CN105138494A - Multi-channel computer system - Google Patents

Multi-channel computer system Download PDF

Info

Publication number
CN105138494A
CN105138494A CN201510527490.5A CN201510527490A CN105138494A CN 105138494 A CN105138494 A CN 105138494A CN 201510527490 A CN201510527490 A CN 201510527490A CN 105138494 A CN105138494 A CN 105138494A
Authority
CN
China
Prior art keywords
module
pcie
disk
computer system
modules
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Granted
Application number
CN201510527490.5A
Other languages
Chinese (zh)
Other versions
CN105138494B (en
Inventor
王磊
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Inspur Beijing Electronic Information Industry Co Ltd
Original Assignee
Inspur Beijing Electronic Information Industry Co Ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Inspur Beijing Electronic Information Industry Co Ltd filed Critical Inspur Beijing Electronic Information Industry Co Ltd
Priority to CN201510527490.5A priority Critical patent/CN105138494B/en
Publication of CN105138494A publication Critical patent/CN105138494A/en
Application granted granted Critical
Publication of CN105138494B publication Critical patent/CN105138494B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Landscapes

  • Multi Processors (AREA)

Abstract

The invention discloses a multi-channel computer system. The system comprises integration application modules, calculation modules, distributed switching modules, a PCIe switching module, a storage control module, disk storage modules, an I/O extension module and basic control modules, wherein the integration application modules are in interconnection with the system based on PCIe signals; the calculation modules are responsible for calculation and processing of application data inside the system and in interconnection with the system through a PCIe signal bus; the distributed switching modules connect all the calculation modules inside the system and improve rapid connection channels outside the system; the PCIe switching module is responsible for receiving, switching and transmitting all PCIe data in the whole system; the storage control module connects the PCIe switching module with the disk storage modules and converts the PCIe signals into disk storage signals; the disk storage modules construct disk arrays out of interior independent physical disks through the storage control module; the I/O extension module provides an extension slot; the basic control modules conduct monitoring and management on all the modules in the system. By means of the multi-channel computer system, the performance of the whole computer system is improved.

