Shulou(Shulou.com)11/24 Report--

2022 is coming to an end.

This year, the annual sales growth rate of new energy vehicles reached about 90%. The coming 2023 is called the "first year of smart car competition". At this time, both car companies and cloud manufacturers have set their sights on the most important tool of smart cars-car cloud.

In the intelligent second half of new energy vehicles, data has gradually become a new "fuel" for cars. How to efficiently mine and make use of the value of data is very important for automobile companies. Data mining and processing are inseparable from the powerful computing power, and its importance, to some extent, is equivalent to the lithium resources in the first half of electrification.

Data mining and processing are inseparable from the support of cloud, and the computing power also comes from cloud computing. Car cloud has become the key to the intelligent second half. The car cloud will grow very fast in 2022, with revenue expected to be 2.5 times that of 2020, according to 21CBR.

From the perspective of the cloud computing industry, automobile cloud has become a must for the majority of cloud manufacturers. In the second half of 2022, cloud manufacturers such as Ali, Tencent, Huawei and Baidu "refreshed" their car cloud strategies in various forms.

This paper attempts to explain three problems:

1. How important is the car cloud?

two。 How big is the car cloud market?

3. Who are the players in the automotive cloud industry?

Importance 2023 will be the first year of intelligent competition for new energy vehicles. In the intelligent era, whether it is intelligent cockpit or autopilot, the amount of data generated by the car will increase exponentially. According to Garner estimates, when autopilot, lidar, cameras and other sensors "observe" the road, at least generate 4TB data every day. In the face of these data, car companies need to find ways to store and process them.

In the testing phase, which is not fully commercialized, processing data is not a big problem because there are not many vehicles involved. Some autopilot companies even use hard drives to copy data and then mine data. However, once the test vehicle reaches a certain scale, it is necessary to build a data center to support the storage and computing requirements. And some visionary companies have embarked on the road of building their own supercomputing centers (private clouds).

For example, in August 2022, Xiaopeng announced the construction of the largest self-driving intelligence center in China, "supporting Shake", which is specially used for data processing of self-driving model training, with 6 billion floating-point computing capacity. The supercomputer Dojo released by Tesla has the computing power of exaflop, that is, tens of billions of operations per second, which Musk called "like a beast."

The practice of self-building private cloud will help car companies to grasp the core data and the technical ability of the whole process algorithm, and create product differentiation. However, this approach is only suitable for IT car companies with relatively strong strength, such as the new power of car building, and it is difficult and time-consuming, and the key is high cost.

In terms of cost, the data generated by a smart car is about 1PB a year, and 28 36TB mobile hard drives are needed for storage alone. The average price of a 36TB mobile hard drive is 6000 yuan, and the total cost is 168000 yuan. As a result, car companies that sell 100000 smart cars a year can have a data storage cost of 16.8 billion a year. According to media reports, a new foreign-funded car company built its own supercomputing center, spending as much as 140 million US dollars on hardware alone. If you include the costs of operation and maintenance, value mining and so on, the cost will rise.

In the future, as their smart cars continue to appear on the market in large quantities, car companies that build their own private clouds will have to continue to expand their private clouds due to the increasing demand for data storage and high computing power, which requires large iterations of hardware and resources. It is bound to require additional investment.

Therefore, the public cloud with strong computing power and strong data management ability has become an important choice for many new energy car companies with the advantage of performance-to-price ratio. Even the private cloud built by Xiaopeng is based on Aliyun's intelligent computing platform.

(screenshot on Aliyun's official Weibo) the market size "the cloud on the car is a deterministic trend." Baidu Intelligent Cloud Automotive Industry Solutions Director recently told the media.

The new car-building forces represented by Weilai, Xiaopeng and ideal have great competitive advantages in intelligence, and they have a lot of demand for mining and using the interactive data in the process of vehicle operation. Compared with the construction of private clouds, such tasks as front-end data collection, data tagging and back-end data services are more willing to be done by cost-effective public cloud vendors.

At present, the car cloud is speeding up the coverage of the head car companies, and 13 of the domestic Top15 car factories are cloud customers.

According to the National Development and Reform Commission, the penetration rate of smart cars in China will reach 82% in 2025, and the number will reach 28 million; in 2030, it will reach 95%, or about 38 million. With the increasing penetration of smart cars, the demand for car clouds will increase. In the future, the big cake of car cloud will grow rapidly.

First of all, in terms of intelligent driving and vehicle-road collaboration, the demand for cloud services of car companies is growing with strong certainty.

In the era of fully autonomous driving, the amount of data generated by each car every day is as high as 4TB. Moreover, for every level of self-driving of smart cars, the demand for computing power increases at least tenfold, resulting in a greater demand for cloud services. In the era of autopilot, the role of cloud computing is also so great that some people even shout: "whoever can mine the value of data efficiently and cheaply will become the king of competition."

Secondly, the continuous digital transformation of car companies will also produce a large amount of data demand.

Within the new energy car companies, there are also a large number of cloud service opportunities. For example, automobile manufacturing is a "depression" of digital transformation, where a large amount of data stays in the enterprise's local server or computer room, without in-depth value mining on the cloud. In the future, the mining and processing of these data will have a bright future.

Third, the car cloud service even helped the car network to give birth to new business type.

In the era of intelligent Internet connection, cloud service makes the link between car companies and users no longer stop at the end of the transaction, and will continue to create value and revenue for car companies. For example, car cloud can promote remote diagnosis, road rescue, fault monitoring, condition detection, OTA cloud services, UBI, used car evaluation and other application scenarios for car companies to achieve greater benefits.

(screenshot on the White Paper on Automotive Cloud Services) so what is the specific size of the automotive cloud market in the future?

