Huawei Technologies Co., Limited

Future Wei Technologies, Inc.

Alibaba Group Holding Limited

Muhui Lin

Jiongjiong Gu

Wesley M Shao

Baifeng Yu

Zhou Yu

Lingzhi Mao

Junjie Wang

Xiaolin Jia

Xin Wu

Ruiling Wang

Lijiang Li

Shunan Li

Zhixiong Guo

Wesley Shao

Embodiments of this application relate to the data access field, including data access methods systems that enable efficient and effective data access. In one embodiment, a method includes: assigning, by an RDMA control service based on user information and a corresponding connection relationship between a switch and a first instance defined by a user, an address segment to the first instance; building, by the RDMA control service, an access control list based on the address segment assigned to the first instance, where the access control list is used for controlling access between different first instances defined by the user; and sending, by the RDMA control service, the access control list to a switch control service, such that the switch control service configures the access control list for the switch. In one embodiment, access between different instances defined by a same user can be effectively controlled, thereby effectively resolving an issue of access isolation for different users accessing an RDMA network node.

Data access methods and systems

A resource management technology is disclosed that may be applied to a host, where the host includes a Central Processing Unit (CPU), an endpoint connected to the CPU, and an Input/Output (I/O) device connected to the endpoint. A method includes: allocating, by the CPU, a target endpoint to a target process, where a virtual device is disposed on the target endpoint; obtaining, by the target endpoint, a performance specification of the target process, and adjusting a performance parameter of the virtual device according to the performance specification, where the adjusted virtual device satisfies requirements of performance specifications of all processes that use the target endpoint; and when the target process needs to access a resource, obtaining, from the I/O device, a resource that satisfies the performance specification of the target process, and providing the obtained resource to the target process for use.

Resource management method, host, and endpoint based on performance specification

Embodiments provide a resource management technology that may be applied to a host, where the host includes a CPU, an endpoint connected to the CPU, and an I/O device connected to the endpoint. A method includes: allocating, by the CPU, a target endpoint to a target process, where a virtual device is disposed on the target endpoint; obtaining, by the target endpoint, a performance specification of the target process, and adjusting a performance parameter of the virtual device according to the performance specification, where the adjusted virtual device satisfies a total requirement of performance specifications of all processes that use the target endpoint; and when the target process needs to access a resource, obtaining, from the I/O device, a resource that satisfies the performance specification of the target process, and providing the obtained resource to the target process for use.

System for resource management using performance specification and description information of the process

A method for extending a Peripheral Component Interconnect Express (PCIe) domain. A configuration space address can be allocated to a PCIe device in an extended domain from a memory address of a root complex endpoint device, a correspondence between the configuration space address and a bus number/device number/function number (BDF) can be established, and a bus number can be allocated from a second bus set of the extended domain to a PCIe device discovered in the extended domain, where the bus number is used for determining the BDF of the PCIe device discovered in the extended domain, so as to access, according to the correspondence between the configuration space address and the BDF and by using the BDF of the PCIe device discovered in the extended domain, a configuration space register of the PCIe device discovered in the extended domain.

Method and apparatus for extending PCIe domain

A fault isolation method, computer system, and apparatus, which are capable of monitoring a state of a second endpoint device in the extended domain, and setting a device state record according to the state of the second endpoint device; after an access request between the second endpoint device and the primary domain is received, querying the device state record according to identifier information that is of the second endpoint device and in the access request, and determining the state of the second endpoint device; and if the state of the second endpoint device is a fault state, discarding the access request to prevent communication between the faulty second endpoint device and the primary domain and prevent spreading a fault to the primary domain, thereby ensuring system reliability.

Fault isolation method, computer system, and apparatus

Per-Function Downstream Port Containment (pF-DPC) is an extension to Downstream Port Containment (DPC) in the Peripheral Component Interconnect express (PCIe) standard. Pf-DPC confines non-fatal errors to specific functions of an end-point device without disabling the link between a PCIe port and the end-point device. PCIe ports configured for pF-DPC may filter (e.g., drop) packets carrying routing identifiers (RIDs) and/or addresses assigned to a function affected by a non-fatal error, while continuing to forward packets carrying RIDs/addresses associated with remaining operable functions over the corresponding link.

Growing community of inventors

Hangzhou, China

Muhui Lin