WO2016067444A1 - Système de gestion de données, ordinateur, procédé et programme de gestion de données - Google Patents

Système de gestion de données, ordinateur, procédé et programme de gestion de données Download PDF

Info

Publication number
WO2016067444A1
WO2016067444A1 PCT/JP2014/079041 JP2014079041W WO2016067444A1 WO 2016067444 A1 WO2016067444 A1 WO 2016067444A1 JP 2014079041 W JP2014079041 W JP 2014079041W WO 2016067444 A1 WO2016067444 A1 WO 2016067444A1
Authority
WO
WIPO (PCT)
Prior art keywords
data
memory
computer
control
change message
Prior art date
Application number
PCT/JP2014/079041
Other languages
English (en)
Japanese (ja)
Inventor
坂倉　隆史
Original Assignee
三菱電機株式会社
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 三菱電機株式会社 filed Critical 三菱電機株式会社
Priority to JP2016556150A priority Critical patent/JP6099844B2/ja
Priority to PCT/JP2014/079041 priority patent/WO2016067444A1/fr
Priority to TW104101432A priority patent/TW201616364A/zh
Publication of WO2016067444A1 publication Critical patent/WO2016067444A1/fr

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F12/00—Accessing, addressing or allocating within memory systems or architectures
    • G06F12/02—Addressing or allocation; Relocation
    • G06F12/06—Addressing a physical block of locations, e.g. base addressing, module addressing, memory dedication
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F12/00—Accessing, addressing or allocating within memory systems or architectures
    • G06F12/02—Addressing or allocation; Relocation
    • G06F12/08—Addressing or allocation; Relocation in hierarchically structured memory systems, e.g. virtual memory systems
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F12/00—Accessing, addressing or allocating within memory systems or architectures
    • G06F12/02—Addressing or allocation; Relocation
    • G06F12/08—Addressing or allocation; Relocation in hierarchically structured memory systems, e.g. virtual memory systems
    • G06F12/10—Address translation

