Wednesday, August 24
12:25 PM - 12:40 PM
At Ant, we have widely adopted Ray for various types of systems. The increasing workloads of these systems brought various scalability challenges. Over the last two years, we have continuously stress-tested the scalability of Ray and made improvements together with the open-source community. We are now able to support clusters with more than 3,500 nodes and 50,000 actors. In this talk, we will discuss the improvements we've made to achieve this scale.
Tao Wang works at Ant Group, focusing on optimizations on Ray.
Come connect with the global community of thinkers and disruptors who are building and deploying the next generation of AI and ML applications.Save your spot