Difference between revisions of "Ray - UC Berkeley RISELab"
m |
m |
||
| Line 12: | Line 12: | ||
* [[Reinforcement Learning (RL)]] | * [[Reinforcement Learning (RL)]] | ||
| − | Ray is a high-performance distributed execution framework targeted at large-scale machine learning and reinforcement learning applications. It achieves scalability and fault tolerance by abstracting the control state of the system in a global control store and keeping all other components stateless. It uses a shared-memory distributed object store to efficiently handle large data through shared memory, and it uses a bottom-up hierarchical scheduling architecture to achieve low-latency and high-throughput scheduling. It uses a lightweight API based on dynamic task graphs and actors to express a wide range of applications in a flexible manner.[http://rise.cs.berkeley.edu/projects/ray/ Ray | UC Berkeley RISELab] | + | Ray is a high-performance distributed execution framework targeted at large-scale machine learning and reinforcement learning applications. It achieves scalability and fault tolerance by abstracting the control state of the system in a global control store and keeping all other components stateless. It uses a shared-[[memory]] distributed object store to efficiently handle large data through shared [[memory]], and it uses a bottom-up hierarchical scheduling architecture to achieve low-latency and high-throughput scheduling. It uses a lightweight API based on dynamic task graphs and actors to express a wide range of applications in a flexible manner.[http://rise.cs.berkeley.edu/projects/ray/ Ray | UC Berkeley RISELab] |
Latest revision as of 21:45, 2 March 2024
Youtube search... ...Google search
Ray is a high-performance distributed execution framework targeted at large-scale machine learning and reinforcement learning applications. It achieves scalability and fault tolerance by abstracting the control state of the system in a global control store and keeping all other components stateless. It uses a shared-memory distributed object store to efficiently handle large data through shared memory, and it uses a bottom-up hierarchical scheduling architecture to achieve low-latency and high-throughput scheduling. It uses a lightweight API based on dynamic task graphs and actors to express a wide range of applications in a flexible manner.Ray | UC Berkeley RISELab