Definitions

  • the present invention relates to a data management system, a computer, a data management method, and a program.
  • the present invention relates to a virtual storage of an operating system, and relates to a data management system, a computer, a data management method, and a program related to FA (Factory Automation) and PA (Process Automation).
  • FA Vectory Automation
  • PA Process Automation
  • the virtual memory mechanism allocates a storage area for storing file data to 4KB pages, and implements a file read or write process using the memory access violation trap process implemented in virtual memory. ing.
  • This virtual storage mechanism is hidden from the application programmer, and the application programmer describes file access processing using file I / O API (Application, Programming, Interface).
  • Patent Document 1 discloses a data transfer width, a data transfer delay, and the like of an SRAM (Static, Random, Access, and Memory) and a memory connected to the IC by arranging a repeater in the IC (Integrated Circuit). It is disclosed that data transfer paths are reconfigured for different storage devices to improve processing efficiency.
  • SRAM Static, Random, Access, and Memory
  • An object of the present invention is to provide a data management system capable of doing so.
  • a data management system includes a control device having a control memory, a computer memory, and a computer connected to the control device via a network, wherein the control memory and the computer memory are common.
  • the controller is A control management unit for transmitting to the computer a first change message indicating that at least part of the data stored in the control memory has been changed to first data;
  • the calculator is With cache memory, The first change message is received, and at least a part of the data stored in the computer memory is changed to the first data based on the first change message, and an address corresponding to the first data
  • the control management unit of the control device transmits a first change message indicating that the data stored in the control memory is changed to the first data to the computer via the network
  • the memory management unit of the computer changes the data stored in the computer memory to the first data based on the first change message, invalidates the address corresponding to the first data
  • the access processing unit of the computer When the access to the first data occurs, the first data is written to the cache memory and the address corresponding to the first data is validated, so that the memory change in the control device using the cloud can be quickly made in the cloud The effect is that it can be reflected in the memory of the computer on the side.
  • FIG. 3 is a diagram for explaining an overview of a data management system according to the first embodiment.
  • FIG. 2 is a diagram showing an outline of a system configuration of a computer according to the first embodiment.
  • FIG. 3 is a diagram for explaining the data management system according to the first embodiment.
  • 1 is a block configuration diagram of a data management system according to Embodiment 1.
  • 5 is a flowchart showing a first change message transmission process and a memory management process in the data management method according to the first embodiment. 5 is a flowchart showing access exception processing in the data management method according to the first embodiment.
  • FIG. 6 is a block configuration diagram of a data management system 801 according to a second embodiment.
  • 10 is a flowchart showing a data management method of the data management system 801 according to the second embodiment.
  • FIG. 9 is a block configuration diagram of a data management system 803 according to a fourth embodiment.
  • 10 is a flowchart showing computer memory rewriting processing S211 according to the fourth embodiment.
  • Embodiment 1 FIG. A data management system 800 and a computer 200 according to the present embodiment will be described with reference to FIGS.
  • Data center 101 is deployed all over the world. For example, about 30 data centers 101 are deployed all over the world. The data centers 101 are connected by a dedicated line 103. Data in the data center 101 is guaranteed to be consistent in all data centers 101.
  • the data center 101 is composed of a huge number of computers 200. Each of thousands of rack mount racks, for example, the rack 104, accommodates about 12 blades on which a few CPUs are mounted. The data center 101 accommodates hundreds of thousands to millions of CPUs, and a control plane for distributing communication data, called “merchant silicon”, is interconnected by a switch that can be implemented by software.
  • the router 105 connects the data center 101 and the user of the data center 101.
  • the router 105 may have a firewall or a VPN (Virtual Private Network) router function from the viewpoint of security.
  • VPN Virtual Private Network
  • the gateway device 106 is a device that connects a network in a factory to the data center 101.
  • a PLC 107 (Programmable / Logic / Controller) is connected to a network in the factory via a gateway device 106.
  • the network in the factory includes a plurality of proprietary networks.
  • the PLC 107 is also connected to these networks, and is connected to the data center 101 via the heterogeneous network and the gateway device 106.
  • FIG. 2 shows an outline of the system configuration of the computer 200 in the blade accommodated in the rack 104.
  • the computer 200 includes a computer memory 201, a CPU 203, a disk device 205, and a NIC 206 (Network, Interface, Controller).
  • a computer memory 201 the memory of the computer 200 is referred to as a computer memory 201
  • the memory of the PLC 107 is referred to as a control memory 501.
  • the computer memory 201 is connected to a controller mounted in a CPU 203 called a north bridge. External I / O devices other than the computer memory 201 are connected to a controller mounted in the CPU 203 called a south bridge.
  • the south bridge includes, for example, a PCI-e 202 (Peripheral Component Interconnect Express), a SATA (Serial Advanced Attachment) 10-Gigabit Ethernet (registered trademark) NIC 206 to which a disk device 205 is connected.
  • a controller group such as PCI-e, GPIO 204 (General, Purpose, Input, Output) to which is connected is mounted.
  • the data management system 800 includes a computer 200 and a PLC 107.
  • the computer 200 and the PLC 107 are connected by a network 600.
  • the computer 200 includes a CPU 203, a computer memory 201, a disk device 205, and a NIC 206.
  • the CPU 203 includes a cache memory 253.
  • the cache memory 253 includes a primary hierarchy cache memory 301, a secondary hierarchy cache memory 302, and a tertiary hierarchy cache memory 303.
  • the disk device 205 is a non-volatile storage device in which a mapped file 2052 is stored.
  • a control memory 501 and a NIC 506 are connected to the PLC 107.
  • the PLC 107 may include the control memory 501 and the NIC 506.
  • the computer 200 and the PLC 107 are connected by a network 600 such as an IP (Internet Protocol) network via the NIC 206 and the NIC 506.
  • a network 600 such as an IP (Internet Protocol) network via the NIC 206 and the NIC 506.
