Patents
Literature
Patsnap Eureka AI that helps you search prior art, draft patents, and assess FTO risks, powered by patent and scientific literature data.

6results about How to "Reduce context switching" patented technology

Method for optimizing user mode file system framework FUSE

The invention relates to a method for optimizing a user mode file system framework (FUSE), which comprises the following steps of: deploying an rcFUSE driving program in a kernel, and deploying an rcFUSE daemon process in a user space; a rcFUSE drive program is loaded, annular channels are created, the rcFUSE drive program receives the request, and each annular channel is distributed to a memory on an NUMA node which is the same as a CPU core ID; mounting a file system based on rcFUSE by a user, and mapping a memory area of the annular channel to a virtual address space of the user; the synchronization requests are directly queued to a pending annular buffer area, the asynchronous requests are initially added to a background queue, and then the asynchronous requests are moved to the pending annular buffer area; and a polling mechanism is adopted in the rcFUSE driving program. The parallelism of request processing is maximized, so that high expandability is realized; context switching and request copy overhead are effectively reduced.
Owner:SICHUAN HUACUNZHIGU TECH CO LTD

Lock strategy adjustment method based on lock scheduling condition and terminal

The invention relates to the technical field of lock scheduling prediction, in particular to a lock scheduling condition-based lock strategy adjustment method and a terminal, and the method comprises the steps of obtaining historically collected system operation data which comprises lock scheduling data, system load data and system scheduling overhead data; training a machine learning model according to the historically collected system operation data; predicting the contention degree of the lock in real time according to the trained machine learning model to obtain a prediction result; adjusting a scheduling strategy of the lock according to the prediction result; according to the technical scheme, the contention degree of the lock can be intelligently predicted, then the lock strategy is dynamically adjusted, the balance between high performance and fairness is achieved, the optimization of the lock scheduling strategy is achieved, unnecessary spinning and context switching can be reduced, the CPU use efficiency is improved, and therefore the overall stability and response capacity of the system are improved.
Owner:CONTEMPORARY NEBULA TECH ENERGY CO LTD

Chain descriptor management method and device for VirtIO equipment of DPU

The invention provides a chained descriptor management method and device for VirtIO equipment of a DPU. A host machine receives a virtual input / output operation request sent by a virtual function; configuring a descriptor in a memory space pointed by a descriptor annular queue buffer area, wherein the descriptor comprises a previous in-chain descriptor address, a next in-chain descriptor address, a descriptor cut-off bit and other basic information; configuring a descriptor head address in a memory space pointed by the available data head annular queue buffer area; and sending a notification signal to the VirtIO equipment through the base address register, receiving a control block which is returned to the completed annular queue buffer area after the VirtIO equipment completes the corresponding virtual input / output operation, and recovering the released descriptors according to information contained in the control block. According to the method provided by the invention, a chain descriptor structure is introduced, batch data transmission and concurrent processing are realized, and the overall performance of VirtIO equipment is improved.
Owner:YUSUR TECH CO LTD

Lock scheduling condition prediction method and terminal

The invention relates to the technical field of lock scheduling prediction, in particular to a lock scheduling condition prediction method and a terminal, and the method comprises the steps: obtaining historically collected system operation data which comprises lock scheduling data, system load data and system scheduling overhead data; training a machine learning model according to the historically collected system operation data; predicting the contention degree of the lock in real time according to the trained machine learning model to obtain a prediction result; and adjusting the scheduling strategy of the lock according to the prediction result. According to the technical scheme, the contention degree of the lock can be intelligently predicted, then the lock strategy is dynamically adjusted, the balance between high performance and fairness is achieved, the optimization of the lock scheduling strategy is achieved, unnecessary spinning and context switching can be reduced, the CPU use efficiency is improved, and therefore the overall stability and response capacity of the system are improved.
Owner:CONTEMPORARY NEBULA TECH ENERGY CO LTD

Message communication processing method and device and electronic equipment

The invention relates to a message communication processing method and device and electronic equipment, and the method comprises the steps: obtaining a first physical address of first original message content under the condition that the first original message content corresponding to any to-be-sent message is received; in the process of accessing the first preset memory based on the first user mode driver, determining a first storage space from a preset sending queue of the first preset memory based on the first preset configuration information, and storing the first physical address in the first storage space; in the process of accessing the first preset memory based on the first network card hardware, under the condition that it is detected that any idle storage space in a preset sending queue is occupied, based on second preset configuration information, taking out the first storage space from the preset sending queue, and analyzing the first storage space to obtain a first physical address; and sending the to-be-sent message based on the first physical address and the first network card hardware. According to the embodiment of the invention, the problem of time delay uncertainty in a message transmission process can be solved.
Owner:CHINA AUTOMOTIVE INNOVATION CORP

A DPDK-based virtual rdma device and method of using the same

The application discloses a virtual RDMA device based on DPDK and a use method thereof. The virtual RDMA device based on DPDK comprises a front-end virtio-rdma device, which is used for receiving an RDMA device operation request initiated by an application program in a virtual machine, encapsulating the request as an instruction conforming to a virtio protocol, and sending the instruction to a back-end vhost-user- rdma device through a vring; and the back-end vhost-user- rdma device is used for simulating the function of a real RDMA network card, the device can acquire and analyze the instruction conforming to the virtio protocol from a circular queue vring, execute corresponding RDMA device operations, and return the operation results to the front-end virtio-rdma device through the vring. The application can get rid of the dependence on the real RDMA device, and can improve the resource utilization, system flexibility and safety.
Owner:KYLIN CORP