Description

A kind of multichannel computer system
Technical field
The present invention relates to computer memory technical field, espespecially a kind of new type multipath computer system exchanged based on fusion architecture support multinode overall situation share I O, storage and distributed network.
Background technology
Along with the growth of current network and data volume explosion type, very high requirement is proposed to the calculating of computer system, network, storage capacity.As shown in Figure 1, in computer system architecture traditional at present, owing to cannot solve the shared use of underlying device between many physical systems from hardware view, can only realize an independent physical system inside by the mode of software virtual machine, multiple virtual machine is to the access of bottom input and output (IO, Input/Output) and shared device.
In large-scale computer system, owing to needing powerful computing power, often there is the computing subsystem of more than physics, although multiple physics subsystem can be made to have access to memory device by the mode of virtual machine, but memory device end needs for each virtual machine sets up corresponding independent disk space, the share and access of whole storage resources cannot be realized, be unfavorable for focus utilization and the management of resource.Meanwhile, existing framework dirigibility is relatively inflexible, and functional module is fixed, and cannot exchange as required between multiple functional module, because which limit the scalability of whole product.In addition, in active computer framework, often network controller is all only connected with single processing unit, when multiple processing unit needs the access carrying out network data simultaneously, can only be realized by the processing unit be connected with network control device, the calculating advantage of multiloop loop system cannot be made full use of like this, cause waste in multi-computing system between computational resource.
Summary of the invention
In order to solve the problems of the technologies described above, the invention provides a kind of multichannel computer system, the performance of whole computer architecture can be promoted.
In order to reach the object of the invention, the invention provides a kind of multichannel computer system, comprise: fusion application module, computing module, distributed Switching Module, quick peripheral assembly interconnecting PCIe Switching Module, storage control module, disk resident storage drive, input and output I/O expansion module and base control module, wherein, fusion application module, interconnects for Based PC Ie Signals & Systems; Computing module, for calculating and the process of application data in responsible whole system, is interconnected by the signal bus of PCIe and system; Distributed Switching Module, for connecting all computing modules in system, improves fast interface channel outward to system; PCIe Switching Module, for PCIe data in responsible system reception, exchange with forwarding; Storage control module, for connecting PCIe Switching Module and disk resident storage drive, is converted to disk storage signal by PCIe signal; Disk resident storage drive, for setting up into disk array by storage control module by the physical disk of internal independence; I/O expansion module, for providing expansion slot; Base control module, for the monitor and managment to each module of internal system.
Further, described fusion application module, specifically for: comprise memory module, Cache accelerating module and I/O and strengthen module, described memory module, Cache accelerating module and I/O strengthen module equal Based PC Ie signal and whole system interconnects.
Further, described distributed Switching Module, specifically for: to connecting all computing modules in system, to the high speed interface channel providing 3 100Gb/s outside system; Set up into distributed network topology by described distributed Switching Module, carry out based on distributed parallel computation.
Further, described PCIe Switching Module, specifically for: comprise processor and PLX exchange chip, be responsible for the reception of PCIe data in whole system, exchange and forward; By PCIe Switching Module by the disk in disk resident storage drive and I/O expansion module and the complete pond of I/O equipment, be supplied to computing module and use.
Further, described storage control module, specifically for: connect PCIe Switching Module and disk resident storage drive, PCIe signal be converted to SAS/SATA signal, and provide for data in magnetic disk on disk resident storage drive read-write, store and control.
Further, described disk resident storage drive, specifically for: comprise multiple 2.5 cun of physical store dishes separately, maximumly on single disk resident storage drive can support 12 2.5 cun of disks, whole system supports 2 disk resident storage drives, can support at most 24 2.5 cun of disks; Independent physical disk by storage control module set up into RAID0,5, the multiple disk arrays such as 6.
Further, described I/O expansion module, specifically for: the expansion slot that 10 PCIe3.0 are provided, the Ethernet of collocation standard, optical-fibre channel, Cache accelerate or computing accelerator card; I/O expansion module, by PCIe Switching Module and computing module and fusion application model calling, by I/O resource pool, is unifiedly supplied to computing module and the use of fusion application module.
Further, described base control module, specifically for: monitor and managment is carried out to module each in system, and provides general network interface and outside supvr or concentrated communication and management equipment to be connected.
Further, described multichannel computer system also comprises: distributed network module; Interconnect between the system that the 100Gb/s link of the both sides of described distributed network module is adjacent with left and right, interconnect between the middle link of 1 100Gb/s and the system of opposite face, and the expansion of double-vane is supported in the left and right sides, build the supercomputer system cluster based on distributed network.
Further, described PCIe Switching Module, also comprises: uplink port, downlink port and management port, and wherein, uplink port connects computing module and fusion application module; Downlink port is used for connecting I/O expansion module and storage control module; Management port connects microprocessing unit by I2C link.
