2024.10.08
On September 24th, the "2024 DataCanvas Computing Power Pack Product Launch Event" was successfully held in Beijing.At the conference, Frost & Sullivan officially released the “China Intelligent Computing Center Operating System Market Research Report” (hereinafter referred to as the “Report”), which deeply analyzes the technological innovations, market insights, and leading practices of intelligent computing centers and their operating systems in the new era of computing intensity. This report systematically fills the research gap in this field for the first time.
DataCanvas was selected as a leading representative vendor of intelligent computing center operating systems with its comprehensive advantages in four dimensions: product strength, operational strength, collaborative strength, and ecological strength.
As China’s first market research report on intelligent computing center operating systems, the "Report" is dedicated to deeply exploring the entirety of intelligent computing centers and their operating systems. It clarifies the basic concepts, key technologies, market positioning, and core values of intelligent computing center operating systems, highlighting their critical role in resource-efficient management, intelligent scheduling, and other key support functions in the intelligent age.
Additionally, the "Report" further evaluates the comprehensive competitiveness of typical vendors in the intelligent computing center operating system industry chain, providing insights into industry leaders and offering direction and advice for the future development of the industry.
New Era of Computation Begins, Intelligent Centers Become Key Focus
Faced with challenges of hardware integration and software compatibility in large-scale model training, as well as the high costs and long training cycles, we need to enhance the efficiency of distributed training, optimize the scheduling of computing resources, and develop clusters of over 10,000 cards and build efficient computing platforms.
——"China Intelligent Computing Center Operating System Market Research Report"
The explosive development of large model technology has triggered a revolution in computing paradigms such as parallel computing, intelligent edge computing, and microservices, resulting in a surge in demand for computing power and ushering in a new era of computing intensity. In this context, intelligent computing centers, as core infrastructures supporting future computing needs, have become a focal point of global strategic competition.
Frost & Sullivan defines an intelligent computing center as "a new type of computing power infrastructure based on advanced artificial intelligence theories and utilizing cutting-edge AI computing architectures.
It provides necessary computing services, data services, and algorithm services for AI applications, effectively supporting the open sharing of data, the construction of intelligent ecosystems, and the gathering of industrial innovation through the production, aggregation, scheduling, and optimized allocation of computing power. This significantly promotes the industrialization of AI, the intelligentization of industries, and the intelligent governance of governments."
Production, Scheduling, and Release of Computing Power in Intelligent Computing Centers
Source: Frost & Sullivan
The "Report" points out that under the guidance of policies and the coordinated planning of local governments, domestic intelligent computing center construction is expected to enter a new wave of construction. Compared with previous data center constructions, this round of new infrastructure will place greater emphasis on the integration of construction and operation, compliance with energy consumption regulations, effective cost control, and security and reliability.
Internationally, countries such as the United States, the European Union, and Singapore have elevated the development of intelligent computing centers to a strategic level, showing unprecedented rapid development in the construction of global intelligent computing centers.
It should also be noted that while the construction of intelligent computing centers is in full swing, it currently faces multiple challenges such as slow standardization processes, unrealized abstraction functions of computing resources, and a lack of powerful tools and platforms to support large model building.
Complex Challenges for Intelligent Centers, Core Value of Operating Systems
The intelligent computing center operating system is specifically designed to meet the service needs of intelligent computing centers, efficiently managing and intelligently scheduling the infrastructure resource pool of intelligent computing centers.
By keeping up with the development and upgrading trends of GPU hardware, optimizing hardware resource utilization and management, it enhances the capability and efficiency of computing power scheduling, and further achieves intelligent management and operations, providing support for the long-term sustainable development of intelligent computing centers.
——"China Intelligent Computing Center Operating System Market Research Report"
As the central nervous system of intelligent computing centers, the intelligent computing center operating system relies on six core values—heterogeneous computing power management, intelligent computing power scheduling, flexible expansion and cost-efficiency, AI optimization acceleration, computing power quota management, and intelligent model and data management. These core values efficiently address the complex challenges of implementing intelligent computing centers, laying a solid foundation for the stable operation and efficient business functionality of intelligent computing centers, and truly realizing "software-defined computing power."
Frost & Sullivan further points out that relying on the precise segmentation and efficient management capabilities of the intelligent operating system for GPU computing power resource pools, the computing power leasing business model is undergoing unprecedented transformation opportunities.
DataCanvas's innovative computing power pack product, based on the intelligent operating system, provides users with unparalleled flexibility. For the first time, it proposes using "degrees" as a unit of measurement, allowing users to flexibly configure and adjust computing resources according to actual needs. This "on-demand purchase, use immediately upon purchase" model effectively "fills the gap in traditional business models in the computing power market, bringing innovative solutions to the industry."
The "Report" believes that the intelligent computing center operating system market is currently in a highly prosperous state. Leveraging deep insights from multiple perspectives such as technology, market, resources, and ecology, Frost & Sullivan uses four dimensions—product strength, operational strength, collaborative strength, and ecological strength—to establish a comprehensive service value analysis model, further detailed into 11 primary indicators and 19 secondary indicators.
This aims to systematically analyze the comprehensive value performance of domestic intelligent computing center operating system vendors, empowering the computing power market to develop more autonomously and maturely.
Source: Frost & Sullivan
As a provider of AI infrastructure, DataCanvas, with its core advantages of low-threshold and high-efficiency AI operations, significantly improved computing power efficiency and utilization, meeting multi-level computing power needs, integrated AI services, and native suitability for high-performance intelligent computing, has been selected as a leading representative vendor of intelligent computing center operating systems.
It demonstrates the best performance in multiple key primary indicators such as technology and performance, economic efficiency, underlying resource scheduling capabilities, software and hardware coordination capabilities, and ecological openness!
Source: Frost & Sullivan
Looking forward, DataCanvas will continue to leverage its leading technological advantages and cutting-edge practical experience in the field of intelligent computing software to provide governments and enterprises with more flexible, efficient, and high-level integrated "algorithm + computing power" AI services.
This will help governments and enterprises navigate the waves of digital transformation with success.