Becoming an Internet Architecture Board Member
At the beginning of 2019, I was honored to become a member of the Internet Architecture Board (IAB) of the Internet Engineering Task Force (IETF). IETF in IP industry is equivalent to 3rd Generation Partnership Project (3GPP) in mobile network, and an IAB member is equivalent to the vice president of the Project Coordination Group (PCG) of 3GPP. I have been engaged in IP R&D since I joined Huawei in 2000.In 2009, I started to participate in IP innovation research and the promotion of IETF standards. It took me 20 years to become an IAB member. These 20 years also map the journey of Huawei router products, from unknown products to the top of the global carrier, which demonstrates how much Huawei has progressed in the IP domain.
In 2005, the product line started investing in the MPLS solution’s R&D. I was responsible for design and development of a series of MPLS features. Just when we were about to complete the development of the Fast Reroute (FRR) feature, a major defect was found: when a link failed, traffic could be properly switched to the backup link within 50 ms; but, after the link recovered, it took one second to switch the traffic back to the original link. After some troubleshooting, it became clear that we hadn’t considered the service logic of traffic switchback during the design process. As a result, the FRR feature became totally unavailable and needed a redesign. In 2006, MPLS-TE primary/backup protection feature received several major complaints because we hadn’t considered the need for secondary protection. When the primary and backup paths were both interrupted continuously, VPN traffic was interrupted continuously. I have worked on numerous features during my career over a decade in MPLS R&D, but these two complaints are more significant. As I’ve always said, setbacks are part of life.
Every time we slipped up, we strived to rise again. We gradually added features and deepened our understanding of networks so that we could be more aware of customer pain points and innovate accordingly. We proposed a challenging goal to provide uninterrupted MPLS tunnels. To achieve this goal, we developed the best-effort path feature of MPLS TE tunnels and allowed users to specify multiple MPLS TE tunnels for load balancing in a tunnel policy. We also allowed services to be switched back to MPLS LDP tunnels when MPLS TE tunnels are interrupted. In the MPLS TE FRR solution, we developed the feature of delayed switchback upon link recovery to ensure zero packet loss. These features played an important role in the key project development for a multinational carrier in Spain. The customer found that our design was more complete, easier to use, and more intuitive than competitors’ offerings.
We will always face hardships and setbacks, but they are essentially the accumulation of experience that allow us to grow and give us a better understanding of service scenarios and technologies. With new technological opportunities such as SRv6, we are in a good position to quickly innovate and formulate standards, taking a lead in the development of next-generation IP technologies.
Long-term accumulation of these hardships is the basis for innovation. As individuals or organizations, we need to continuously improve ourselves and stay ready.
The industry recognizes the dedication of Huawei employees. A can-do attitude is pivotal to our success. However, when it comes to divergent technological innovations in the IP industry, it is critical to select a technological route. In the past, Huawei’s R&D was driven by market projects. To try to win bids for projects, marketing engineers would receive numerous requirements, making technological development erratic and inaccurate. R&D faced great pressure in delivery and was overburdened. After long-term accumulation, we are now able to understand the network and customer pain points and must learn to make strategic efforts.
At the beginning of SRv6 technology development, the product line was divided on technology selection. In 2017, we spent a lot of time discussing whether to use SR-MPLS6 or SRv6, which was still quite new. Even after some SRv6 features were delivered, Kevin Hu (President of Huawei’s Data Communication Product Line) held multiple meetings for discussions. Through several rounds of discussion, we gradually came to understand that SR-MPLS6 was just a supplement of the MPLS feature. An IP network strategy oriented towards 5G and cloud can be built only on SRv6. With this realization, we reached a consensus and made every effort to implement SRv6.
R&D is difficult for such emerging technologies, but what’s more challenging is keeping up with market expansion and industry guidance. In the past, Cisco led the IP industry, and others followed. This was the first time that Huawei took the lead in a technical transformation. We defined application scenarios, aligned with customer strategies, and built an industry ecosystem. We started from scratch and advanced cautiously, but the whole journey was full of excitements.
More importantly, we jointly developed technical innovations with customers to bring benefits of SRv6 to customers, which were well recognized. We had deployed nearly 20 SRv6 commercial networks in 2019. Under the guidance of the expert committee for promoting large-scale IPv6 deployment, we set up an IPv6+ technology innovation team to integrate new technologies such as SRv6 into China’s IPv6 industry strategy.
To further promote industry consensus, we have organized multiple SRv6 industry forums. Because of unforeseen circumstances, the second forum was held a week ahead of schedule. We got the draft of the presentation slides only two days before the meeting. We had to align and communicate with the speakers day and night. Working so much overtime gave a sore throat the night before the meeting. At this time, the meeting affairs team leader called and wanted to determine the seat arrangement for the guests. Because I had a speech the next day, I asked her to go through the possible arrangements, and I responded only with “yes” or “no,” to protect my voice. After the forum, a familiar feeling arose. After working at Huawei for a long time, I feel that our work is often like a glass — smashed to pieces suddenly and placed in front of us, and then we put it back together piece by piece, until at the last moment, the glass was finally ready. Able to be stitched together and displayed beautifully in front of the world.