Further, described distributed Switching Module comprises PCIe exchange control unit and 100Gb network controller, wherein, PCIe exchange control unit is up passes through the link of 8 PCIe3.0x4 and computing module or fusion application model calling, and the descending link passing through 1 PCIe3.0x16 is connected with 100Gb network controller; The up link by PCIe3.0x16 of 100Gb network controller is connected with PCIe exchange control unit, PCIe signal is converted to network signal and outside network device carries out communication, the descending high speed interface channel providing 3 100Gb/s, for carrying out the expansion of network.
Compared to the prior art, the present invention introduces based on novel PC Ie switching technology in new computer system architecture, solve in computer system architecture traditional at present owing to cannot solve the shared use of underlying device between many physical systems from hardware view,, multiple virtual machine inner the independent physical system problem to the access of bottom IO and shared device can only be realized by the mode of software virtual machine, release the new computer framework based on MultiHost multinode overall situation share I O and storage.
Distributed network architecture is incorporated in multichannel Computer System Design in new computer system architecture, by the distributed network module in system not only can realize between inside computer system multiple information processing generic module based on distributed exchanges data, distributed network module externally also provides the communication link of 3 100Gb/s bandwidth simultaneously, can set up into based on distributed network between multiple novel computer system architecture, at this moment each novel computer system is using as the computing unit of in distributed network, and the distributed network based on distributed network modular assembly also supports that double-vane is expanded, the extension of network of novel computer framework and the expansion of calculated performance aspect can be strengthened further.
In computer body system structure, introduce 100Gb/s exchange interconnecting channels at a high speed, as the exchanges data path of core, the data transfer bandwidth of the whole computer architecture of Integral lifting and transfer efficiency, meet the demand of following large data age.
Other features and advantages of the present invention will be set forth in the following description, and, partly become apparent from instructions, or understand by implementing the present invention.Object of the present invention and other advantages realize by structure specifically noted in instructions, claims and accompanying drawing and obtain.
Accompanying drawing explanation
Accompanying drawing is used to provide the further understanding to technical solution of the present invention, and forms a part for instructions, is used from and explains technical scheme of the present invention, do not form the restriction to technical solution of the present invention with the embodiment one of the application.
Fig. 1 is the schematic diagram of the convention computer architecture that in prior art, global resource is shared.
Fig. 2 is the schematic diagram of the novel computer framework that in a kind of embodiment of the present invention, global resource is shared.
Fig. 3 is the configuration diagram of multichannel computer system in a kind of embodiment of the present invention.
Fig. 4 is the configuration diagram that in a kind of embodiment of the present invention, multichannel computer system supports double-vane expansion.
Fig. 5 is the structural representation of PCIe Switching Module in a kind of embodiment of the present invention.
Fig. 6 is the structural representation of distributed Switching Module in a kind of embodiment of the present invention.
Embodiment
For making the object, technical solutions and advantages of the present invention clearly understand, hereinafter will be described in detail to embodiments of the invention by reference to the accompanying drawings.It should be noted that, when not conflicting, the embodiment in the application and the feature in embodiment can combination in any mutually.
Can perform in the computer system of such as one group of computer executable instructions in the step shown in the process flow diagram of accompanying drawing.Further, although show logical order in flow charts, in some cases, can be different from the step shown or described by order execution herein.
Fig. 2 is the schematic diagram of the novel computer framework that in a kind of embodiment of the present invention, global resource is shared.As shown in Figure 2, introduce in new computer system architecture based on novel PC Ie switching technology, by quick peripheral assembly interconnecting (PCIe, PeripheralComponentInterconnectExpress) Switching Module realize based on MultiHost multinode the overall situation share I O and storage resources share, now no matter for multinode or arbitrarily virtual machine will be clipped to one piece of same disk storage space, instead of traditional separate, stored block.
Fig. 3 is the configuration diagram of multichannel computer system in a kind of embodiment of the present invention.As shown in Figure 3, comprise fusion application module, computing module, distributed Switching Module, PCIe Switching Module, storage control module, disk resident storage drive, I/O expansion module and base control module, wherein,
Fusion application module: be divided into memory module, Cache accelerating module, I/O to strengthen module three kinds, various module is all that Based PC Ie signal and whole system interconnect, and user can change between modules according to the actual needs of oneself;
Computing module: calculating and the process of being responsible for application data in whole system, the processor in computing module can select X 86 processor or the arm processor of main flow Intel Company on market, is interconnected by the signal bus of PCIe and whole system;
Distributed Switching Module: whole distributed Switching Module internally connects all computing modules, the high speed interface channel of 3 100Gb/s is externally provided, distributed network topology can be set up into by distributed Switching Module, can realize based on distributed parallel computation, expand the computing ability of whole novel computer system;
