Forward-looking: The most intriguing idea to come out of Microsoft's Build conference this year is a forward-looking concept the company is calling the "hybrid loop," which defines a set of hybrid apps that span the gap between the cloud and client devices on the edge. At a basic level, hybrid apps will be able to run both locally on a client PC and in the cloud and, most importantly, shift dynamically between the two.
If there's ever a place where you're likely to hear the direction that computing and applications are headed, Microsoft's developer-focused Build conference is probably it. True to form, this year's Build featured several interesting new concepts that managed to provide a sense of where Microsoft's vision for software is headed, while simultaneously hinting strongly at some key technologies (most notably Arm-based processors) that they believe will be necessary to power that future.
For hybrid applications, the cloud is meant to be seen as an additional computing resource, much like a GPU can be used in addition to the core CPU powering a typical PC. For example, these types of apps would be able to make decisions at runtime whether to do AI inferencing on the client devices or on the Azure cloud.
In addition, because most of these new apps are likely to incorporate some level of AI within them, they're also expected to be able to leverage NPU (Neural Processing Units) or other types of AI hardware accelerators. Thus, hybrid apps and the hybrid loop concept can be written to leverage a local device's CPU, GPU and NPU and, as needed or directed, also use cloud-based computing resources.
It's a fascinating theory, though how well it can be translated into real-world capabilities remains to be seen. After all, the concept of cloud-based processing for client devices has been around for a while now. It helps power virtual desktops and thin clients, among other applications, and forms the basis for Microsoft Windows 365 and their notion of a Cloud PC. For certain types of environments or situations, these solutions can work quite well. However, performance limitations, scaling concerns and more have constrained them to niche deployments up until now and they make up just a tiny percentage of the overall client computing market.
Still, leveraging some of the principles of hybrid cloud environments and applying them to the client certainly seems like it should work. Dynamically allocating resources has been a key part of cloud computing architectures since the early days, so it just makes sense to want to apply some of these principles to client computing devices. Plus, one of the huge benefits of running in the cloud is that it makes platform-specific application issues much less of a concern. By their very nature, cloud-based apps are cross-platform and hardware platform agnostic.
One big challenge is that up until now applications haven't been specifically written or optimized to run in these kinds of environments. Part of the Microsoft vision involves using cloud-native development practices to create these AI-powered, hybrid client/cloud apps. In fact, several announcements at Build focused on new tools for developers to start getting used to these ideas. Microsoft talked about leveraging its OnnX Runtime and Azure ML tools along with a new prototype AI tool chain to allow developers to create applications that fit into this hybrid loop model.
One practical challenge is that most current PCs don't have dedicated AI processing. Intel and AMD have both been talking about this and will likely have more dedicated AI silicon in future parts.
One practical challenge is that most current PCs don't have dedicated AI processing. Intel and AMD have both been talking about this and will likely have more dedicated AI silicon in future parts. In addition, Nvidia's CUDA efforts show that GPUs can be used for AI applications, but most of those only run on servers. The one current AI processing option for Windows-based PCs are the Qualcomm Snapdragon SoCs used in Windows on Arm-based devices, including Microsoft's Surface Pro X.
Despite very modest shipments to date for this entire category, Microsoft is a big believer in the potential for Arm-based PCs and they made that clear through via several announcements at Build.
First, Microsoft previewed Project Volterra, which is a Snapdragon-powered tiny desktop PC/development kit. Project Volterra leverages Qualcomm's newly released Neural Processing SDK for Windows and can be used to start building these type of AI-capable hybrid apps. The Qualcomm tool leverages the Snapdragon's integrated Hexagon DSP, Kryo CPU and Adreno GPU that mobile developers have been using to create AI-powered smartphone apps for several years now. Recognizing that the vast majority of PCs still have x86 CPUs, Microsoft also said they would be bringing NPU-specific support to all their other Windows development tools.
Speaking of development tools, the other big Arm-related announcement from Build was the release of a complete set of Arm native development tools including Visual Studio 2022, VSCode, Visual C++, Modern .NET6 and Java, and more. The company also said they were working to get a number of open-source projects, including Python, node, git and more to natively target Arm architecture. It's all part of the company's effort to make Arm an equal citizen to x86 in the PC client world.
Even with the potential growth in Arm-based PCs and dedicated AI processors across PCs, there are still questions about whether current client PC architectures can leverage cloud resources in an effective way. Though nothing was announced, it seems as if Microsoft is working on new types of cloud PC designs and architectures that could provide a better way to achieve this. We shall see.
Microsoft doesn't expect the hybrid loop concept to take off overnight, so many of these efforts are expected to extend over several years. Still, it is quite interesting to see the direction that the company sees the computing world moving in. There are clearly some big hurdles to overcome, but bringing together cloud computing, edge computing, client computing, AI and more into an organized whole that runs across a very heterogeneous set of hardware resources is a vision I expect we'll be thinking about for some time to come.
Bob O'Donnell is the founder and chief analyst of TECHnalysis Research, LLC a technology consulting firm that provides strategic consulting and market research services to the technology industry and professional financial community. You can follow him on Twitter @bobodtech.