For those of you who have not yet heard, earlier this month our product won Best of Interop in the Performance Optimization category. As you might imagine, this ratification of our technology has the whole team here at Talari pretty jazzed.
The other finalists in our category were Riverbed and Certeon – WAN Optimization companies both. Riverbed bills themselves as the IT performance company for networks, applications and storage. Certeon, too, focuses on how their virtualized WAN Optimization solution accelerates applications. The technology powering Talari's WAN Virtualization solution, by contrast, focuses on reliability and performance predictability, enabling customers to get more bandwidth at lower cost by leveraging public Internet connections for part or all of their enterprise WAN.
So how is it that our WAN Virtualization solution "beat" two WAN Optimization companies in a category titled "performance optimization"?
As judge Michael Biddick of InformationWeek Analytics wrote: "Performance of the WAN is key – but price is also part of the equation". This is spot-on.
Cost always matters. If it didn't, everyone would have at least 3 Gigabit WAN links to each of their locations, would they not? While the WAN Optimization folks are quite right when they note that bandwidth alone does not solve all WAN performance problems, it certainly helps greatly in many, many cases. You can never be too rich, too thin, or have too much network bandwidth.
"Backhoes don't obey Moore's law" is a true statement, but for the right - read: high enough - price, a carrier will run a fiber link anywhere for you. Even large enterprises obsessed with performance and doing huge first-time file transfers don't deploy enormous amounts of WAN bandwidth to all their locations - because it is simply too expensive to do. And for the great majority whose IT budgets are about flat, increasing spending on their MPLS networks by anywhere near what's needed to keep up with ~40% / year growth in WAN traffic demands is simply not possible, even as the carriers seem finally to be dropping their MPLS prices by 10% - 20% / year after going several years with pricing roughly flat. Yet as the move to private and public cloud computing architectures continues, even more WAN bandwidth is going to be necessary.
WAN Virtualization technology is quite complementary to WAN Optimization. Where WAN Optimization is ideal for solving speed-of-light latency issues, WAN Virtualization excels at combating the effects of packet loss and jitter (i.e. worst case latency), as well as frequently preventing loss and high jitter from occurring at the last-mile in the first place, in addition to cost effectively enabling much more bandwidth.
In fact, about two-thirds of our WAN Virtualization customers are WAN Optimization customers as well. We typically don't sell to the folks who've just deployed WAN Optimization in the last 6 - 12 months, any more than we sell to customers who've recently signed a 2 or 3 year MPLS contract without low minimum commitments. Rather, even if their focus is on providing application acceleration, since you can't add a second vendor's WAN Optimization technology to further optimize a network where WAN Opt already exists, for those folks who deployed WAN Optimization 18+ months ago, our WAN Virtualization technology is a great answer to "what do I do next?".
Cost matters in wide area networks, it's just that cost savings can't be at the expense of application performance and performance predictability, and we here at Talari Networks are grateful to the folks at InformationWeek Analytics for pointing this out.
