Technology cloud report original.
The ensuing price war in 2023 will once again increase the competitiveness of domestic public clouds. Alibaba Cloud, Tencent Cloud, JD Cloud, and Mobile Cloud took the lead in reducing prices, and also passed the chill to more cloud service providers. There is no doubt that this is a cruel test for cloud service providers.
In the stage where the public cloud is highly involved, exchanging prices for markets is undoubtedly an effective means of competition, but how long can this effect last? In other words, can the competitive dimension of the public cloud market be leveled in the same price war? Is there still untapped fertile land in the public cloud market?
For these problems, Qingyun Public Cloud seems to have a new way of solving problems.
Long-distance running based on leading technology
Qingyun Technology, born in 2012, is one of the first cloud service providers in China to launch public cloud services. At that time, Silicon Valley gave birth to many unicorn companies in the field of cloud computing, while domestic enterprises were still accustomed to the traditional IT chimney structure. Cloud computing, as a brand-new technological change, brought infinite prospects to the start-up Qingyun.
In that year, the domestic Internet business and mobile Internet business grew explosively, and the flexible, elastic, and cost-effective public cloud entered the field of vision of Internet companies.
Data shows that in 2013, it generally took 15 minutes to create a virtual machine in the industry, but Qingyun public cloud only took 6 seconds and was billed by the second. With leading technology, Qingyun public cloud quickly seized the Internet market.
Entering 2014, Qingyun has insight into some new changes in the market: traditional enterprises have begun to pay attention to cloud migration, and top customers in banking, retail, transportation and other fields have entered the vision of cloud service providers.
Qingyun made a major change at the right time, combined with the public cloud service model, launched a private cloud delivery method, and took the lead in proposing the technical proposition of "unified public and private" architecture hybrid cloud in China, thus further widening its leading edge over its opponents.
Because of the hybrid cloud delivery model of the "public-private integration" architecture, Qingyun's public cloud and private cloud can provide enterprises with a consistent experience, and can also integrate the elasticity, flexibility, security and reliability of the cloud. Between the "convenience" and "security" of the cloud, it is difficult to make a trade-off, so as to achieve "having both the fish and the bear's paw".
Accompanied by this important period, Qingyun has also continued to iteratively upgrade and innovate from the single public cloud service in the early stage, and has developed into a comprehensive and rich cloud technology stack in popular technical fields such as overlay storage, hyper-convergence, SDN, and SD-WAN. Private clouds complement each other, paving the way to cloudification for medium and large enterprise customers with flexible product and service architectures and scenario-based solutions.
At that time, many domestic government, enterprise, financial and other industry organizations chose to put some non-core business or ecological related business on the Qingyun public cloud to make full use of the elastic resources of the public cloud to respond to business demands faster. At the same time, based on industry norms, Compliance requirements, put the core business in Qingyun private cloud.
However, some traditional enterprises with a deeper understanding of cloud computing digital technology are willing to use Qingyun public cloud as the digital base of the entire business, and gradually classify related services by function to the cloud.
For example, Jiangsu Traffic Control, as the vanguard of the digital transformation of the transportation industry, has built a new transportation infrastructure based on Qingyun public cloud, realizing unified command and dispatch of high-speed traffic; a large bank provides inclusive financial services based on Qingyun public cloud, which meets compliance requirements , It also responds to the national financial innovation strategy and gives full play to the supporting role of finance in economic development.
In 2017, as enterprise multi-cloud and hybrid cloud deployments became increasingly normalized, the emergence of containers made it easier for enterprises to achieve agile development and continuous delivery in hybrid cloud heterogeneous infrastructure environments, and cloud-native technologies represented by containers officially launched stage of history.
This time, Qingyun is still at the forefront of technological innovation. Since its launch in 2018, its container platform KubeSphere has become one of the three major open source container platforms in the world. It is the only open source container platform with world influence initiated by a Chinese company. The container platform, whose open source community users cover more than 100 countries and regions around the world, is used by tens of thousands of enterprise users in the production environment.
This not only lays a solid technical foundation for Qingyun to lead the cloud native, but also lays a killer weapon for its hybrid cloud strategy to further lead the market.
With the support of cloud-native technology, Qingyun public cloud, as an important supplement to the hybrid cloud strategy, further provides enterprise customers with a wider range of unified cloud services, helping enterprises to build their own business applications more conveniently and efficiently, and focus on business innovation .
It is not difficult to find that in the past ten years, Qingyun has experienced the "golden decade" development wave of cloud computing from public cloud, private cloud, hybrid cloud to cloud native, from technology iteration to the evolution of customer needs, Qingyun has always been Stepping on the beat of the times, relying on leading technology and strategic layout, it has repeatedly subverted the old technology structure, business path and business model of China's local cloud computing industry.