PCIe Switching Module: be responsible for the reception of PCIe data in whole system, exchange and forward.Primarily of processor, PLX exchange chip composition in crosspoint design, owing to considering the quantity of the bandwidth sum data channel that data are transmitted, the employing of exchanges data core is combined by 3 PLX exchange chips and is formed; Can realize the disk in disk resident storage drive and I/O expansion module and the complete pond of I/O equipment by PCIe Switching Module, be supplied to computing module and use;
Storage control module: be mainly used to connect PCIe Switching Module and disk resident storage drive, its effect PCIe signal is converted to traditional disk storage signals such as SAS/SATA by memory controller be connected to disk resident storage drive, provide for data in magnetic disk on disk resident storage drive read-write, store control;
Disk resident storage drive: the logic magnetic disc that disk resident storage drive stores as data in whole system, be made up of multiple 2.5 cun of physical store dishes separately, maximumly on single disk resident storage drive can support 12 2.5 cun of disks, whole system supports 2 disk resident storage drives, can support at most 24 2.5 cun of disks.Each independently physical disk on disk resident storage drive by storage control module can set up into RAID0,5, the multiple disk arrays such as 6, the read-write of data simultaneously also controls by disk control module;
I/O expansion module: expansion slot I/O expansion module providing 10 PCIe3.0, the Ethernet of standard of can arranging in pairs or groups, optical-fibre channel, Cache accelerate or computing accelerator card, can the I/O extended capability of overall enhanced computer system.I/O expansion module, by PCIe Switching Module and computing module and fusion application model calling, realizes the I/O resource pool on it, can unify to be supplied to computing module and the use of fusion application module;
Base control module: as the control module in whole system, realize the monitor and managment to each module of internal system on the one hand, such as realize for basic operations such as the monitoring of the duties such as each module Primary Component voltage, temperature and firmware renewal, remote on-offs, also provide general network interface and outside supvr or concentrated communication and management equipment to be connected simultaneously.
Fig. 4 is the configuration diagram that in a kind of embodiment of the present invention, multichannel computer system supports double-vane expansion.As shown in Figure 4, distributed network module can be had in each system in novel computer architecture, distributed network module also provides the communication link of 3 100Gb/s bandwidth outward, interconnect between the system that wherein the 100Gb/s link of both sides is adjacent with left and right, and 1 of centre interconnects between the link of 100Gb/s and the system of opposite face, support the expansion of double-vane in the left and right sides based on this framework simultaneously, the supercomputer system cluster based on distributed network can be built, powerful calculating and extended capability are provided.
Fig. 5 is the structural representation of PCIe Switching Module in a kind of embodiment of the present invention.As shown in Figure 5, PCIe Switching Module is responsible for reception, the exchange of PCIe data in whole system and is forwarded.Primarily of processor, PLX exchange chip composition in Switching Module design, can realize the disk in disk resident storage drive and I/O expansion module and the complete pond of I/O equipment by PCIe Switching Module, be supplied to computing unit and use.Three generic ports are divided in PCIe Switching Module, be respectively uplink port, downlink port and management port, wherein: uplink port is used for connecting and calculates and fusion application module, and each connecting link provides the connection bandwidth of PCIe3.0x4, and transfer rate can reach 32Gb/s; Downlink port is used for connecting I/O expansion module and storage control module, and each connecting link provides the connection bandwidth of PCIe3.0x8, and transfer rate can reach 64Gb/s; Management port connects microprocessing unit by I2C link, and microprocessing unit, based on the design of the low power processor such as general ARM, is mainly used to carry out basic configuration, the supervision of key state and control to module.
Fig. 6 is the structural representation of distributed Switching Module in a kind of embodiment of the present invention.As shown in Figure 6, whole distributed Switching Module internally connects all computing modules, the high speed interface channel of 3 100Gb/s is externally provided, distributed network topology can be set up into by distributed Switching Module, can realize based on distributed parallel computation, expand the computing ability of whole novel computer system.Inner primarily of PCIe exchange control unit and 100Gb network controller composition at distributed module, wherein: PCIe exchange control unit is up passes through the link of 8 PCIe3.0x4 and calculating or fusion application model calling, one-way transmission speed can reach 32Gb/s, the descending link passing through 1 PCIe3.0x16 is connected with 100Gb network controller, and one-way transmission speed can reach 108Gb/s; The up link by PCIe3.0x16 of 100Gb network controller is connected with PCIe exchange control unit, PCIe signal is converted to network signal and outside network device carries out communication.The descending high speed interface channel providing 3 100Gb/s, for carrying out the expansion of network.
Although the embodiment disclosed by the present invention is as above, the embodiment that described content only adopts for ease of understanding the present invention, and be not used to limit the present invention.Those of skill in the art belonging to any the present invention; under the prerequisite not departing from the spirit and scope disclosed by the present invention; any amendment and change can be carried out in the form implemented and details; but scope of patent protection of the present invention, the scope that still must define with appending claims is as the criterion.

