Author | Liu Baodan From performance to market confidence, Meituan is walking out of a three-year low point, but Wang Xing is not stopping there - he has even bigger plans. Going overseas has become a must for Chinese companies. Meituan, which has been warming up for 8 years, has finally made up its mind to put going overseas on the agenda. Recently, Meituan began recruiting senior engineers for international silver enterprise direct connection. After the model was successful in the Hong Kong market, Meituan officially kicked off its overseas expansion, accelerated recruitment and put the first stop of the overseas expansion in Saudi Arabia in the Middle East. Going overseas is a critical turning point, which means that after more than ten years of capacity accumulation, Meituan has to export its local life capabilities to the world, which is as significant as the replication of TikTok by ByteDance. In the wave of Internet companies going overseas, Meituan went overseas later because local life patterns are more important than social, e-commerce and other industries. However, Wang Xing must make this move. Against the background of intensified domestic competition and the shrinking of community group buying, he must find a new growth story. On his entrepreneurial journey, Wang Xing is still determined to create a new business legend in this global adventure. A must-have question. Meituan has fought a beautiful takeaway battle in Hong Kong. On May 6, Measurable AI, a market research firm, released the latest data showing that by March 2024, according to the number of orders, KeeTa, the takeaway business of Meituan in Hong Kong, has a market share of 44%, rising to the largest takeaway platform in Hong Kong. However, Hong Kong is only a stopover for Meituan's overseas expansion, and Meituan has set its real meaning of going overseas in Saudi Arabia. Wall Street news learned that Meituan has been recruiting people around the direction of going overseas in the past two months. The positions include engineers, overseas human resources and operation experts, international payment and transaction product managers, mainly responsible for payments, employee management and related products in overseas markets. More importantly, the recruitment of local talents. More than a month ago, Meituan posted relevant recruitment information on LinkedIn and the Middle East recruitment platform Baye.com, with Riyadh, the capital of Saudi Arabia, as the place of work. From the city selection, Meituan did not choose the United States with a larger market space, nor did it choose Southeast Asia where culture and food are more similar, but chose Saudi Arabia. It can be seen that Meituan's overseas expansion strategy still has a heavy experimental component and is more cautious. Wang Xing is not fighting an unprepared battle. For this overseas expansion, Meituan has been planning for many years. As early as 2016, Wang Xing began to consider the issue of going overseas and visited Silicon Valley, Berlin, Israel, Jakarta and other places. In 2017, Meituan officially laid out overseas accommodation business, first connecting hotels in nearly 100 countries overseas to the Meituan application. At that time, the domestic and foreign takeaway wars were in full swing, and with Meituan's listing in Hong Kong in 2018, Wang Xing's overseas strategy was forced to be shelved. Since then, Meituan has also made a series of international investments, including Swiggy in India, Gojek in Indonesia, and Opay in Nigeria, involving food, taxis, payments and other fields, to prepare for going overseas. Along with the frequent news reports of Meituan's victory in Hong Kong, Meituan's overseas plan was finally brought to an unprecedented strategic height in 2024, and Wang Xing once again rushed to the forefront. In February, Meituan put the home business group, the in-store business group and other businesses into the core local business sector, and appointed Wang Putong as CEO, while Wang Xing personally took charge of overseas business, which ensured the landing of the overseas expansion strategy in the organizational structure. In fact, before the confirmation of the overseas expansion strategy, Wang Xing personally visited the Middle East last May and met with members of the Saudi royal family, laying the foundation for Meituan's layout in Saudi Arabia.
In today's weather is good. Today's weather is good.
Facing the certainty opportunity brought by AI, Lenovo has put forward the vision of AI for All, and computing power is the most important part of it.
Recently, Lenovo has officially released a new computing infrastructure with servers, storage, data networks, edge computing, including the new generation server Lenovo QLoud WR5220 G5, NetApp AFF A all-flash series, medical emergency and disaster recovery integration solutions, Lenovo QLoud 100G core switch and so on.
This release further enriches Lenovo's "One Horizontal and Five Vertical" AI-oriented infrastructure map. Through Lenovo's development of computing power, it can be known that Lenovo's computing power is entering a stage of rapid development.
In August of last year, Lenovo officially released the AI computing power strategy, proposing to build an AI-oriented computing infrastructure. In February of this year, Lenovo first announced the "One Horizontal and Five Vertical" strategy at the Mobile World Congress (MWC) in Barcelona, revealing Lenovo's development blueprint for computing infrastructure. "One Horizontal" refers to Lenovo's Wanquan heterogeneous intelligent calculation platform, and "Five Vertical" includes servers, storage, data networks, software and hyper-convergence, and edge computing infrastructure products and solutions.
