Virtualization Reborn

Timing is everything in the tech business and because of that, it’s not uncommon to see old ideas come back to life in new forms. Many important or innovative ideas just reach the market before the world is ready for them and end up having little or no impact initially. However, eventually some make their presence felt.

As a long-time follower of the thin client business, I’ve often felt this way about these devices. In case you aren’t familiar with them, thin clients are basic computing devices with limited or even zero local computing power and local storage, that rely on a network connection to a server or other computing host. Software is executed on this remote host and the results are sent back over the network connection and then displayed on the screen attached to the thin client.

Thin clients, which are arguably a rethinking of the mainframe and terminal concept, have been around in some form or other since the late 1990s. They have gone through numerous enhancements and revisions but have never really had the level of impact that many thought or expected they would. In many business environments, thin clients serve a critical function. However, their role is still more secondary and they’ve had essentially no impact on consumer devices—until recently.

In a classic case of “what’s old is new again,” we’re starting to see vendors leverage the thin client computing model in a variety of different devices, from wearables to smart TVs to connected cars. Of course none of them are calling their devices thin clients—because they don’t necessarily fit the “traditional mold” of a thin client and the term “thin client” has some baggage attached to it—but the principles behind the devices are the same.[pullquote]We’re starting to see vendors leverage the thin client computing model in a variety of different devices, from wearables to smart TVs to connected cars.”[/pullquote]

At the core of the thin client experience is the concept of virtualization, where pieces of hardware are essentially modelled in software, running in a different location, in order to make it appear that the computing is ha ppening at the thin client device. So, for example, in the case of some of the new connected car initiatives from Apple and Google, the car’s entertainment system may appear to be running software locally, when in fact, the software is running on a connected smartphone and the screen on the entertainment system is updated remotely.

Similarly, some of the applications for streaming TV from phones or tablets happens on the mobile device, and the “thin client” device connected to the TV, such as Google Chromecast or Apple TV, decodes the video signal and displays it on the TV. In the case of wearables, it’s expected that many of them will leverage the computing capability and screen of a nearby smartphone, just as thin clients require the greater computing power of a networked attached server.

In some ways, you can think of this as a new means of delivering a video signal to a display. In fact, the need for more standardization around various types of wireless video standards was the inspiration for my recent column on screenless devices (see “Screen Overload to Drive Screen-less Devices.”) But in the case of thin clients, it’s not actually just sending regular video—instead, the concept is called “screen scraping.” With screen scraping, the host device renders the video, much like a video card inside a PC, and then sends compressed packets over to the thin client, which decodes the packets and sends the signal over to an attached display.

The benefit behind the thin client concept is that you leverage the greater computing power of the host device and create a simple, low-cost client that can work alongside it. With traditional models, that host tended to be a high-cost server located in a secure data center, but in the new thin client computing models, it’s smartphones and tablets that have become the host devices. They now have sufficient computing power to drive these new types of flexible compute models and, of course, that capability will only grow over time.

Thin clients and virtualization have always been very powerful, intriguing technologies that seemed to offer the potential to radically reshape how and where computing occurs. Now that these technologies can be deployed in more mobile forms, it seems like their time really has come.

Published by

Bob O'Donnell

Bob O’Donnell is the president and chief analyst of TECHnalysis Research, LLC a technology consulting and market research firm that provides strategic consulting and market research services to the technology industry and professional financial community. You can follow him on Twitter @bobodtech.

13 thoughts on “Virtualization Reborn”

  1. Very nice column. Some thoughts…

    It’s very interesting thinking of Chromecast and Apple TV in thin client terms. In these cases the devices are essentially emulating DSP’s and other component level processors, instead of a general purpose instruction set. This puts the device closer to the TV as a component of the TV. Really cool.

    You’ve already pointed out that thin client services are a return of the mainframe mode. I vehemently agree. The cloud model is the same, but with a twist. When the cloud is viewed as a “peripheral” it fits very nicely with the idea of a “personal” computer. When the computer is the peripheral to the cloud anything truly “personal” is negated. Unless you own the cloud or mainframe (intentionally ridiculous). Unless the service provider develops all the services you may need, it comes back to a modern version of “picking up your printout”.

    There would be those that would claim that “most people” would never need anything more than that, and they would be right. Forgive the stroll down memory lane, but the PC Revolution wasn’t about “most people”, it was about “all people”. “Most People” from a device latitude perspective is the antithesis of “personal”. It seems that this cloud push is almost a “corralling”. This is far less of a concern on the appliance level (TV) but it does concern me on a more general computing level.

      1. So funny! And true! One must learn their tools to get the most out of them. We’ve progressed since then, but that doesn’t invalidate the “all people” aspect.

      2. I remember it the same way, too much complexity that required the care and management of nerds and geeks, and not truly consumer-facing. I could argue that the personal computing revolution really began with the iPhone and iPad. The iPad especially, it’s the first personal computer my 80 year old father can actually use reasonably well.

        1. Your 80 year old father, God bless him, is not all people either. Just as “geeks only” aren’t either. This also wasn’t about Apple, it was about client/server and the cloud.

    1. Thanks…yes, I’ve got a whole other column idea on the concept of cloud-based big data really being pretty similar to big brother, regardless who “owns” it.

  2. I’ve had an app on my iPad called CloudOn for several years. It is exactly what you describe, a virtual MS Windows machine. It works but is so clunky as to be practically useless on the iPad. Almost anything to do Word or Excel is better.

    Thin clients like a Chromebook might do better but I don’t think simple virtual machines running desktop software are going to be very useful for tablets in general.

  3. Wow, awesome weblog structure! How long have you been running a blog for?
    you make running a blog look easy. The full glance of your site is
    fantastic, let alone the content material! You can see similar here
    najlepszy sklep

Leave a Reply

Your email address will not be published. Required fields are marked *