2016 Predictions For Datacenter Networking


(Image source: Morgue File)

No analyst is allowed to get past January 1st without posting their list of predictions for the new year, so I have no choice but to jump into that pool. 2015 was an interesting year for networking, and 2016 will most likely continue that trend. We’ll see a year where the stalwarts like Cisco Systems, Juniper Networks and Alcatel-Lucent Enterprise will see additional challenges from non-traditionals. We’ll see new networking strategies emerging. And we’ll see consolidation as competition heats up.

  1. New competition. While networking has been fairly staid in the past twenty years, the vendor mix is shifting a bit. While shipments are up, the revenue is down, reflecting the reality of tighter competition among both the primary vendors as well as the secondary vendors (and new entrants). Non-traditionals like Hewlett-Packard Enterprise are picking up steam, and even Huawei might be able to break into North America, the largest market, in 2016. Change will happen at the switch level long before it gets to the core. Expect to see competition continue to be tight as vendors battle for share of a market in transition. Does this mean that investors should short stock and customers should start ripping out equipment? Absolutely not. But it does mean the purchases should be scrutinized and negotiations should be more thorough as there may be more options than ever before.
  1. Software-Defined Networking (SDN). OK, I’ll take the bullet here. If you asked me about SDN in 2014, I’d have told you that 2015 would be the year of SDN, but I would have forgotten the asterisk. 2015 was the year that everyone talked about SDN, but current networking infrastructure really is preventing the shift from happening quickly. So while the market is headed in that direction, it is taking a little longer than most had expected—but it is still happening. 2016 will see more SDN adoption, but the real action is going to be in pockets, mostly around new emerging workloads like Big Data, private cloud and IoT that benefit from different network architectures. Expect to see continued discussion ramping up on SDN, some possible new entrants to the market and more customers taking it seriously—even if they aren’t starting deployments just yet.
  1. Network Function Virtualization (NFV). Expect to see more traction here (over SDN). Because this is being pushed by carriers and they will be the primary customers, the acceptance is further ahead and deployments are beginning to happen. The proof of concepts (POCs) are typically being driven by actual customer environments which means once the concepts are proven out, commitments will be made. While NFV is the smaller opportunity for changing network trajectories, it will win the race and see more deployment action in 2016.
  1. Litigation. Cisco Systems successfully defended its patents against Arista Network’s infringement in 2015, and it is probably a safe bet that others will take to the courts in 2016 as well. While the market is not generally known for litigation, the dynamics of a more competitive market will probably bring us additional litigation in 2016—especially against a backdrop of changing infrastructure and new technology. When new technologies come to market or are beginning to be productized, lawsuits typically follow. This is not necessarily a bad thing as establishing patent ownership can help remove uncertainty for customers on the fence about buying into new technologies.
  1. Changing structures. Because of the changes we are seeing in network traffic—more East-West, server-to-server traffic and less North-South server-to-core traffic—we will probably see more people adopting leaf/spine topologies. Some of the new networking products (like VMware’s NSX) lend themselves well to this type of decentralized architecture. But don’t expect that to crush the traditional core-centric network topology. Instead this will happen more along the edges and with the deployment of new workloads like Big Data. Over time we expect leaf/spine to take a greater role in the network, but for 2016 it will be seen more in pockets—but the trend is clear.
  1. Open networking. Just as open source has played an important role in the development of the internet and now cloud computing, open source has an important part to play in networking. Splitting up the traditional “black box” into hardware and software gives customers a little more flexibility, but most importantly it will give them more control. As 2016 brings more complexity, look for customers to investigate more open networking solutions as a way to try to insert the right services into the right places in order to make all of this stuff actually work.
  1. Consolidation. No, not network consolidation (that is happening regardless), this is vendor consolidation. In 2015 we’ve seen acquisitions (Hewlett-Packard Enterprise and Aruba) and partnerships (Cisco Systems and Ericsson) as the market streamlines. We anticipate that 2016 will bring additional consolidation as larger players with deep pockets and smaller startups with interesting technology do their mating dance. Consolidation can have some positive benefits as the new technologies get more stable financial support from the acquirers. We don’t expect to see major players combining; the Cisco Systems/Apple model is probably the smarter route as there are already too many large tech mergers clouding the IT waters.

After a relatively interesting 2015, it feels like 2016 could bring even more changes to networking. Companies like Alcatel-Lucent, Cisco Systems, Dell, Huawei, Hewlett-Packard Enterprise, Juniper Networks and even VMware should have plenty to say in 2016, so fasten your seat belts as it should be an interesting ride.