Claims (11)

1. a multichannel computer system, it is characterized in that, comprise: fusion application module, computing module, distributed Switching Module, quick peripheral assembly interconnecting PCIe Switching Module, storage control module, disk resident storage drive, input and output I/O expansion module and base control module, wherein
Fusion application module, interconnects for Based PC Ie Signals & Systems;
Computing module, for calculating and the process of application data in responsible whole system, is interconnected by the signal bus of PCIe and system;
Distributed Switching Module, for connecting all computing modules in system, improves fast interface channel outward to system;
PCIe Switching Module, for PCIe data in responsible system reception, exchange with forwarding;
Storage control module, for connecting PCIe Switching Module and disk resident storage drive, is converted to disk storage signal by PCIe signal;
Disk resident storage drive, for setting up into disk array by storage control module by the physical disk of internal independence;
I/O expansion module, for providing expansion slot;
Base control module, for the monitor and managment to each module of internal system.
2. multichannel computer system according to claim 1, it is characterized in that, described fusion application module, specifically for: comprise memory module, Cache accelerating module and I/O and strengthen module, described memory module, Cache accelerating module and I/O strengthen module equal Based PC Ie signal and whole system interconnects.
3. multichannel computer system according to claim 1, is characterized in that, described distributed Switching Module, specifically for: to connecting all computing modules in system, to the high speed interface channel providing 3 100Gb/s outside system;
Set up into distributed network topology by described distributed Switching Module, carry out based on distributed parallel computation.
4. multichannel computer system according to claim 1, is characterized in that, described PCIe Switching Module, specifically for: comprise processor and PLX exchange chip, be responsible for the reception of PCIe data in whole system, exchange and forward;
By PCIe Switching Module by the disk in disk resident storage drive and I/O expansion module and the complete pond of I/O equipment, be supplied to computing module and use.
5. multichannel computer system according to claim 1, it is characterized in that, described storage control module, specifically for: connect PCIe Switching Module and disk resident storage drive, PCIe signal is converted to SAS/SATA signal, and provide for data in magnetic disk on disk resident storage drive read-write, store control.
6. multichannel computer system according to claim 1, it is characterized in that, described disk resident storage drive, specifically for: comprise multiple 2.5 cun of physical store dishes separately, maximumly on single disk resident storage drive can support 12 2.5 cun of disks, whole system supports 2 disk resident storage drives, can support at most 24 2.5 cun of disks; Independent physical disk by storage control module set up into RAID0,5, the multiple disk arrays such as 6.
7. multichannel computer system according to claim 1, is characterized in that, described I/O expansion module, specifically for: the expansion slot that 10 PCIe3.0 are provided, the Ethernet of collocation standard, optical-fibre channel, Cache accelerate or computing accelerator card;
I/O expansion module, by PCIe Switching Module and computing module and fusion application model calling, by I/O resource pool, is unifiedly supplied to computing module and the use of fusion application module.
8. multichannel computer system according to claim 1, it is characterized in that, described base control module, specifically for: monitor and managment is carried out to module each in system, and provides general network interface and outside supvr or concentrated communication and management equipment to be connected.
9. multichannel computer system according to claim 1, is characterized in that, described multichannel computer system also comprises: distributed network module; Interconnect between the system that the 100Gb/s link of the both sides of described distributed network module is adjacent with left and right, interconnect between the middle link of 1 100Gb/s and the system of opposite face, and the expansion of double-vane is supported in the left and right sides, build the supercomputer system cluster based on distributed network.
10. multichannel computer system according to claim 4, is characterized in that, described PCIe Switching Module, also comprises: uplink port, downlink port and management port, and wherein, uplink port connects computing module and fusion application module; Downlink port is used for connecting I/O expansion module and storage control module; Management port connects microprocessing unit by I2C link.
11. multichannel computer systems according to claim 3, it is characterized in that, described distributed Switching Module comprises PCIe exchange control unit and 100Gb network controller, wherein, PCIe exchange control unit is up passes through the link of 8 PCIe3.0x4 and computing module or fusion application model calling, and the descending link passing through 1 PCIe3.0x16 is connected with 100Gb network controller; The up link by PCIe3.0x16 of 100Gb network controller is connected with PCIe exchange control unit, PCIe signal is converted to network signal and outside network device carries out communication, the descending high speed interface channel providing 3 100Gb/s, for carrying out the expansion of network.
CN201510527490.5A 2015-08-25 2015-08-25 A kind of multichannel computer system Active CN105138494B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201510527490.5A CN105138494B (en) 2015-08-25 2015-08-25 A kind of multichannel computer system