  • the primary hierarchy cache memory 301 is a primary cache memory in which an object operated by the CPU core is stored, and is also referred to as an L1 cache memory.
  • the storage area 3011 is an area for storing the synchronization state / attribute information of the primary hierarchy cache memory 301.
  • the secondary tier cache memory 302 is a secondary cache memory that is a first back store, and is also referred to as an L2 cache memory.
  • the storage area 3021 is an area for storing the synchronization state / attribute information of the secondary hierarchy cache memory 302.
  • the tertiary hierarchy cache memory 303 is a tertiary cache memory that is a second back store, and is also referred to as an L3 cache memory.
  • the storage area 3031 is an area for storing synchronization state / attribute information of the tertiary hierarchy cache memory 303.
  • the computer memory 201 is a third back store.
  • the storage area 2011 is an area for storing the synchronization state / attribute information of the computer memory 201.
  • the computer memory 201 is, for example, a DRAM (Dynamic, Random, Access, Memory) called DDR4 (Double, Data, Rate4).
  • the disk device 205 functions as a fourth back store.
  • a storage area 2051 is a storage area for synchronization state / attribute information of the disk device 205.
  • the control memory 501 of the PLC 107 is a fifth back store.
  • the storage area 5011 is an area for storing synchronization state / attribute information of the control memory 501 of the PLC 107.
  • the control memory 501 is a DRAM called DDR3 (Double / Data / Rate 3), for example.
  • Operating system virtual storage is a system in which virtual addresses are allocated and managed in a memory area of a memory such as DRAM, which is a valuable resource of the system.
  • memory and storage devices such as magnetic disk storage devices that are non-volatile storage are mapped in units of pages of 4 KB units and cannot be accommodated on the memory, a memory area is virtually allocated and accommodated using the storage device as an auxiliary storage device Export pages that cannot.
  • the virtual memory gives a virtual large memory area to the execution program.
  • Virtual memory releases all pages used by lower-order processes according to the working set and process priority, which keeps the number of pages used by processes constant, in order to use memory more efficiently. It has been implemented with policies such as a method and a method of releasing pages that are not frequently used.
  • DDR4 Dynamic RAM
  • DIMM Direct, Inline, Memory, Module
  • the capacity of the DDR4 reaches 1 TB.
  • the capacity of 1 TB is a memory capacity that cannot possibly be consumed by the current operating system or application execution. Therefore, a storage device for saving pages that cannot be accommodated becomes unnecessary, and in many cases, the memory is positioned only as a storage area that can be accessed at high speed. The memory area reserved by the application program is released as it is. Therefore, by using DDR4 as the computer memory 201 that is the third back store, the processing speed of the computer 200 is improved.
  • the operating system uses a virtual storage mechanism in file read processing or write processing.
  • a storage area in which file data is stored is allocated to 4 KB pages, and a file access process or a memory access violation trap process implemented in the virtual storage is used.
  • Write processing is realized.
  • This virtual memory mechanism is hidden from the application programmer, and the application programmer describes the file access processing using the file I / O API.
  • the data management system 800 includes a control device 500 having a control memory 501 and a computer 200 having a computer memory 201 and connected to the control device 500 via a network 600.
  • the control memory 501 and the computer memory 201 store common data.
  • the common data stored in the control memory 501 and the computer memory 201 is, for example, PLC data acquired by the PLC 107.
  • the control data 501 and the computer memory 201 are handled as if the memory is shared by mapping the PLC data to the virtual address.
  • the computer 200 includes an application execution unit 251, an OS execution unit 252, a memory management unit 254, a cache memory 253, and a computer memory 201.
  • the OS execution unit 252 includes an access processing unit 2521.
  • control device 500 has a PLC 107.
  • the PLC 107 includes a control management unit 510 and a control memory 501.
  • the control management unit 510 transmits a first change message 601 indicating that at least a part of the data stored in the control memory 501 has been changed to the first data 614 to the computer 200.
  • the first data 614 will be described later.
  • the control management unit 510 includes the first data 614 and the address 613 corresponding to the first data 614 in the first change message 601.
  • the first change message 601 is a message transmitted from the control device 500 to the computer 200.
  • the second change message 602 is a message transmitted from the computer 200 to the control device 500.
  • the computer 200 includes a cache memory 253.
  • the memory management unit 254 receives the first change message 601, changes at least a part of the data stored in the computer memory 201 to the first data 614 based on the first change message 601, and the first data The address 613 corresponding to 614 is invalidated.
  • the address 613 corresponding to the first data 614 is set as the first address 613. That is, the memory management unit 254 receives the first change message 601 including the first data 614 and the first address 613, and based on the received first change message 601, the first address in the computer memory 201 The data corresponding to 613 is changed to the first data 614, and the first address 613 in the computer memory 201 is invalidated.
  • the access processing unit 2521 writes the first data 614 to the cache memory 253 and validates the first address 613 corresponding to the first data 614.
  • the application execution unit 251 executes an application software group.
  • the OS execution unit 252 executes an operating system including a virtual storage mechanism.
  • the memory management unit 254 manages a memory in which PLC data is stored. Software that manages the memory like the memory management unit 254 is referred to as a pager.
  • the memory management unit 254 is referred to as a PLC pager, for example.
  • the control management unit 510 of the PLC 107 performs data management of the PLC 107.
  • the control management unit 510 is also referred to as a PLC data management unit.
  • the control management unit 510 communicates with the memory management unit 254 of the computer 200 and changes data in the control memory 501 of the PLC 107.
  • the data change performed by the memory management unit 254 and the control management unit 510 includes data reference and data update.
  • FIG. 5 shows a data format 611 of the first change message and the second change message exchanged between the control management unit 510 and the memory management unit 254 according to the present embodiment.
  • the control memory 501 of the PLC 107 is mounted on a 16-bit space.
  • the number of data 612 to be changed in units of 2 bytes is stored.
