It’s hard to look at Network World or any other tech site without seeing a bunch of articles on software defined networks. Over the past couple of years, we’ve seen almost every network vendor, large and small, lay out its vision for SDNs and then back it up with new products to support the vision. One of the vendors that has been absent from the SDN tournament, though, is Huawei – that is, until this week.
Earlier this week, the giant Chinese equipment manufacturer threw its hat into the SDN game with a new switch series, the S12700 Agile Switch, specifically designed for migrating to an SDN. There are two products currently in the Agile Switch line – a big one (S12708) and a bigger one (S12712). From my briefing with Huawei, it appears that these products will be focused on implementing SDNs across the campus network rather than the data center.
The approach Huawei took in building these products is similar to the company that it’s looking to take a chunk out of, and that’s Cisco. Instead of using merchant silicon, Huawei is using its own custom chip – the Huawei Ethernet Network Processor (ENP). The use of its own ENP gives Huawei a number of unique features, such as unified wired/wireless management, remote broadband remote access server (BRAS) capabilities, and Huawei’s iPCA (packet conservation algorithm) monitoring service for looking at network-wide service flows. One might say that these are “proprietary” features, but there aren’t standards to define these yet. So sure, call them proprietary, but it’s not like there’s a standardized way to deliver these features. The “agility” of the switch also comes from the ENP as it enables programmability of the network to allow companies to create custom services quickly.
Under normal circumstances, the programmability and performance of these switches should put Huawei on the short list of any company looking to deploy an SDN in the campus. With 38TB of switching capacity and 384 10 Gig-E ports at full line rate, Huawei can go toe-to-toe with any vendor out there. However, these aren’t normal circumstances, as there’s a tremendous amount of misinformation and speculation in the media about what Huawei is and is not.
I’ve read in the media that Huawei is still a relatively minor player in the enterprise market, and though that may be here in the U.S., Huawei Enterprise did almost $2 billion in revenue last year. Sure, it’s not Cisco-size, but that makes it bigger than almost all of the other enterprise infrastructure vendors. I believe half of this revenue comes from its home market of China, with the U.S. market being the company’s smallest revenue contributor, which is why there’s this perception that Huwaei is a minority player.
Also, I’ve heard from a number of people that Huawei equipment is nothing but a cheap copy of other vendors, primarily Cisco - the tech equivalent of Japanese transistor radios of 70s. While it’s true that Huawei got caught once copying Cisco source code, that was a long time ago, and the company has since focused on building quality products. The Agile Switch line is a great example of Huawei engineering.
Lastly, there’s the issue of security and whether Huawei has created secret back doors to spy on the U.S. Obviously, I can’t verify whether Huawei has or has not, but I find it unlikely. However, the threat alone has been enough to keep Huawei off the radar of most U.S. companies. Last year, a Congressional report was released recommending that U.S. telecoms boycott Huawei equipment, and that created another flurry of media reports about the company that led to more misinformation. One company I talked to thought it was illegal for U.S. companies to buy Huawei equipment.
Herein lies the challenge for Huawei. To be successful, it needs to build its channel in the U.S., which is dominated by its biggest rival, Cisco. To build channel, the channel needs to hear customers want Huawei. However, customers won’t want Huawei without support from the channel. Being successful here in the U.S. is going to require Huawei landing a few big U.S. accounts to establish itself. The new products are a good start, but now the work really begins.