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201510527490.5A CN105138494B (en) 2015-08-25 2015-08-25 A kind of multichannel computer system

Publications (2)

Publication Number Publication Date
CN105138494A true CN105138494A (en) 2015-12-09
CN105138494B CN105138494B (en) 2018-11-20

Family

ID=54723845

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201510527490.5A Active CN105138494B (en) 2015-08-25 2015-08-25 A kind of multichannel computer system

Country Status (1)

Country Link
CN (1) CN105138494B (en)

Cited By (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN105515667A (en) * 2015-12-11 2016-04-20 浪潮(北京)电子信息产业有限公司 High-availability computer system
CN106020737A (en) * 2016-06-16 2016-10-12 浪潮(北京)电子信息产业有限公司 Globally-shared-disc high-density storage frame system
CN106844263A (en) * 2016-12-26 2017-06-13 中国科学院计算技术研究所 It is a kind of based on configurable multiprocessor computer system and implementation method
CN107220195A (en) * 2017-05-26 2017-09-29 郑州云海信息技术有限公司 A kind of multichannel computer system architecture for supporting high density NVMe to store
CN108701110A (en) * 2016-04-01 2018-10-23 英特尔公司 A kind of mechanism for the PCIE cable Topology Discoveries in rack scale architecture environment
WO2018209889A1 (en) * 2017-05-16 2018-11-22 杭州海康威视数字技术股份有限公司 Extended storage device based on pcie bus

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101540685A (en) * 2008-06-06 2009-09-23 曙光信息产业(北京)有限公司 PCIe shared storage blade for blade server
CN101540686A (en) * 2008-06-06 2009-09-23 曙光信息产业(北京)有限公司 PCIe switching device for blade server
CN103179385A (en) * 2013-03-29 2013-06-26 苏州皓泰视频技术有限公司 Extensible network video record (NVR) system based on multi-core processor
CN104639469A (en) * 2015-02-06 2015-05-20 方一信息科技(上海)有限公司 Computing and storing cluster system based on PCIE (Peripheral Component Interconnect Express) interconnection