  • a combination of an address 613, which is a 16-bit address corresponding to the number of data 612, and 2-byte data 614 corresponding to the address 613 is stored.
  • first data 614 corresponding to the first address 613 is stored, and second data 614 corresponding to the second address 613 is stored.
  • corresponding addresses 613 and data 614 are examples of corresponding first addresses 613 and first data 614.
  • the data management system 800 includes the control device 500 including the control memory 501, the computer memory 201 and the cache memory 253, and the computer 200 connected to the control device 500 via the network 600. .
  • the control memory 501 and the computer memory 201 store common data, that is, PLC data.
  • the control management unit 510 transmits a first change message 601 indicating the change of the PLC data to the memory management unit 254 of the computer 200.
  • the first change message 601 in S110 is transmitted by UDP (User Datagram Protocol) if the reliability of the layer 2 is high. This is to eliminate the delay time due to 3-Way handshake and window control that occurs in the case of TCP (Transmission, Control, Protocol). Further, in order to secure, the control management unit 510 that manages PLC data and the memory management unit 254 that is a PLC pager may hold a shared secret key in advance and encrypt / decrypt data.
  • ⁇ Memory management process S200> In S ⁇ b> 211 to S ⁇ b> 213, the memory management unit 254 of the computer 200 receives the first change message 601 from the control device 500, and first stores at least a part of the data stored in the computer memory 201 based on the first change message 601. And the memory management process for invalidating the address corresponding to the first data is executed.
  • the memory management unit 254 receives the first change message 601. As described with reference to FIG. 5, the first change message 601 includes an address 613 indicating a change location and data 614 after the change. In S212, the memory management unit 254 changes data corresponding to the address 613 in the computer memory 201 to data 614 corresponding to the address 613 based on the received first change message 601. That is, the memory management unit 254 rewrites the computer memory 201 with the content of the first change message 601.
  • the memory management unit 254 invalidates the address 613 in the computer memory 201.
  • the memory management unit 254 returns to S211 and waits for the first change message 601.
  • invalidating the address 613 in the computer memory 201 means invalidating the page table entry of the page including the data 614.
  • the page table is generated immediately after the data management system 800 starts operation.
  • the page table is data managed by the kernel of the operating system, and is composed of data called page table entries that constitute a page table that is about twice as large as the physical memory.
  • the access exception process S220 in the memory management method according to this embodiment will be described with reference to FIG.
  • ⁇ Access exception processing S220> when access to the data 614 occurs, the access processing unit 2521 of the OS execution unit 252 performs access exception processing that writes the data 614 to the cache memory 253 and validates the address 613 corresponding to the data 614. Execute.
  • the access processing unit 2521 causes the invalidated page of the computer memory 201 to be read into the multi-layer cache memory 253. That is, the access processing unit 2521 causes the multi-layer cache memory 253 to read the data 614 corresponding to the address 613 in the computer memory 201.
  • the access processing unit 2521 validates the page table entry of the page including the data 614. By the validation of the page table entry by the access processing unit 2521, the address 613 corresponding to the data 614 in the computer memory 201 is validated.
  • the virtual storage management method of the computer in the data management system includes a data management software specifying means corresponding to a 4 KB page area of physical memory, and a page table entry management means for storing mapping information to the virtual address of the page area.
  • a data management software specifying means corresponding to a 4 KB page area of physical memory
  • a page table entry management means for storing mapping information to the virtual address of the page area.
  • the data management system it is possible to cause a change in data generated in the PLC to occur in the memory of the computer and the multi-layer cache memory, and to change the memory on the PLC side immediately. Can be reflected in the memory of the computer and the multi-layer cache memory.
  • the contents of the computer memory and the contents of the control memory of the PLC can be shared via the network using the control management unit and the memory management unit. Sharing the computer memory and the control memory of the PLC means that it can be immediately reflected in the processing of the computer application. Therefore, according to the data management system according to the present embodiment, information of a large number of PLC memories, for example, information from sensors and devices provided in a production line and connected to the PLC is collected in a data center called a cloud. it can. According to the data management system according to the present embodiment, information in a large number of PLC control memories can be reflected in a wide variety of processes by applications installed on the cloud.
  • PLC data such as status information on the production facilities of companies deployed globally and the controllers and devices constituting the individual production facilities is stored in a virtual memory area.
  • the data management system according to the present embodiment provides means for flexibly reconfiguring an optimal data coherency maintenance mechanism for status information in consideration of usage requirements for status information, communication delay, and communication data amount.
  • Embodiment 2 differences from the first embodiment will be mainly described.
  • the same components as those described in the first embodiment are denoted by the same reference numerals, and the description thereof may be omitted.
  • the CPU 203 of the computer 200 changes data to the computer memory 201 managed by the memory management unit 254 that is a PLC pager.
  • the factory control is enabled by the CPU 203 of the computer 200.
  • the data change executed by the CPU 203 includes data reference and data update.
  • FIG. 8 is a block configuration diagram of the data management system 801 according to the present embodiment.
  • a data management system 801 illustrated in FIG. 8 includes a change determination unit 255 in addition to the configuration of the data management system 800 described in the first embodiment.
  • the data management system 801 uses the second change message 602 that is not used in the first embodiment.
  • the memory management unit 254 changes at least a part of the data stored in the computer memory 201 to the second data 614.
  • a second change message 602 indicating that this has been done is transmitted to the control device 500.
