Recently, the International Data Corporation (IDC) released the "IDC China 2021H1 Artificial Intelligence Public Cloud Service Market Research Report", Huawei Cloud's one-stop AI development platform ModelArts ranks first in China's machine learning public cloud service market, with a market share of 30.6% %, a year-on-year increase of 1.6%. So far, ModelArts has topped the market for three consecutive times.
Huawei Cloud's one-stop AI development platform, ModelArts,
ranks No. 1 in China's machine learning public cloud service market.
Data source: "IDC China 2021H1 Artificial Intelligence Public Cloud Service Market Research Report"
reports that in China's machine learning public cloud service market, Huawei Cloud Favored by government and enterprise customers, its competitive advantage in the AI cloud service market has gradually become prominent.
Huawei Cloud's one-stop AI development platform ModelArts can provide data processing, algorithm development, model training, model management, model deployment and other AI model development process technical capabilities; take the lead in supporting MLOps; continue to build large-scale model training and optimization capabilities; AI ecology, access to ecological development tools, connect the supply, demand, and learning parties of the AI market through HUAWEI CLOUD AI Gallery, and help solve the "scarcity of AI computing power, shortage of AI talents, difficulty in AI development, Difficult to apply” dilemma and lack, promote AI to enter the core production system of thousands of industries.
For AI developers, it provides hard-core black technology service
ModelArts artificial intelligence cluster service to support large-scale distributed training. To help developers and enterprise customers solve the challenges of technology, cost, and resources when training large models, HUAWEI CLOUD's one-stop AI development platform ModelArts provides artificial intelligence cluster services, in-depth optimization of all scenarios, and supports large-scale distributed training. From the perspective of the end-to-end whole process, compared with offline development of large models, the development efficiency based on ModelArts can be increased by more than 4 times, and it can solve the problem of resource availability for super-large model training.
The cloud-native model development tool chain helps improve the efficiency of AI development. Huawei Cloud's one-stop AI development platform ModelArts supports the combination of cloud resources and development tool chains. For AI beginners, the online CodeLab supports second-level access to resources and can be switched on demand; for in-depth developers, it supports remote development on the cloud and custom development environments. With the help of cloud-native resource scheduling, developers can further build algorithm development paradigms and capabilities based on ModelArts cloud-native.
Facing the AI ecology, build an AI development community of "knowledge" + "training"
Based on the one-stop AI development platform ModelArts, HUAWEI CLOUD has built the developer ecological community AI Gallery, which is a social platform for AI asset sharing and an AI development community of "knowledge" + "practice training". AI Gallery brings together more than 10 kinds of algorithms, models, datasets, workflows, and more than 50,000 AI assets to ensure that all participants in the AI development and application ecological chain can efficiently realize their own business value and reduce the cost of all industries. The learning threshold for developers in the field of artificial intelligence accelerates the application and practice of AI.
Facing the AI industry and fully supporting the development of MLOps in the whole process
Huawei Cloud's one-stop AI development platform ModelArts fully supports MLOps, changing the original process of developing AI models in stages and realizing a collaborative iterative development model of full-process automation. ModelArts has also become the first AI development platform in China that supports MLOps.
Based on the concept of MLOps, the Workflow tool of ModelArts, a one-stop AI development platform of Huawei Cloud, provides functions such as operation record, monitoring, and continuous operation. Through the Workflow tool, the full life cycle management of AI development, operation and post-operation monitoring can be realized, the iteration speed from AI development to implementation can be accelerated, and the effect experience can be continuously optimized. At the same time, developers can use the AI Gallery to realize the rapid construction and capability sharing of Workflow.
Huawei Cloud's one-stop AI development platform, ModelArts, helps enterprises realize the full lifecycle management of AI applications, greatly improves the efficiency of AI development, and accelerates the implementation of AI in various industries.
In the field of manufacturing, HUAWEI CLOUD and Bosch Huayu have developed Usecase, an intelligent identification tool state based on ModelArtsWorkflow. Through the tool voice recognition algorithm, the efficiency of tool usage is increased by more than 5% under the condition of ensuring the yield rate. The Tool Sound Recognition Usecase based on the MLOps concept can easily expand the solution for different tools or even other production tools, and apply the sound detection algorithm to a wider range of state maintenance scenarios in the manufacturing industry.
In the field of drug research and development, relying on ModelArts, a one-stop AI development platform of Huawei Cloud, and EIHealth, a one-stop medical research and development platform of Huawei Cloud, the Shanghai Institute of Materia Medica, Chinese Academy of Sciences, and Huawei Cloud have trained the large molecular model of Pangu drugs on Huawei Cloud, empowering the whole process. AI drug design greatly improves the efficiency of new drug research and development.
In the field of transportation, Shenzhen Airport quickly built 60+ dispatching rules based on HUAWEI CLOUD airport intelligent body to realize automatic and intelligent seat allocation, increase the airport bridge rate by 5%, and help 2.6 million passengers avoid taking shuttle buses every year.
In the future, HUAWEI CLOUD will persist in innovation, build an open and win-win HUAWEI CLOUD global ecosystem, and open up technology to customers from all walks of life through cloud services, so that cloud is everywhere, intelligence is omnipresent, and a smart world is built together. Cloud base.