From the perspective of car enterprises, Sullivan data show that in 2021 alone, the annual budget of smaller domestic car companies, such as car cloud, is about 10 million yuan, while the budget of large group-level car enterprises is about 50 million yuan, and shows a situation of rapid growth. According to an interview by Automotive Heart with VP of the information technology department of an automotive group, the company's public cloud procurement has grown sharply year after year, with cloud storage alone increasing its 2022 budget by about 30 per cent over 2021 and cloud computing by 26 per cent.

From the overall data, according to the 21st Century Economic report, the total revenue of the national automobile cloud market in 2022 is expected to be 2.5 times that of 2020. According to the "2021 China Automotive Cloud Market tracking report", the overall market size of the automotive cloud industry in 2021 is as high as 33.52 billion yuan; by 2026, this figure will soar to 80 billion yuan, a nearly three-fold increase in three years.

(note: LaaS, PaaS and SaaS are the three forms of car cloud services)

The automotive cloud market is also an important business growth point for domestic cloud manufacturers.

Entering 2022, Chinese cloud manufacturers bid farewell to the era of high growth and enter the ceiling of development. The revenue growth of leading cloud manufacturers such as Aliyun, Huawei Cloud and Tencent Cloud has generally dropped to less than 20% from 50% last year. The rising automotive cloud market is undoubtedly a visible cake for these cloud manufacturers.

Recently, cloud service providers have refreshed their solutions for automotive cloud services.

On September 6, 2022, Baidu officially launched Group Cloud, NetLianyun and supply chain Collaborative Cloud for the automotive industry, while Aliyun and Xiaopeng Automobile jointly completed the driving Intelligence Center "supporting Shake" in August 2022. In November 2021, Tencent Cloud official announced the launch of car cloud, and announced a strategic upgrade in June 2022, proposing "car-cloud integration". In June 2021, byte Jump also formed a car cloud team. Huawei is betting on the car track and has built a stack of self-driving cloud services to form an integrated solution of manufacturing, production, supply, marketing, after-sales and application.

Baidu, Ali, Tencent, Byte Jump, Huawei and other leading cloud service companies, all joined the car cloud. Not only big factories, according to incomplete statistics, domestic and foreign cloud manufacturers have clearly launched the car cloud, there are now more than 12.

According to Sullivan, Huawei Cloud, Aliyun, Baidu Cloud and Tencent Cloud are in the first echelon, accounting for nearly 70% of the total. Specific to various areas of car cloud, in the field of self-driving, Huawei Cloud accounts for 26.9%, Aliyun 16%, Tencent Cloud 15%; in the field of car networking, Huawei Cloud 20.2%, Aliyun 18.9%, Tencent Cloud 18.3%; in the field of car-road collaboration, Huawei Cloud 24%, Baidu Cloud 22%, Aliyun 18%.

Specifically, Huawei Cloud set up a "vehicle networking laboratory" as early as 2014, which is early for intelligent network-connected electric vehicles and other application scenario storage technology, with a first-mover advantage. And Huawei has "car-building experience", so it has a great advantage. In May 2022, Huawei Cloud announced that it had won the first place in China's automotive cloud market.

Baidu Cloud has been committed to the development of intelligent cars and has advantages in self-driving and other aspects. At present, Baidu's "three clouds" are committed to building their own digital base, opening up car companies and upstream and downstream supporting enterprises, and providing intelligent cloud solutions.

Ali has accumulated a lot of mature experience in self-driving and CAE simulation, and has landed products such as "Xiaoman Donkey" and self-driving trucks. The Zhiji intelligent car jointly developed by Ali and SAIC is also Aliyun's key enabling products.

Tencent Cloud aims at car companies. Through the social ecology of Wechat + WeCom and Tencent travel service Mini Program, you can install a complete set of self-developed products into car scenarios such as smart cockpits and establish data-driven closed loops. To form the integration of cloud, car and mobile, and reduce the migration costs of car companies.

The goal of Volcano engine Auto Cloud is to catch up with Tencent in 2025. In terms of business layout, although volcano engine has launched intelligent driving cloud, intelligent cockpit cloud and intelligent marketing cloud, it focuses on marketing cloud to achieve the purpose of differentiated competition.

In short, in the face of the booming smart car market, cloud manufacturers can touch the automobile industry across the river, enter a more profound sea of digital intelligence of the real economy, and greatly improve their revenue capacity. However, it is still a problem for cloud manufacturers to gain the absolute trust of new energy car companies.

New energy car companies have high requirements for the confidentiality of their core technologies, which makes it more difficult for them to fully integrate with public cloud services. In addition, data protection is also a hurdle that cloud manufacturers need to face. The data collected by new energy vehicles are stored in the cloud, if there is a lack of effective protection, it is likely to be lost or leaked due to network attacks, thus invading users' privacy.

Full-text reference

[1] "from cloud to cloud, cloud services enable the upgrading of intelligent networking in the automotive industry", Ernst & Young, Huawei

[2] "China Automotive Cloud Market tracking report 2021", Leopard Research Institute, Frost Sullivan

[3] "car Cloud, four Kingdoms Kill", Photon Planet

[4] "car Cloud, the Hidden Battlefield of self-driving", the Heart of the car

[5] "Infinite Game" of car Cloud, Finance and Economics

[6] "Huawei, Ali, Tencent, byte and other giants have entered the car cloud. Is the spring of the car cloud coming? "talking about AI

[7] Why does Ali Yun "float still"? Snow Leopard Financial Agency

This article comes from the official account of Wechat: che Bai think Tank (ID:EV100_Plus), author: Chen Zhongshan, Editor: a Feng, typesetting: fat Tiger

Welcome to subscribe "Shulou Technology Information " to get latest news, interesting things and hot topics in the IT industry, and controls the hottest and latest Internet news, technology news and IT industry trends.