  • the control management unit 510 receives the second change message 602 transmitted from the memory management unit 254, and at least part of the data stored in the control memory 501 based on the second change message 602 is converted into the second data 614. change.
  • the change determination unit 255 controls the control memory 501 based on the second address 613 corresponding to the second data 614. It is determined whether or not at least a part of the data stored in is changed to the second data 614.
  • the memory management unit 254 transmits a second change message 602 to the control device 500 when the change determination unit 255 determines that at least a part of the data stored in the control memory 501 is changed to the second data 614. .
  • the CPU 203 changes or updates the data in the computer memory 201.
  • the application execution unit 251 of the computer 200 executes an application program that changes or updates PLC data in the computer memory 201.
  • the data at the second address 613 in the computer memory 201 is changed to the second data 614.
  • the change determination unit 255 determines whether or not the change or update of the computer memory 201 is reflected in the control memory 501 of the PLC 107.
  • the change determination unit 255 refers to the memory management data managed by the engineering environment of the PLC 107 and determines whether or not the data update to the PLC 107 is necessary.
  • the change determination unit 255 refers to the memory management data managed by the engineering environment of the PLC 107, determines whether or not the second address is a meaningful address, and whether or not the data update to the PLC 107 is necessary based on the determination result Determine whether.
  • the process ends.
  • the process proceeds to S412.
  • the change determination unit 255 notifies that the computer memory 201 has been rewritten by the memory management unit 254, which is a PLC pager, with a virtual address, that is, the second address 613.
  • the memory management unit 254 generates a second change message from the second data 614 on the memory corresponding to the virtual address, that is, the second address 613.
  • the memory management unit 254 transmits the generated second change message to the registered IP connection.
  • control management unit 510 receives the second change message 602 transmitted from the memory management unit 254.
  • control management unit 510 stores data corresponding to the second address 613 in the control memory 501 based on the second address 613 and the second data 614 included in the received second change message 602. Change to second data 614.
  • the change in the data in the control memory 501 is reflected in the cache memory of the PLC 107.
  • the data in the memory may be reflected in the cache memory.
  • the change determination unit determines whether or not to reflect the change in the control memory, so the second change message is not transmitted for an unnecessary change.
  • the factory can be efficiently controlled by the CPU of the computer.
  • Embodiment 3 FIG. In the present embodiment, differences from the first embodiment will be mainly described. In the present embodiment, the same components as those described in the first embodiment are denoted by the same reference numerals, and the description thereof may be omitted.
  • the data management system 802 includes a plurality of computers 200. That is, the control management unit 510 of the PLC 107 connects to the memory management unit 254 that is the PLC pager of each computer 200 of the plurality of computers 200 via the network 600.
  • PLC data that is common data is stored in the control memory 501 of the PLC 107 and the computer memory 201 included in each computer 200 of the plurality of computers 200.
  • the control management unit 510 of the PLC 107 is connected to the plurality of memory management units 254, and the PLC data is stored in a plurality of computer memories 201 in a multiplexed manner.
  • the first change message 601 is transmitted to each of the plurality of memory management units 254, so that the control memory 501 of the PLC 107, Synchronization can be established with the computer memories 201 of the plurality of computers 200.
  • the control memory of the PLC 107 is transmitted by transmitting the second change message 602 to the PLC 107.
  • 501 can be synchronized with each computer memory 201 of the plurality of computers 200. Note that an exclusive control mechanism between a plurality of PLC pagers does not have to be specially prepared. This is because updating data simultaneously from a plurality of computers does not occur in operation.
  • Embodiment 4 FIG. In the present embodiment, differences from the first embodiment will be mainly described.
  • the same components as those described in Embodiment 1 are denoted by the same reference numerals, and the description thereof may be omitted.
  • a block configuration diagram of the data management system 803 according to the present embodiment will be described with reference to FIG.
  • a data management system 803 illustrated in FIG. 10 includes a disk device management unit 256 that is a disk device pager that manages data of the disk device 205 and the disk device 205 in addition to the configuration of the data management system 800 described in the first embodiment. .
  • the computer 200 includes a disk device 205, and the computer memory 201 and the disk device 205 store PLC data that is common data.
  • the computer memory rewriting process S212 is a process corresponding to the computer memory rewriting process S212 of FIG.
  • the computer memory rewriting process S212 according to the present embodiment includes S212a and S212b.
  • the memory management unit 254 stores the data corresponding to the first address 613 in the computer memory 201 based on the received first change message 601 in the first The first data 614 corresponding to the address 613 is changed. Further, the memory management unit 254 changes the data of the disk device 205 mapped to the first address 613, which is the same virtual address, to the first data 614 corresponding to the first address 613, so that the disk device pager The contents of the first change message 601 are notified to the disk device management unit 256.
  • the disk device management unit 256 changes the data of the disk device 205 mapped to the first address 613 to the first data 614.
  • the same physical memory page is doubly mapped by the memory management unit that is the PLC pager and the disk device management unit that is the disk device pager. Therefore, a system with higher safety can be provided.
  • the computer may include a plurality of disk devices, and the same physical memory page may be mapped to the plurality of disk devices in a multiple manner.
  • the data on the control memory of the PLC is expanded on the memories of all the data center computers. That is, (1) When the PLC data is updated, the PLC-side memory is updated, and the physical PLC data main body stored in the disk device of the data center is updated. (2) When the memory on the PLC side is updated, a first change message of the memory is transmitted to the data center side, and the memory on the data center side is changed based on the first change message. Therefore, the application program enables data operation of a specific PLC by memory access at a virtual address, which gives the application programmer very high convenience.

