Analyst Opinion – Over the next 2-3 years, many more applications will be available for smartphones, which will become increasingly capable and complex. In fact, we expect smart phones commonly available in 2011/12 to incorporate many of the ey subsystems available in the PC of today, including enhanced graphics processors, full multimedia subsystems, complete Internet access and application processing (AJAX, Dynamic XML, Java, Flash, Silverlight), and connection speeds approaching current broadband levels. With all this capability and complexity, will we soon find ourselves in the same dilemma with our smart phone and Internet access devices that we now face with the PC – trying to remain secure and prevent system crashes? And if so, is virtualization a solution?
Smartphones will undergo a fundamental change in philosophy going forward. We do not expect that the large, monolithic OS common in the PC environment will be duplicated in future smartphone designs. We expect a virtualized environment to evolve. Virtualization has many benefits in a smartphone device. It allows various subsystems to be abstracted and isolated from each other, preventing a potential crash in the graphics processor from bringing down the voice phone channel. It also isolates any applications that may be running from affecting key functions like email and SMS/MMS. It can also isolate mission critical data from access and corruption by errant applications or even malware. And it allows leverage of existing drivers and low level functions to be abstracted and utilized by many higher level functions.
Although the installation of virtualization in most phones today is impractical due to resource limitations, newer generation devices should have more than enough processing power and memory to accomplish this. In fact, some current platforms have already moved in this direction, running their “trusted” apps in an isolated environment (BlackBerry, J2ME). Virtual devices could even have more than one phone instance running at the same time (multiple SIMs) for isolating business and personal use/billing, or domestic and international needs.
To be sure, the move to virtualization is not a panacea, nor will it be easily accomplished. It will require a redesign of the phone architecture, which would require a Hypervisor running at the lowest level of the hardware. Some mobile-targeted hypervisors already exist, VMWare for example, but have not yet been incorporated into shipping devices.
Applications would require a custom design effort targeting specific processors (such as ARM and OMAP) and specific OSes (Symbian, Android, Windows Mobile, Blackberry). However, once implemented, the manufacturer of the device would have the ability to essentially reconfigure a device with whatever OS and other environment they choose without doing a device
redesign. This could have enormous implications for vendors that currently offer different OSes on essentially the same platform. Virtualization could also aid in device upgrades and management functions, making the devices more cost effective and decreasing the overall TCO for business users. Finally, truly isolating the applications could prevent the “blue screen”
crash and re-boot scenario so prevalent in PCs, which is disastrous in a smart phone.
The bottom line: We expect to see an increasing interest in virtualization technology for smart phone devices, following the same track as we are currently seeing in the PC environment. However, although the virtualization of smart phone devices is attractive, it will take a considerable effort to achieve. We expect a gradual uptake of virtualization, starting in high-end
devices over the next 3-4 years, and eventually making its way into the mid-tier device market. This will ultimately be of great benefits to users who will obtain better, more capable devices that operate more safely and more robustly.
Jack Gold is the founder and Principal Analyst at J.Gold Associates, an information technology analyst firm based in Northborough, MA, covering the many aspects of business and consumer computing and emerging technologies.