The 5G revolution is on its way for the consumer; however, looking for that new device in a store will be futile because the industry is still building the next-generation infrastructure. The 5G revolution means that mobile technology will go beyond connecting your phone or laptop. The new world of mobility will power the Internet of Things and technology such as connected cars, smart cities, factories and things that we can’t even dream of yet.
“At Mobile World Congress 2017, we were talking a lot about what 5G will enable. All of that requires way more bandwidth, very low latency and a much better responsiveness at the endpoint near the device or the user,” said Sandra Rivera (pictured), corporate vice president and general manager of the Network Platforms Group at Intel. “All the innovations from a radio perspective will enable 5G, but of course, the rest of the infrastructure has to support it as well.
Rivera spoke with host Jeff Frick (@JeffFrick) and guest host Scott Raynovich (@rayno) of theCUBE, SiliconANGLE Media’s mobile live streaming studio, during the Open Networking Summit in Santa Clara, California. (*Disclosure below.)
This week, theCUBE features Sandra Rivera in our Women in Tech feature.
The foundation for building 5G
While it is true that 5G is happening now, most of the work is taking place behind the scenes. The industry is building networks and devices to support and scale, with the new technology is not quite ready to go-to-market, Rivera explained. The goal for 5G networking is to have a true convergence of computing and communications. Intel is using its capabilities in the cloud and in the core of the network to scale out to both the edge and the access network.
The idea of bringing programmable, scalable and flexible computing closer to those endpoints is the foundation for building 5G. Therefore, Intel is driving the orchestration with Software-Defined Networking and Network Function Virtualization, according to Rivera.
So while 5G networks continue to develop, Intel is investing now with a broad ecosystem of partners. Rivera explained that the mobile broadband standard 3GPP for 5G would not be ready until the end of 2018. Once the protocols and requirements are in place, she noted that true compliant 5G devices would arrive in 2019 and ramp and scale by 2020.
“I think that there’s always some healthy skepticism about, ‘Are we over investing? Are we investing too early?’ But if you look at the amount of work that we have to get done in what is a relatively short window of time, we feel like we need to speed up,” said Rivera.
Intel’s strategy for 5G is to construct an end-to-end platform beginning with modem technology that will go into client devices, such as smartphones and tablets. However, Rivera also believes Intel needs to prepare for drones, robots and cars, and future devices that come along will also need to connect to the platform.
Intel powers most of the world’s cloud infrastructure, so the company is architecting the 5G infrastructure to process IoT data at the edge of the network and in the core of the network, where the switching, routing, authentication and the security functions will operate. Once the cycle of processing data is complete, the data will go back into the cloud and the datacenter, Rivera explained.
“That’s Intel! It really is end-to-end. We have this broad view and the scalable architecture where it’s a consistent, common tool chain and a very broad access to the ecosystem. Developers [will have the ability] to take you through an end-to-end portfolio of services and capabilities that you require,” Rivera said.
Open source and the developer community contributions
During the Open Networking Summit, Intel also doubled down on its commitment to developers by announcing that it was moving its Data Plane Development Kit to The Linux Foundation to share with other developers in an attempt to accelerate development and commercial adoption. Invented in 2010, DPDK contains a set of libraries and optimized drivers for running high-performance packet processing, general purpose CPUs.
If you’re in the network business, it’s all about moving the packets, so you need high-performance packet processing,” Rivera said.
DPDK offers optimized libraries for cue and buffer management and well as flow classification for quality of service. Rivera also noted that it could run on a standard server CPU, which is a very powerful capability as it means there is no longer a need for purpose-built silicon to run those functions.
“Intel is a big believer in open source, open standards and a big enabler and investor in broad ecosystems,” Rivera said.
The company is consistently the number one contributor to many of the projects, including the actual Linux kernel, Rivera added. From Intel’s networking project perspective, the leadership of The Linux Foundation is the right choice.
“The Linux Foundation is demonstrating and coalescing the industry around some of the big problems and challenges, as well as opportunities, that we face together,” Rivera concluded.
Watch the complete video interview below, and be sure to check out more of SiliconANGLE’s and theCUBE’s coverage of Open Networking Summit. (*Disclosure: SiliconANGLE Media’s theCUBE is a media partner at Open Networking Summit. Neither The Linux Foundation nor other conference sponsors have editorial control over content on theCUBE or SiliconANGLE.)