Qnap Systems, Inc.

Chin-Hsing Hsu

A container system is disclosed. The container system includes a host, a daemon and an API server running in the containers of the host, and a container manager running in the host. The host has a file system. The container manager creates a source directory in the file system and mounts the source directory to a share directory of the daemon. The API server receives a request command with respect to a target directory in the file system, and the container manager bind-mounts the target directory to a sub-directory of the source directory according to the request command. In addition, the container manager responds the API server the location of the target directory in the share directory for accessing the required data through the share directory in the containers.

Container system, method of dynamically mounting host data to container, and application program for the same

A containerized application management system and management method, applied to a network, are provided. The system includes a master node, a first worker node with a first change tracking program pod deployed thereon, and a second worker node with a second change tracking program pod deployed thereon. The method includes: receiving a pod create request, and sending a content mutate request by the master node when content of the pod create request meets a preset condition; receiving the content mutate request and changing the content to form a mutated pod create request, and forwarding the mutated pod create request by the master node to the first or second worker node; and enabling the first or second worker node to create a pod according to the mutated pod create request. The pod includes an initial container and at least one container.

Containerized application management system and management method

A data caching method, a node, a machine learning framework and a file system client based on hyper-converged infrastructure are provided. In the data caching method, a computing program requests a computing node to prefetch computing data prior to using the computing data. Then, the computing node acquires and stores the computing data in a cache memory as requested by the computing program. Subsequently, as the computing program requests the computing node to discard the computing data, the computing node discards the computing data from the cache memory as requested by the computing program.

Data caching method and node based on hyper-converged infrastructure

A distributed storage system and a data synchronization method are used with a network. The system includes a first network host and a second network host. A first file system directly writes data generated by the first network host into a first software-simulated persistent memory. The data in the first software-simulated persistent memory is stored into a first remote block device and cached, respectively. The cached data is stored into a first persistent storage by asynchronous writeback mechanisms. The first remote block device transmits the received data to the second software-simulated persistent memory through the network. The data transmitted to the second software-simulated persistent memory is cached, and the cached data is stored into a second persistent storage by asynchronous writeback mechanisms. The second network host replaces the first network host to provide services when the first network host is out of service.

Distributed storage system and data synchronization method applied thereto

A multi-controller storage system and a storage apparatus for performing data access to a non-volatile memory with multiple controllers are provided. When a host transmits a command to a first controller and the first controller assigns the command to a second controller, the first controller transfers data from the host to the second controller through the first controller through the non-volatile memory express over remote direct memory access protocol (NVMe over RDMA) to directly write the data into the second controller.

Multi-controller storage system and storage apparatus

A network application program product and a method for processing an application layer protocol are provided. The network application program product includes a network application program executed in a user mode and a packet filtering module executed in a kernel mode. The network application program transmits connection information, associated with network connection to the network application program, to the packet filtering module. The packet filtering module captures a to-be-processed packet on the network connection according to the connection information; analyzes the to-be-processed packet into at least one protocol data unit; compares the protocol data unit with preset code content; and transmits the protocol data unit to the network application program or the data access module according to the comparison result for later processing.

Network application program product and method for processing application layer protocol

A method for managing backup data units stored in a data server is provided. The backup data may include a first backup data unit established by copying a source data unit into the data server in response to a data backup command received at a first time point. The method includes: acquiring a specified storage space in response to a data backup command received at a second time point; copying the first backup data unit to the specified storage space in a lightweight copy manner; writing differential data contents between the source data unit presented at the second time point and the source data unit presented at the first time point into the lightweight-copied first backup data unit in the specified storage space; and defining the resulting data stored in the specified storage space as a second backup data unit, wherein the second time point is later than the first time point.

Method and system for managing backup data

A data writing method of a flash disk array is provided. The data writing method includes steps of: obtaining a first data quantity; receiving intermediate data to be written into the flash disk array; comparing a second data quantity of the intermediate data with the first data quantity; and writing the intermediate data into the flash disk array when the second data quantity is not less than the first data quantity. At most one data block is allocated from each of flash drives. Some of the allocated data blocks are written with the intermediate data, while others of the allocated data blocks are not written with the intermediate data and provided as reserved blocks. The number of the reserved blocks has at least two selectable values dependent on residual lives of the flash drives.

Data writing method of flask disk array based on residual lives of flash drives thereof

A method, a data storage system, and a computer-readable recording medium for disk array data distribution are proposed. The method includes the following steps. The space of a disk array composed of multiple flash storage devices is divided into multiple RAID extents with a same data distribution pattern, where each of the RAID extents includes a first region having multiple first stripes and first strips and also a second region having multiple second stripes and second strips. The first strips in each of the first stripes are evenly distributed among the flash storage devices in a first rotation pattern, and the second strips in each of the second stripes are unevenly distributed among the flash storage devices in a second rotation pattern.

Growing community of inventors

Taipei, Taiwan

Chin-Hsing Hsu