Future Wei Technologies, Inc.

Huawei Technologies Co., Limited

Vid Scale, Inc.

Guangyu Shi

Norbert Egi

Jian Ye Chen

Liufei Wen

Hao Gong

Masood Mortazavi

Haiyong Xie

Thomas Boyle

Robert Lasater

Xiangyang Xu

Guoqiang Wang

Hongbo Zhang

Jianming Wu

Yu Liu

Ravishankar Ravindran

Jian He

Huawei Liu

Xiaoke Ni

Xiancai Li

Vineet Chadha

John Peters

Xinwen Zhang

Stephen P Morgan

Lingyuan Fan

Chi Young Ku

Youshui Long

Pengwei Wang

Gopinath Palani

Kopylov Vladislav

Hao Wu

Xun Yang

Cheng He

Zhefeng Yan

Yunfei Zhang

Venu Gopala Rao Kotha

Zhi Li

Xueliang Chen

Byung Choi

Mengmeng Chen

Qiang Gu

Cong Shi

Chi Yong Ku

Ron Chung Hu

Shueng-han Gary Chan

Raju Joshi

Fu Li

Jianyun Zhu

Shuo Guo

Avraham Kapuya

A method of communicating data over a Peripheral Component Interconnect Express (PCIe) Non-Transparent Bridge (NTB) comprising transmitting a first posted write message to a remote processor via the NTB, wherein the first posted write message indicates an intent to transfer data to the remote processor, and receiving a second posted write message in response to the first posted write message, wherein the second posted write message indicates a destination address list for the data. Also disclosed is a method of communicating data over a PCIe NTB comprising transmitting a first posted write message to a remote processor via the NTB, wherein the first posted write message comprises a request to read data, and receiving a data transfer message comprising at least some of the data requested by the first posted write message.

Scalable direct inter-node communication over peripheral component interconnect-express (PCIe)

A method comprising receiving a plurality of data records, storing the data records as data segments in a storage element, receiving a plurality of descriptors for each data segment, wherein each descriptor describes an aspect of data contained in the data segments, employing a first user-defined function to resolve a first minimum descriptor for each data segment and a first maximum descriptor for each data segment, composing a lightweight index for the data segments, wherein the lightweight index comprises the first minimum descriptor for each data segment and the first maximum descriptor for each data segment, and appending the lightweight index to the data segments in the storage element.

Serialization scheme for storing data and lightweight indices on devices with append-only bands

A data processing apparatus and a data processing method are provided. The apparatus includes M protocol stacks and at least one distribution service module, and the M protocol stacks separately run on different logic cores of a processor and are configured to independently perform protocol processing on a data packet to be processed. The distribution service module receives an input data packet from a network interface and sends the data packet to one of the M protocol stacks for protocol processing, and receives data packets processed by the M protocol stacks and sends the data packets outwards through the network interface. The present disclosure implements a function of parallel protocol processing by multiple processes in user space of an operating system in a multi-core environment by using a parallel processing feature of a multi-core system, thereby reducing resource consumption caused by data packet copying.

Data processing method and a computer using distribution service module

A peripheral component interconnect express PCI-e network system having a processor for (a) assigning addresses to the PCI-e topology tree, comprising: traversing, at a given level and in a breadth direction, down-link couplings to an interconnection; ascertaining, at the level, which of the down-link couplings are connected to nodes; assigning, at the level, addresses to nodes of ascertained down-link coupling having nodes; and (b) propagating, a level, comprising: traversing, at the level and in a depth direction, down-link couplings to the interconnection of the PCI-e network, ascertaining, at the level, which of the downlink couplings are coupled to other interconnections in the depth direction, consecutively proceeding in the depth direction, to a next level of the down-link coupling of a next interconnection; and alternatively repeating (a) and (b) until the nodes are assigned addresses within the PCI-e tree topology network.

Method and system for aggregation-friendly address assignment to PCIe devices

Systems and methods for offloading computations from a CPU directly to an accelerator engine are disclosed. One embodiment includes determining a function of an application to be offloaded from a CPU to an accelerator engine, locating data within a file necessary to perform the functions, programming a logic of the accelerator engine based on the function to be offloaded, programming a DMA engine to move a copy the data from a secondary storage device to the accelerator engine, and processing the data at the accelerator engine using the programmed logic.

Acceleration framework with direct data transfer mechanism

A method implemented by a network element (NE) in a network, comprising composing a first network storage entity by mapping a plurality of logical storage units to a plurality of physical storage units in a physical storage system according to a first storage metric associated with the plurality of physical storage units, arranging the plurality of logical storage units sequentially to form a logical circular buffer, and designating a current logical storage unit for writing data and an upcoming logical storage unit for writing data after the current storage unit is fully written, and rebalancing the physical storage system while the physical storage system is actively performing network storage operations by relocating at least one of the logical storage units to a different physical storage unit according to a second storage metric associated with the plurality of physical storage units.