During this process, Qingyun Public Cloud has also maintained a long-distance running attitude. From serving Internet users at the earliest, to serving medium and large enterprise customers based on the "public-private integration" hybrid cloud architecture, and now providing convenient cloud-native practices, it has always served enterprises. Provide a path to obtain continuous value, and meet the needs of different enterprises for in-depth cloud use and cloud management in a neutral and flexible manner.
New World of Nuggets Computing Power
It is undeniable that China's cloud computing market has entered a mature stage after more than ten years of development. With the leveling of cloud computing IaaS layer technology, more and more competitors have entered this field. In order to improve the reuse rate of public cloud IT infrastructure investment, more cloud service providers have entered the stage of price competition, trying to squeeze out competitors.
In the face of intensified market competition, Qingyun Public Cloud was inevitably coerced, but it did not choose to participate in the price war.
In Qingyun's view, price is not the only indicator for customers to go to the cloud and use the cloud. Enterprise customers pay more attention to comprehensive factors such as technology, service, operation and maintenance, and response support. Qingyun Public Cloud always puts the interests of customers first, continues to meet various needs of customers, and creates more value.
"In fact, customers still hope to get good services, and hope that public cloud vendors can provide them with the service support they need.
For example, traditional companies generally do not have a comprehensive understanding of public cloud than Internet companies. At this time, they need more early-stage business communication, architecture consulting and service support. They will also compare the overall TCO and cost performance, so we will give it to them. A complete solution for customers”, said Shen Ou, vice president of Qingyun Technology.
Because of this, Qingyun Public Cloud has set its sights on more traditional industries, providing services for industry customers who value service quality more. In addition to the financial industry that has been focusing on investment, education, medical care, energy, manufacturing and other industries are becoming new opportunities and markets for Qingyun Public Cloud.
Not only that, with the maturity of the cloud computing market and service segmentation, Qingyun also has insight into a new change: with the rapid development of cloud native technology, PaaS is decoupling from IaaS, which will gradually differentiate into application service providers and There are two groups of computing resource service providers.
Among them, cloud native technology allows enterprises to build applications across different IaaS resources, regardless of the underlying infrastructure. The cloud-native requirements are both industry-oriented and business-focused. Therefore, in the near future, upper-layer application services will become a huge market.
In the face of the imminent explosion of the application service market, Qingyun has already held the entry ticket of cloud native.
Promoting the development of public cloud through cloud native will be a key step: on the one hand, Qingyun Public Cloud can provide enterprises with efficient and convenient cloud native services through its own accumulated cloud native technology and products; on the other hand, through more open ecological cooperation , Qingyun Public Cloud will connect to more excellent third-party manufacturers to meet the diverse and personalized business needs of customers, and ultimately bring more innovative and cutting-edge value.
At the IaaS level, computing power resources are becoming a core component of the country's new digital infrastructure. The current "Digital from the East and Computation from the West" project that the country is fully promoting is a new engine that uses computing power to promote the development of the digital economy.
Statistics from the Ministry of Industry and Information Technology show that by the end of last year, the total scale of my country's computing power reached 1.8 trillion floating-point operations per second, the total scale of storage capacity exceeded 1000 EB (1 trillion GB), and the scale of the core computing power industry reached 1.8 trillion yuan. According to the calculation of the China Academy of Information and Communications Technology, every 1 yuan invested in computing power will drive 3 to 4 yuan of GDP economic growth.
The rapid development of the computing power service market has also allowed Qingyun Public Cloud to find a way to break through.
Shen Ou once revealed in an interview that when the country proposed "new infrastructure" in 2020, Qingyun had already seen the digital economy's demand for computing power centers and multiple computing power through research on relevant policies and industry needs. The layout two years in advance has enabled Qingyun to have the ability to serve the three elements of computing power center construction (architecture/energy management, multi-computing power operating system, and multi-computing power cluster operation). Related products and technologies have been used in many government-related projects be put into practice.
At present, Qingyun has built a comprehensive computing power infrastructure, created and continuously upgraded a series of high-performance, flexible and agile, safe and reliable computing power products and services, including enterprise-level e-series cloud servers, high-performance computing EHPC, QKE Container engine, distributed storage, and enterprise cloud platform, super intelligent computing platform, etc., to meet the computing power needs of enterprises for general computing, high-performance computing, scientific computing, etc., to help enterprises achieve unified and efficient massive data storage and management, and comprehensive support Enterprise business digital innovation. And this is also one of the manifestations of Qingyun's integration and innovation of public cloud and private cloud.
Based on a complete computing power infrastructure, Qingyun can provide vigorous computing power for enterprises in the fields of industrial simulation, atmospheric environment, medical medicine, education and scientific research, transportation and logistics, autonomous driving, smart government affairs, etc., and drive business innovation.
Taking Qingyun Super Intelligent Computing Platform as an example, the core code of the entire platform is independently developed by Qingyun and insists on iteration. It can not only provide massive expansion of multiple computing power, but also support the construction of cross-regional data center networks and distributed clouds, and provide computing power Unified scheduling and unified operation services. And this ability comes from the refinement of the computing power of Qingyun public cloud.
