1, introducing a raft of improvements and fixes across performance, libraries, and the developer experience.

  • C# Dev Kit Previews .NET Aspire Orchestration

    Microsoft's dev team has been busy updating the C# Dev Kit, a Visual Studio Code extension that enhances the C# development experience by providing tools for managing, debugging, and editing C# projects.

  • Hands On: New VS Code Insiders Build Creates Web Page from Image in Seconds

    New Vision support with GitHub Copilot in the latest Visual Studio Code Insiders build takes a user-supplied mockup image and creates a web page from it in seconds, handling all the HTML and CSS.

  • Naive Bayes Regression Using C#

    Dr. James McCaffrey from Microsoft Research presents a complete end-to-end demonstration of the naive Bayes regression technique, where the goal is to predict a single numeric value. Compared to other machine learning regression techniques, naive Bayes regression is usually less accurate, but is simple, easy to implement and customize, works on both large and small datasets, is highly interpretable, and doesn't require tuning any hyperparameters.

Subscribe on YouTube

Upcoming Training Events