Rebalancing and elastic storage scheme with elastic named distributed circular buffers

An apparatus comprising at least one receiver configured to receive a traffic flow, receive information comprising a set of functions and an order of the set from a controller, and a processor coupled to the at least one receiver and configured to assign the traffic flow to one or more resources, determine a processing schedule for the traffic flow, and process the traffic flow by the set of functions, following the order of the set, using the one or more resources, and according to the processing schedule.

Improving the architecture of middleboxes or service routers to better consolidate diverse functions

A method for dynamically building a column store database from a row store database. The method includes establishing the row store database for storing data, wherein each row includes a plurality of attributes, and wherein data in row store database is current to a temporal point in time. The method includes establishing the column store database including data structured to satisfy received analytic queries. The method includes beginning from an initial state of the column store database, for each subsequently received analytic query, importing a targeted amount of data from a corresponding temporal state of the row store database into the column store database to satisfy the corresponding subsequently received analytic query.

Method and system for adaptively building a column store database from a temporal row store database based on query demands

A system comprising a unified interconnect network, a plurality of process memory modules, and a plurality of processor modules configured to share access to the memory modules via the unified interconnect network. Data may be communicated between a plurality of processor modules and a plurality of shared resource pools via a unified interconnect network, wherein the communications comprise a protocol that is common to all resource pools, and wherein each resource pool comprises a plurality of resource modules each configured to perform a common function. Further, a network interface controller (NIC) module may be configured to receive data from a plurality of processor modules via a unified interconnect network, and provide core network connectivity to the processor modules.

Disaggregated server architecture for data centers

An approach to offloading probabilistic computations is described. An application server comprising a memory and a processor and coupled to a network-attached storage device configured to create a dedicated process in response to a procedural call to a virtual machine container based on a data request is disclosed. The processor forwards the data request to the network-attached storage device, programs one or more virtual machines to perform a probabilistic computation based on the procedural call, and directs the probabilistic computation to a first virtual machine of the one or more virtual machines. The request for data is transformed into a modified call using a virtualized lookup call.

Offloading probabilistic computations in data analytics applications

Embodiments of the present invention provide a data processing system and a data processing method. An MMIO address of a data request is obtained, where data stored in a PCIe storage device can be directly obtained from the MMIO address, and according to the MMIO address, a network adapter can directly read the data from the PCIe storage device of the data processing system, and transmit the data to a second data processing system, or directly write the data received from the second data processing system into the PCIe storage device. Therefore, the processing system can implement transmitting data directly from the PCIe storage device to the network adapter without using memory. During data transmission between the two data processing systems, a usage ratio of a resource, such as memory and a CPU, is reduced, and efficiency of data transmission is improved.

Data processing system and data processing method

Embodiments of the present invention provide a solution for managing inter-domain resource allocation in a Peripheral Component Interconnect-Express (PCIe) network. One processor among a plurality of link processors is elected as a management processor. The management processor obtains information about available resources of PCIe network. When a resource request from a request processor is received, the management processor allocates a resource of the available resources to the requesting processor. The management processor instructs one or more link processors to program one or more inter-domain NTBs through which the traffic between the allocated resource and the requesting processor is going to flow according to the memory address information of the allocated resource, to allow cross-domain resource access between the requesting processor and the allocated resource.

Resource management for peripheral component interconnect-express domains

An apparatus for initialization. The apparatus includes a management I/O device controller for managing initialization of a plurality of I/O devices coupled to a PCI-Express (PCIe) fabric. The management I/O device controller is configured for receiving a request to register a target interrupt register address of a first worker computing resource, wherein the target interrupt register address is associated with a first interrupt generated by a first I/O device coupled to the PCIe fabric. A mapping module of the management I/O device controller is configured for mapping the target interrupt register address to a mapped interrupt register address of a domain in which the first I/O device resides. A translating interrupt register table includes a plurality of mapped interrupt register addresses in the domain that is associated with a plurality of target interrupt register addresses of a plurality of worker computing resources.

Delivering interrupts through non-transparent bridges in a PCI-express network

A device comprises a receiver configured to receive a join-lookup remote procedural call (RPC) for a file, wherein the join-lookup RPC requests a join operation of sub-files associated with the file, and a transmitter configured to transmit the file in response to the Join-Lookup RPC. A distributed file system (DFS) client comprises a transmitter configured to transmit a join-lookup RPC for a file, wherein the join-lookup RPC requests a join operation of sub-files associated with the file, and a receiver configured to receive the file in response to the Join-Lookup RPC. A method comprises receiving a join-lookup RPC for a file, wherein the join-lookup RPC requests a join operation of sub-files associated with the file, and transmitting the file in response to the Join-Lookup RPC.

File joining on back end device

Growing community of inventors

Cupertino, CA, United States of America

Guangyu Shi