I find it fascinating how the communications industry likes to fixate on certain topics or themes and declare certain technologies dead, without much thought to how this might happen and the time it actually takes. Voice started the move to IP over a decade ago, but there's still a heck of a lot of TDM out there. Communications is moving to the cloud, but most customers still buy premises-based equipment. SIP was supposed to commoditize the IP phone, but customers are still paying a premium for feature-rich phones.
The latest wave to hit our industry is the notion that software will kill off all hardware based platforms. The video industry, in particular has come under tremendous fire as a bunch of software-based startups have emerged and are calling for the head of the hardware-based MCU. However, I think this statement is grossly simplified and, while software can bring unprecedented flexibility and portable scale for video applications, hardware platforms aren't going away.
First, I do want to make it clear, I'm not downplaying the value of software. If video is truly to become the new voice, software must play a key role in doing this. It's simply far too limiting to have hardware-only platforms for an infrastructure that seeks to enable ubiquitous video. So software-based platforms are here to stay and we're likely to see hybrid environments through 2014 and then software platforms taking on an even larger role moving into 2015.
The other thing software brings is a level of agility that can't be replicated with hardware platforms. Want to bring up more video? Spin up some more software based MCUs. Want to move the infrastructure to a new location? Then run the software on a virtual machine and migrate it over. Software creates agility; IT agility is a top initiative for almost every CIO today, and you can't have IT agility if the communications layer isn't also agile.
However, companies currently evaluating video solutions shouldn't quickly disregard hardware platforms. Despite advancements in off-the-shelf hardware and general-purpose processors, these platforms cannot outperform dedicated hardware platforms that have been optimized for those specific functions. The hardware engineers over at companies like Polycom and Cisco have spent thousands of hours making sure their products have the right amount of memory and processors and that the I/O is specifically optimized for performance.
Again, I'm not saying that software-based platforms don't perform well; most of them perform very well. I'm just saying that if density and performance of the video infrastructure is far more important than agility or cost, then a hardware solution is the right answer.
From a use case perspective, hardware solutions best meet the needs of demanding environments such as telemedicine, video call centers or any other situation where quality of experience is a critical factor for success. They are also still preferred in environments where high capacity is needed in a relatively confined geography (density) that might exist within a large enterprise that has close geographic demand, or a service provider that has large demand in a dense geography like NYC and the Northeast.
Ultimately, for most companies, "hardware"- versus "software"-based video isn't an "either/or"--both have relevancy, it's just understanding where each model fits best. If performance or density is an absolute must, then go with a hardware-based solution. However, if agility and flexibility outweigh performance, then a software model fits best.
By leveraging the strengths of both hardware and software, companies can finally get to a model where video can be pervasive across its user base. Ideally, customers would want to look for a vendor that offers both hardware and software platforms to ensure the solutions work together to offer consistent functionality.
Follow Zeus Kerravala on Twitter and Google+!
@zkerravala
Zeus Kerravala on Google+