In April, Lenovo officially launched the Wanquan Heterogeneous Intelligent Calculation Platform, which integrates Lenovo's five technology innovations and is the infrastructure base for AI 2.0 model training and inference. Chen Zhenkuang, Vice President of Lenovo Group and General Manager of China's Infrastructure Business Group, said that the platform is the core of Lenovo's infrastructure strategic framework in the AI 2.0 era.
With the release of these new products, it will undoubtedly accelerate the landing of Lenovo's computing infrastructure.
Take the new generation server Lenovo QLoud WR5220 G5 as an example, it is a 2U2S server, providing computing power for scenarios such as cloud computing, big data, AI in medium-sized data centers, virtualization, online trading, high-performance computing, key business flows, and business collaboration.
For example, the new NetApp AFF A all-flash storage series and other products can help generate AI, virtualization, and enterprise databases, and other customer IT workloads, with performance improvements of up to 2 times, and data availability verified as 6 nines.
Behind the accelerated layout, Lenovo is targeting the market opportunities brought by AI artificial intelligence technology iteration, especially the booming demand for AI computing power. IDC predicts that by 2026, China's smart computing scale will reach 1271.4 EFLOPS, with a five-year compound growth rate of 52.3%, while the compound growth rate of general computing power during the same period will be 18.5%.
As there are two sides to every coin, behind market opportunities are challenges. Major technology companies around the world are betting on AI, and Lenovo must continue to enhance its strength to win more market share in the AI computing power market and cope with fierce market competition.
Lenovo is one of the most comprehensive companies in AI layout, from AI intelligent terminals, AI-oriented infrastructure to AI-native solution services. According to the financial report, the company's R&D expenses will reach 14.8 billion yuan in 2023, and research and development investment is about 6 times its pre-tax net income, indicating its determination.
As a technology company with a 40-year development history, Lenovo has successfully passed through multiple technology cycles. Whether the company can reproduce the glory of the past in the AI era, time will give the answer.
The following is a record of the exchange (after editing):
Q: What requirements does the rapid development of AI technology, represented by large models, pose to China's computing infrastructure?
Chen Zhenkuang: Computing power is a core driving force in today's AI era. At the end of 2022, the large model ChatGPT appeared internationally, which triggered the demand for the entire large language model, and achieved rapid development in AI computing power. In the past, computing power was based on general computing power. Large models have enabled artificial intelligence to enter the 2.0 era, with an annual demand growth rate of 10 times for computing power.
The entire computing infrastructure still cannot fully meet the demand for computing power, especially with CPU-based computing infrastructure. New acceleration computing computing infrastructure such as GPU and TPU has received great attention. In the past year and a half, accelerated computing has grown from around 20% to over 1/3 last year, which is the development of computing power.
Q: How will China's computing power industry develop in the future? How will Lenovo be ready to embrace this trend? What are the layouts?
Chen Zhenkuang: Computing power not only requires servers, but also places more stringent requirements on high-speed storage and data scheduling. The scheduling and coordination between computing nodes and clusters have high requirements for high-speed connection networks. To support large-scale training, inference in various scenarios, computing power elasticity is also essential. At the same time, energy-saving and heat dissipation requirements are also very high.
We first proposed the "One Horizontal and Five Vertical" strategic framework this year. The "One Horizontal" refers to Lenovo's Wanquan heterogeneous intelligent calculation platform, which can help our users more efficiently schedule computing power when using computing infrastructure. At the same time, the heterogeneous intelligent calculation platform can perform GPU virtualization, support collective communication algorithm libraries, and achieve power failure resumption of training for minutes. It can unify scheduling between traditional HPC computing infrastructure and AI computing infrastructure across clusters.
"Five verticals" includes server, storage, data network, software and hyper-converged, edge infrastructure products and solutions. "Five verticals" already has a very comprehensive layout at Lenovo in the past. But we are reorienting it with AI as the guide, rebuilding it, and upgrading it comprehensively to make it more capable of developing computing power in the AI era with artificial intelligence and full AIization.
Q: In April, Lenovo and the China Intelligent Computing Industry Alliance launched the Heterogeneous Computing Ecological Industry Alliance. What are the plans for the future? What promotion for the Heterogeneous Computing Industry Alliance?
