17963716. METHOD AND SYSTEM FOR MULTI-CORE LOAD SCHEDULING IN AN OPERATING SYSTEM (OS) LESS COMMUNICATION NETWORK simplified abstract (SAMSUNG ELECTRONICS CO., LTD.)

METHOD AND SYSTEM FOR MULTI-CORE LOAD SCHEDULING IN AN OPERATING SYSTEM (OS) LESS COMMUNICATION NETWORK

Organization Name

SAMSUNG ELECTRONICS CO., LTD.

Inventor(s)

Mahantesh Kothiwale of Bangalore (IN)

Aneesh Deshmukh of Bangalore (IN)

Jitender Singh Shekhawat of Bangalore (IN)

Nayan Ostwal of Bangalore (IN)

Nitin Anand of Bangalore (IN)

Srinivasa Rao Kola of Bangalore (IN)

METHOD AND SYSTEM FOR MULTI-CORE LOAD SCHEDULING IN AN OPERATING SYSTEM (OS) LESS COMMUNICATION NETWORK - A simplified explanation of the abstract

This abstract first appeared for US patent application 17963716 titled 'METHOD AND SYSTEM FOR MULTI-CORE LOAD SCHEDULING IN AN OPERATING SYSTEM (OS) LESS COMMUNICATION NETWORK

Simplified Explanation

The abstract describes a method and system for multi-core load scheduling in an operating system (OS) less communication network. The method involves initializing multiple threads to process incoming packets and synchronizing these threads for simultaneous processing. The method also includes determining the CPU load on each core and the data rate of incoming packets. Based on this information, flags are enabled or disabled, and threads are determined by a reinforcement learning (RL) agent. The method then processes the functionalities of the incoming packets based on the determined threads.

• The method initializes multiple threads to process incoming packets.
• The threads are synchronized for simultaneous processing.
• The CPU load on each core and the data rate of incoming packets are determined.
• Flags are enabled or disabled based on the determined load and data rate.
• Threads are determined by an RL agent based on the enabled flags.
• The method processes the functionalities of the incoming packets based on the determined threads.

Potential Applications

• This technology can be applied in operating systems less communication networks to efficiently schedule the processing of incoming packets across multiple cores.
• It can be used in high-performance computing systems to optimize the utilization of CPU cores and improve overall system performance.
• The method can be implemented in cloud computing environments to enhance the scalability and responsiveness of network communication.

Problems Solved

• The method addresses the challenge of load scheduling in operating systems less communication networks, where traditional methods may not be applicable.
• It solves the problem of efficiently utilizing multiple CPU cores to process incoming packets in a synchronized manner.
• The method helps in optimizing the processing of incoming packets based on the CPU load and data rate, ensuring efficient resource allocation.

Benefits

• The method improves the overall performance and efficiency of operating systems less communication networks by effectively utilizing multiple CPU cores.
• It enables simultaneous processing of incoming packets, reducing latency and improving network responsiveness.
• The use of an RL agent allows for dynamic and adaptive load scheduling based on real-time conditions, leading to optimized resource allocation.

Original Abstract Submitted

A method and system for multi-core load scheduling in an operating system (OS) less communication network is disclosed. The method comprises initializing a plurality of threads for processing corresponding functionalities of incoming packets. The method further comprises synchronizing the plurality of initialized threads with each other for simultaneous processing of the one or more incoming packets. The method further comprises determining central processing unit (CPU) load on each of the plurality of cores and an ingress data-rate of one or more incoming data packets. The method further comprises enabling or disabling at least one flag based on the determined load and the ingress data-rate and determining at least one thread based on the enabled flag by the RL agent. The method further comprises processing the corresponding functionalities associated with the one or more incoming packets based on the at least one determined thread.