Huawei has announced two new chips for artificial intelligence applications.
The Ascend AI IP and chip series, the world’s first AI IP and chip series that natively serves all scenarios, providing optimal TeraOPS per watt. The Ascend series delivers excellent performance per watt in every scenario, whether it’s minimum energy consumption or maximum computing power in data centers. Their unified architecture also makes it easy to deploy, migrate, and interconnect AI applications across different scenarios.
The Ascend 910 and Ascend 310 chips, which were announced at today’s event, mark Huawei’s leading AI capabilities at the chip level – the bottom layer of the stack. These chips will help greatly accelerate AI adoption in all industries.
Apart from the Ascend series of chips, Huawei’s full-stack AI portfolio also includes the following:
CANN (Compute Architecture for Neural Networks): A chip operators library and highly automated operators development toolkit
MindSpore: A unified training and inference framework for device, edge, and cloud (both standalone and cooperative)
Application enablement: Full-pipeline services (ModelArts), hierarchical APIs, and pre-integrated solutions
Huawei’s AI portfolio includes its new Ascend series of AI chips – the world’s first AI IP and chip series designed for a full range of scenarios (see below for more information). The portfolio also includes new products and cloud services that are built on Ascend chip capabilities. With its full-stack AI portfolio, Huawei aims to provide pervasive intelligence to help drive industry development and build a fully connected, intelligent world.
Ten future changes: Driving Huawei’s AI strategy
Huawei predicts that by 2025, the world will see upwards of 40 billion personal smart devices, and 90% of device users will have a smart digital assistant. Data utilization will reach 86% and AI services will be readily available, as prevalent as the air we breathe. According to Huawei, AI has become a new general purpose technology and will change all industries and organizations on earth.
Huawei has defined ten changes that will help pave the way. They include:
1. Faster model training
2. Abundant and affordable computing power
3. AI deployment and user privacy
4. New algorithms
5. AI automation
6. Practical application
7. Real-time, closed-loop system
8. Multi-tech synergy
9. Platform support
10. Talent availability
Huawei’s AI strategy has five areas of focus:
1. Invest in AI research: Develop fundamental capabilities for machine learning in domains like computer vision, natural language processing, and decision/inference, etc. Huawei places special emphasis on machine learning that is:
data and power-efficient (i.e., less data, computing, and power needed)
secure and trusted
2. Build a full-stack AI portfolio:
Deliver abundant and affordable computing power
Provide an efficient and easy-to-use AI platform with full-pipeline services
Make the portfolio adaptive to all scenarios, both standalone and cooperative scenarios between cloud, edge, and device
3. Develop an open ecosystem and talent: Collaborate widely with global academia, industries, and partners.
4. Strengthen existing portfolio: Introduce an AI mindset and techniques into existing products and solutions to create greater value and enhance competitive strengths.
5. Drive operational efficiency at Huawei: Apply AI to massive volumes of routine business activities for better efficiency and quality.
Xu envisions that Huawei’s full-stack AI portfolio will speed up AI adoption in all industries to provide pervasive intelligence for a fully connected, intelligent world.