Today, the Qingyun Super Intelligent Computing Center platform has taken the lead in landing in a national supercomputing center. Among them, Qingyun undertook the construction of its supercomputing public service cloud, government affairs administrative service cloud and public cloud, providing HPC high-performance computing, EHPC elastic high-performance computing, standard cloud services (including IaaS/PaaS/GPU/AI) and distributed cloud-based Architecture of regional cloud services.
It is precisely based on the earlier computing power layout and large-scale production practice experience that Qingyun has reliable support in terms of computing power scheduling, management, and operation, and can quickly meet the diverse computing power needs of customers. and respond. As a result, Qingyun public cloud has also obtained a new development opportunity.
Breakthrough of AI Computing Power
At the beginning of 2023, the rapid development of large-scale AI models represented by ChatGPT has further promoted the surge in demand for computing power. From the earliest models such as ResNet and inception to today's GPT model, the amount of parameters of large models has continued to grow, and the amount of data has accelerated exponentially, and the required AI computing power has also increased exponentially.
However, the explosive growth of AI computing power has also brought new troubles to users in various industries:
First, large-scale AI model training and reasoning require large-scale, high-performance computing resources and huge computing requirements, resulting in extremely high computing costs;
Second, the application of AI technology in various industries continues to expand, and real-time response and decision-making have also become the core needs of enterprises. For example, when encountering response bottlenecks in the process of AI training and reasoning, it greatly affects the operating speed;
Third, traditional physical servers and data center architectures are difficult to support large-scale AI training, and enterprises are under increasing pressure on operation and maintenance;
Fourth, the extremely high technical threshold has caused great resistance to the use of AI technology by many companies.
Facing the wave of AI, Qingyun Public Cloud seized the opportunity to further break through and expand its territory from the huge demand for AI computing power.
Shen Ou said that he hopes to enable customers to obtain stable and high-quality underlying infrastructure efficiently and conveniently at a very low cost to provide support for AI large model training and reasoning, including large-scale high-performance computing resources, high-speed data storage and Management services, high-performance network services, etc.
It is worth noting that AI large model training and reasoning is itself a complex technical system. In order to help customers shield complex technologies, the cooperation between cloud vendors and large model vendors will be one of the development trends.
In this regard, Qingyun Public Cloud will connect to professional large-scale model manufacturers or AI team professional service capabilities through an open ecological cooperation model, and jointly provide scenario-based solutions for customers' AI-related businesses, assisting customers in completing data collection, computing power platform construction, Model training and reasoning, customized models, public-private hybrid cloud deployment, etc., finally build or implement AI applications that meet real needs and form their own differentiated competitiveness.
At the same time, AI is also a very broad field. Its requirements for computing power are complex and diverse. Qingyun Public Cloud will also provide a complete computing power infrastructure covering general computing power, scientific computing, and high-performance computing. , to provide diversified computing power support for enterprise customers.
For example, the new generation of computing e4 cloud server to be released by Qingyun can be well applied to business scenarios with higher performance requirements for computing and network.
It is reported that the e4 cloud server adopts the latest fourth-generation Intel® Xeon® Platinum processor, which has a comprehensive performance improvement. Compared with the previous generation (e3), the CPU performance has increased by 50%, the storage IOPS performance has increased by 35%, and the physical network performance has increased by 100%. %, the network delay is reduced by 30%, and it supports acceleration engines such as AMX and QAT. The performance of deep learning and training reasoning is greatly improved, the efficiency of SSL decryption is increased by 6 times, and the compression and decompression is increased by 8-9 times. Performance computing, big data, database and other business scenarios.
In general, in terms of computing power operation, Qingyun public cloud gives full play to its computing power dispatching and management capabilities, cooperates with operators, large-scale government, enterprise and scientific research institutions, and plays a good role in computing power operation; in terms of computing power scene capabilities, Qingyun Based on public cloud capabilities, a comprehensive computing power infrastructure has been built to provide enterprises with diversified computing power support; in terms of AI computing power ecology, Qingyun Public Cloud is actively cooperating with AI large-scale model companies to drive development through positive business models , Create more value for enterprise customers.
epilogue
In the era of highly introverted public cloud, Qingyun does not blindly follow the price war, but finds another way. Through the forward-looking layout and continuous leading technical capabilities, it has blazed a new path of cloud native and computing power, thus giving the market new imagination. .
[About Science and Technology Cloud Report]
Focus on original enterprise-level content experts - technology cloud reports. Founded in 2015, it is the top 10 media in the cutting-edge enterprise IT field. Recognized by the Ministry of Industry and Information Technology, Trusted Cloud, one of the official media designated by the Global Cloud Computing Conference. In-depth original reports on cloud computing, big data, artificial intelligence, blockchain and other fields.