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Engineering & Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Programmable Controllers (AREA)
  • Memory System Of A Hierarchy Structure (AREA)
  • Information Retrieval, Db Structures And Fs Structures Therefor (AREA)

Abstract

L'invention concerne un dispositif de commande (500) doté d'une unité de gestion de commande (510) qui envoie à un ordinateur (200) un premier message de modification (601) indiquant que les données stockées dans une mémoire de commande (501) ont été modifiées pour être remplacées par des premières données. L'ordinateur (200) comprend : une unité de gestion de mémoire (254) qui modifie les données stockées dans une mémoire d'ordinateur (201) pour les remplacer par les premières données, selon le premier message de modification (601), et désactive une adresse correspondant aux premières données ; et une unité de traitement d'accès (2521) qui, si un accès aux premières données est généré, écrit les premières données dans une antémémoire (253), puis active l'adresse.
PCT/JP2014/079041 2014-10-31 2014-10-31 Système de gestion de données, ordinateur, procédé et programme de gestion de données WO2016067444A1 (fr)

Priority Applications (3)

Application Number Priority Date Filing Date Title
JP2016556150A JP6099844B2 (ja) 2014-10-31 2014-10-31 データ管理システム、計算機、データ管理方法及びプログラム
PCT/JP2014/079041 WO2016067444A1 (fr) 2014-10-31 2014-10-31 Système de gestion de données, ordinateur, procédé et programme de gestion de données
TW104101432A TW201616364A (zh) 2014-10-31 2015-01-16 資料管理系統、計算機、資料管理方法及程式產品

