- At its annual Connect conference, Huawei set down its plan to become one of the five dominant public cloud platform providers, opposite IBM, Google, Microsoft, and Amazon.
- Huawei’s cloud ambitions, however, aim not to dominate but to create an open, independent platform that augments and works with other clouds while maximizing differentiated Huawei functionality and expertise.
What is a computer? Is the cloud a computer and vice versa? In many ways, yes. Both a computer and the cloud represent a programmable resource, for example. Both dole out capabilities in the form of services. And both are finite in their scale and bound to the purpose of those who program them. Sure, the cloud can be seen as a never-ending cluster of computers slung together. But both, at the end of the day, return zeros and ones in exactly the same way.
If you’d asked me that question prior to attending Huawei’s annual Connect conference, I probably wouldn’t have been willing to make that kind of connection. During his opening keynote address, Guo Ping, Huawei’s current acting CEO, changed my mind when he rolled out the often-cited ‘mother of all failed predictions’ made by IBM’s Thomas Watson back in 1943:
“I think there is a world market for maybe five computers.”
Everyone of course duly chuckled. But, he then proceeded to explain why Mr. Watson was right – assuming of course that you replace the word computer with cloud. Though not named outright, Mr. Ping certainly was referring to hyper-scale public cloud providers IBM, Google, Microsoft, and Amazon. So, which company is the fifth cloud? No surprise, Huawei wants to be that fifth provider. But it intends to do so in a fairly unorthodox manner.
Huawei does not intend to build an ecosystem of customers and partners through differentiated applications (a la Microsoft Azure); nor does it want to dominate by consuming and housing data (as is the case with Google and Amazon). Instead, the company intends to succeed by first building a comparable cloud in terms of scale and reach (a natural first step) and then opening that platform up to other clouds, to create what the company refers to as ‘intelligent connections’ between devices and the cloud and between clouds themselves.
Bear in mind that this view isn’t akin to Cisco’s proposed cloud of managed clouds (Cisco InterCloud). Huawei simply wants to create an open, independent platform that augments and works with other clouds while supporting use cases and markets where Huawei can offer differentiated functionality. This means a strong emphasis on the global telecom services business and further expansion close to home in support of large governmental and private organizations, both of which are reflected in customers such as China Telecom and China Central TV.
Can Huawei really match its cloud rivals in scope and scale? I don’t think so; at least, it does not seem likely that Huawei can catch up with the likes of Microsoft and IBM in delivering differentiated functionality such as artificial intelligence (AI). The company’s newly unveiled AI capabilities, branded as Huawei Enterprise Intelligence (EI) services, certainly move the company in the right direction, but when compared with a family of AI services from the other four cloud providers, it’s clear that Huawei is just beginning its journey.
Does that matter though, given Huawei’s open, partner-centric desire to create intelligent connections between clouds? I don’t think so. Huawei is welcome and able to make use of the capabilities provided by strategic partners like Microsoft and SAP, filling any existing technology gaps. The company has also shown itself quite willing to make use of open source software as a means of achieving such interplay.
In my opinion, where Huawei is most likely to succeed with its cloud play is in helping partners and customers apply Huawei’s significant hardware expertise to trenchant problems like cross-cloud security, AI-scale hardware, and IoT edge computing. A case study in the latter of these two can be seen in the newly announced Schindler CUBE edge device. Built together with Huawei, the Schindler CUBE uses AI routines to provide real-time analytics and filter data for connected elevators and escalators.
If Huawei can scale and replicate this sort of partner-first approach in supporting Schindler’s internal efforts, I think it stands a good chance of securing a long-lasting position as a hyper-scale public cloud provider. I imagine we won’t have to wait another 74 years to see if Mr. Ping’s reappraisal of Mr. Watson’s initial prediction plays out.