首页|Patent Issued for System and methods for tag-based synchronization of tasks for machine learning operations (USPTO 11842197)
Patent Issued for System and methods for tag-based synchronization of tasks for machine learning operations (USPTO 11842197)
扫码查看
点击上方二维码区域，可以放大扫码查看
原文链接
NETL
NSTL
From the background information supplied by the inventors, news correspondents obtained the followingquote: “A hardware-based machine learning (ML) system typically includes multiple cores/subsystems(processing blocks and tiles), each having its own processing units and on-chip memory (OCM) for executingML applications/operations. A ML application typically includes multiple tasks, wherein each tasks mayinclude a plurality of programmable instructions grouped and transmitted from a compiler to an instructionstreaming engine and executed by the ML system to complete the task. Some of the tasks may be executedindependently on the processing tiles without relying on other tasks to be executed, while some other tasksmay have dependency on other tasks and can only be scheduled after the other tasks are executed by theprocessing tiles first. As such, scheduling of the tasks that have dependencies needs to be synchronized.
BusinessCyborgsEmerging TechnologiesMachine LearningMarvell Asia Pte Ltd.
NETL
NSTL