At the end of 2006, the MPLS feature started to be widely used in the market. We needed to add many features features. However, because of organizational changes, the MPLS development team had only two experienced employees. I had to find a workaround and struggled to cope with the situation. I even felt a bit depressed at one point. At the retreat meeting for the product line, I finally got angry and complained about the difficulties brought by the organizational changes to the MPLS feature. Chen Jie, my colleague and old friend, comforted me and said, “I understand you very well. You are upset because of the passion you have for MPLS.” The word passion struck a chord with me. During this intense period of work, technologies and products seemed to bring us endless trouble and pain. But when I heard Chen Jie’s words, I felt positive, even moved. In the days that followed, this feeling encouraged me, and we worked hard to re-build an MPLS team. I was later transferred to different positions focusing on different fields, such as solutions, key account development, planning, architecture, standards, research, and industry. Every time I got a new position, I would focus on team development, and these team members have become the backbone of Huawei IP.
In 2010, when promoting innovation in IP protocol, I communicated with a predecessor at the Huawei US Research Center. He told me that Cisco’s success was attributed to a close combination of products, standards, and markets. In contrast, Nortel Networks proposed many good innovative ideas, but its products failed to keep up. That’s why NT failed. Alcatel-Lucent has done a good job developing products and solutions, but has limited influence in developing standards. I was deeply impressed by these words. How do we organically combine Huawei’s products, standards, and market? David Wang later became President of the Network Product Line, where he also pointed out the shortcomings of IP standard innovation. The product line strengthened the operation of the IPSSC and established project teams for key standards, such as SRv6, telemetry, and 5G transport. The team members are key to areas such as pre-research, standards, products, solutions, marketing, and sales, where they form a joint innovation force. The product line developed the NetCity mechanism, carried out joint innovation with key customers, and made products and solutions more competitive through fast iteration. In 2019, the Genome in a Bottle (GIAB) mechanism was launched. Experts from different fields, such as product management, solution architecture, and standards, were arranged to hold scenario-specific and future-oriented strategic workshops with key customers. These moves have made our work more organized and effective.
While promoting SRv6 innovation, our team members contributed innovative ideas in many emerging fields, such as network slicing, iFIT, DetNet, SFC, BIER6, and APN6. I was always excited to communicate with team members about their ideas, and these innovative ideas have gone beyond the scope of SRv6. A strategic blueprint for IPv6+ has gradually become clear, along with a huge opportunity for the 5G and cloud era. I feel strongly about the strength of the team. I’m glad that we have a strong team that paves the way for future innovation and development.
In 2008, I left the familiar SE position and went to Europe for business development with Tier-1 carriers. I was full of anxiety at that time. When I spoke with customers, they would sincerely tell me, “Do not look down on yourself. Our relationship is not ‘we give requirements and you develop products and solutions based on them.’ We value Huawei’s status as a multinational company, and hope to learn from your industry experience and get some guidance from you.” When I participated in the IETF standards, an industry predecessor from Huawei’s research center in Canada told me, “Huawei must send its best IGP and BGP experts to the IETF. The language barrier isn’t a problem. The most important thing is the technology. Huawei supports countless large-scale networks all over the world. Such experience is unparalleled.” Boosted by these words of encouragement, we have made the impossible possible on the road of protocol innovation.
In IETF meetings, a person can typically present a maximum of three to five topics. To open up innovation of IP standards, I made 15 presentations at the IETF 88 meeting in 2013, covering almost all important working groups in the routing field, talking to experts in various fields. As we have published many articles and presentations, the peers who attended the IETF 88 meeting said that Huawei’s innovation in the routing domain with its lack of innovation is going to have a big impact. Over the past few years, Huawei has made significant contributions to the standardization of SDN transition, SRv6, telemetry, and other fields. These contributions have been well received by peers. In early 2019, I was honored to become a member of the IETF IAB, the first IAB member from Huawei and even from an Asian enterprise. What’s more, I’m glad that our work in IETF has attracted the attention of customers, especially our innovation in the strategic direction of IPv6+. Customers are impressed by Huawei’s new ideas and feel that Huawei has changed.
In 2019, Huawei maintained its leading position in the global carrier IP market. IPv6+ innovation and standardization were in full swing. Huawei’s commercial deployment of SRv6 far exceeded that of competitors. We are giving it everything we’ve got to lead in innovation, and are assuming major responsibilities to promoting IP technology innovation and continuously contributing to the IP standards community. A multitude of IP heroes have progressed their way up from zero. They have matured, gained confidence, and become more firm. Our young colleagues may be as hesitant as I once was, but I’m sure a new generation of IP heroes will emerge among them.
We are facing one of the best development opportunities in history. Let’s strive ahead with confidence.