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
PCT/JP2014/079041 WO2016067444A1 (fr) 2014-10-31 2014-10-31 Système de gestion de données, ordinateur, procédé et programme de gestion de données

Publications (1)

Publication Number Publication Date
WO2016067444A1 true WO2016067444A1 (fr) 2016-05-06

Family

ID=55856826

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/JP2014/079041 WO2016067444A1 (fr) 2014-10-31 2014-10-31 Système de gestion de données, ordinateur, procédé et programme de gestion de données

Country Status (3)

Country Link
JP (1) JP6099844B2 (fr)
TW (1) TW201616364A (fr)
WO (1) WO2016067444A1 (fr)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2019057097A (ja) * 2017-09-20 2019-04-11 株式会社東芝 コントローラ

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JPH07507174A (ja) * 1993-07-15 1995-08-03 ブル・エス・アー 情報システムにおけるメモリ管理方法、該方法を利用する情報システム
JP2000207278A (ja) * 1999-01-14 2000-07-28 Hitachi Ltd 分散共有主記憶を有する情報処理装置
JP2013073335A (ja) * 2011-09-27 2013-04-22 Toshiba Corp 情報処理装置

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JPH07507174A (ja) * 1993-07-15 1995-08-03 ブル・エス・アー 情報システムにおけるメモリ管理方法、該方法を利用する情報システム
JP2000207278A (ja) * 1999-01-14 2000-07-28 Hitachi Ltd 分散共有主記憶を有する情報処理装置
JP2013073335A (ja) * 2011-09-27 2013-04-22 Toshiba Corp 情報処理装置

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2019057097A (ja) * 2017-09-20 2019-04-11 株式会社東芝 コントローラ

Also Published As

Publication number Publication date
JPWO2016067444A1 (ja) 2017-04-27
TW201616364A (zh) 2016-05-01
JP6099844B2 (ja) 2017-03-22

Similar Documents

Publication Publication Date Title
Alian et al. Application-transparent near-memory processing architecture with memory channel network
US20210112003A1 (en) Network interface for data transport in heterogeneous computing environments
EP3140743B1 (fr) Méthodologie de cohérence de cache à base de répertoire d'interconnexion de système cube de mémoire hybride
US7979645B2 (en) Multiprocessor system for memory mapping of processing nodes
EP3073384B1 (fr) Attribution de mémoire fork-safe à partir de fichiers de mappage en mémoire avec un comportement de mémoire anonyme
KR20210147911A (ko) 캐시 일관성 상호 연결을 갖는 메모리
US7844746B2 (en) Accessing an effective address and determining whether the effective address is associated with remotely coupled I/O adapters
JP5841255B2 (ja) 仮想化入力／出力のためのプロセッサローカルコヒーレンシを有するコンピュータシステム
KR20210147976A (ko) 확장 가능하고 일관성 있는 메모리 장치를 위한 시스템 및 방법
US20160092362A1 (en) Memory network to route memory traffic and i/o traffic
US20130013889A1 (en) Memory management unit using stream identifiers
US20220283975A1 (en) Methods and apparatus for data descriptors for high speed data systems
US9996498B2 (en) Network memory
US10346345B2 (en) Core mapping
TW201502801A (zh) 網路上的記憶體分享
CN103870435A (zh) 服务器及数据访问方法
US20220179792A1 (en) Memory management device
JP6099844B2 (ja) データ管理システム、計算機、データ管理方法及びプログラム
EP2979193B1 (fr) Système de mémoire partagée
US20190121780A1 (en) Noncoherent interprocessor communication remapping node controller
US7904693B2 (en) Full virtualization of resources across an IP interconnect using page frame table
US9535851B2 (en) Transactional memory that performs a programmable address translation if a DAT bit in a transactional memory write command is set
Ros et al. Emc 2: Extending magny-cours coherence for large-scale servers
CN117280332A (zh) 通过具有多个片上系统的结构避免死锁
CN113849262A (zh) 用于无需复制而在虚拟机之间移动数据的技术

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 14905060

Country of ref document: EP

Kind code of ref document: A1

ENP Entry into the national phase

Ref document number: 2016556150

Country of ref document: JP

Kind code of ref document: A

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 14905060

Country of ref document: EP

Kind code of ref document: A1