Chen Zhenkuang: As a leading enterprise in the calculation industry, Lenovo has a responsibility to promote the development of the entire industry. I think that members of the industry alliance contribute to different dimensions at three levels.
The first level is AI chip. There are many good AI chip companies at the international level. There are also many domestically. In particular, in the past year or two, they have become increasingly mature on the inference side and increasingly advanced in training.
The second level is system and equipment. Like the role Lenovo plays, we apply different computing power of chips based on different application scenarios, training and inference scenarios. We apply these chips more efficiently to contribute to the industry.
The third level is software and application. For example, big model companies, including ecological companies such as making enterprise and industry models above big models.
The three levels of chip layer, system equipment layer, and application software layer are indispensable. The entire industry ecological alliance also needs to exert the capabilities of all members in several different dimensions, set standards together, and integrate better with each other.
Q: What are the differences in the positioning of Lenovo Wentai and Lenovo ThinkSystem four servers this time? What are their respective characteristics? What are the technological upgrades compared to the two Wentai servers released at the computing power conference last year?
Chen Zhenkuang: Today we mainly released four servers, two Lenovo Tian servers and two Lenovo ThinkSystem servers.
The two Lenovo Wentai servers were released last year for the next generation of products. The first one is WR5220 G5, which is currently the most used product in the market and has higher performance while being low power consumption. Compared with last year, we have integrated more smart management engines to reduce unplanned shutdowns caused by components and reduce the frequency of component replacement.
The second one is WA5480 G5, which is a server specially designed for AI training and inference. It supports up to 10 double-width PCA GPUs and can currently support a variety of international and domestic GPUs of various brands. It is the most suitable server for AI training and inference in AI scenarios.
The other two are Lenovo ThinkSystem servers. One is SD520 V4, which is a high-density server, suitable for high-density deployment, high-performance computing, and so on. The other is Lenovo ThinkSystem SR630 V4, which is a 1U2CPU model. The 1U server is mainly used in cloud computing and computing leasing scenarios.
Q: What is the future market plan for Lenovo's All-Inclusive Heterogeneous Intelligent Computing Platform?
"One horizontal" is a comprehensive upgrade of the "five verticals" capability. It is to restructure and upgrade the underlying infrastructure that we have laid out comprehensively in the past for AI-oriented purposes. It will be ubiquitous.
Q: Lenovo's server has grown very rapidly in recent years. What is the reason behind it? In addition, we see more companies doing server business. Where is Lenovo's advantage?
IDC's "China x86 Server Market Report for the First Quarter of 2024" shows that Lenovo's x86 server sales surged 200.2% year on year, and its market share has risen to the third place in the domestic market.
This year is the beginning of Lenovo's fifth decade. Lenovo has gone through a cycle of entrepreneurship, the PC era, the international era, and the diversified era in every decade. Our product services, whether it is PC or infrastructure, are available in more than 180 countries and regions around the world. The capability constructed is a comprehensive product capability of "end-edge-cloud-net-intelligence". Lenovo's entry into the top three is the result of the steady promotion of the AI-oriented infrastructure strategy.
Q: How is the overall development of China's AI server market? What challenges does it face? This time, we launched new full-stack computational infrastructure products. What are the highlights that you can share with us?
The Chinese accelerated server market size is expected to reach US$ 9.4 billion in 2023, with a 104% increase compared to 2022. Domestic chip performance has significantly improved, and the relevant industry chain has gradually formed. In the meantime, the application scenarios for artificial intelligence chips are becoming more diversified, and targeted artificial intelligence chips are being invested in more precise market segments.
Lenovo's new generation of servers are specially crafted for our customers, with three core values of outstanding computing power, extreme energy efficiency and AI intelligent operation and maintenance. Firstly, in terms of performance, the number of processor cores has increased by 2.25 times, and the performance of AI loads has improved by 2 times. In the cloud server application scenario, the output performance per cabinet has increased by 42%; memory bandwidth has increased by 14% and fully supports CXL 2.0, with E3.S capacity doubling. In terms of energy efficiency, the energy consumption per processor core has been reduced by 70%, and the entire line supports liquid cooling mode. By covering 98% of power-consuming parts, the data center PUE (power usage effectiveness) has been reduced to below 1.1. In terms of AI intelligent operation and maintenance, specific key component logs, such as memory and hard disk, can be intelligently analyzed to effectively avoid or reduce the frequency of component failure, thereby reducing unplanned downtime for customers.