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101540685A (en) * 2008-06-06 2009-09-23 曙光信息产业(北京)有限公司 PCIe shared storage blade for blade server
CN101540686A (en) * 2008-06-06 2009-09-23 曙光信息产业(北京)有限公司 PCIe switching device for blade server
CN103179385A (en) * 2013-03-29 2013-06-26 苏州皓泰视频技术有限公司 Extensible network video record (NVR) system based on multi-core processor
CN104639469A (en) * 2015-02-06 2015-05-20 方一信息科技(上海)有限公司 Computing and storing cluster system based on PCIE (Peripheral Component Interconnect Express) interconnection

Cited By (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN105515667A (en) * 2015-12-11 2016-04-20 浪潮(北京)电子信息产业有限公司 High-availability computer system
CN108701110A (en) * 2016-04-01 2018-10-23 英特尔公司 A kind of mechanism for the PCIE cable Topology Discoveries in rack scale architecture environment
CN108701110B (en) * 2016-04-01 2023-01-31 英特尔公司 Mechanism for PCIE cable topology discovery in rack-scale architecture environment
CN106020737A (en) * 2016-06-16 2016-10-12 浪潮(北京)电子信息产业有限公司 Globally-shared-disc high-density storage frame system
CN106844263A (en) * 2016-12-26 2017-06-13 中国科学院计算技术研究所 It is a kind of based on configurable multiprocessor computer system and implementation method
CN106844263B (en) * 2016-12-26 2020-07-03 中国科学院计算技术研究所 Configurable multiprocessor-based computer system and implementation method
WO2018209889A1 (en) * 2017-05-16 2018-11-22 杭州海康威视数字技术股份有限公司 Extended storage device based on pcie bus
US11100037B2 (en) 2017-05-16 2021-08-24 Hangzhou Hikvision Digital Technology Co., Ltd. Extended storage device based on PCIe bus
CN107220195A (en) * 2017-05-26 2017-09-29 郑州云海信息技术有限公司 A kind of multichannel computer system architecture for supporting high density NVMe to store

Also Published As

Publication number Publication date
CN105138494B (en) 2018-11-20

Similar Documents

Publication Publication Date Title
CN105138494A (en) Multi-channel computer system
US10409766B2 (en) Computer subsystem and computer system with composite nodes in an interconnection structure
US10346156B2 (en) Single microcontroller based management of multiple compute nodes
US11971836B2 (en) Network-on-chip data processing method and device
CN101188599B (en) Implementation method for load balance design of electric monitoring front system of power plant
WO2012130134A1 (en) Computer system
CN109242754A (en) A kind of more GPU High performance processing systems based on OpenVPX platform
CN204833236U (en) Support memory system of hybrid storage
CN104135514B (en) Fusion type virtual storage system
CN102929363B (en) A kind of method for designing of high-density blade server
CN104601684A (en) Cloud server system
CN104317770A (en) Data storage structure and data access method for multiple core processing system
CN104967577B (en) SAS switch and server
CN109586956B (en) FCoE switch capable of flexibly configuring ports and method
CN106844263B (en) Configurable multiprocessor-based computer system and implementation method
CN107220195A (en) A kind of multichannel computer system architecture for supporting high density NVMe to store
CN116185641B (en) Fusion architecture system, nonvolatile storage system and storage resource acquisition method
CN105099776A (en) Cloud server management system
CN106020737A (en) Globally-shared-disc high-density storage frame system
CN111079908B (en) Network-on-chip data processing method, storage medium, computer device and apparatus
CN104679714A (en) Supercomputer cluster based on ATCA (advanced telecom computing architecture)
CN105357461A (en) OpenVPX-based ultra-high-definition video recording platform for unmanned aerial vehicle
CN211149445U (en) High-speed data processing platform
CN104898775A (en) Calculation apparatus, storage device, network switching device and computer system architecture
CN102508797B (en) Flash memory control expanding module, controller, storage system and data transmission method thereof

Legal Events

Date Code Title Description
C06 Publication
PB01 Publication
C10 Entry into substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant