CN100349142C - Remote page access method for use in shared virtual memory system and network interface card - Google Patents
Remote page access method for use in shared virtual memory system and network interface card Download PDFInfo
- Publication number
- CN100349142C CN100349142C CNB2004100427427A CN200410042742A CN100349142C CN 100349142 C CN100349142 C CN 100349142C CN B2004100427427 A CNB2004100427427 A CN B2004100427427A CN 200410042742 A CN200410042742 A CN 200410042742A CN 100349142 C CN100349142 C CN 100349142C
- Authority
- CN
- China
- Prior art keywords
- network interface
- page
- interface unit
- getting
- long
- 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.)
- Expired - Lifetime
Links
Images
Landscapes
- Memory System Of A Hierarchy Structure (AREA)
Abstract
本发明公开了一种用于虚拟共享存储系统的远程取页方法及网络接口卡。该方法将网络接口卡上的存储器区域映射到用户进程的虚地址空间；源节点的用户进程直接访问其网络接口卡，产生并向目的节点的网络接口卡发送远程取页请求消息；目的节点的网络接口卡直接读取目的节点中的所需页；目的节点的网络接口卡向源节点的网络接口卡返回远程取页应答消息；源节点的网络接口卡直接将所述页写到源节点的内存区域。该网络接口卡增加了帧头分析逻辑、RDMA操作逻辑和虚实地址转换逻辑。在本发明中，用网络接口卡上的硬件处理大部分的协议开销，用户进程和网络接口卡可以双向直接访问，可以在不打断远程节点CPU的当前工作的情况下实现远程取页操作。
The invention discloses a remote page fetching method and a network interface card for a virtual shared storage system. The method maps the memory area on the network interface card to the virtual address space of the user process; the user process of the source node directly accesses its network interface card, generates and sends a remote page fetching request message to the network interface card of the destination node; The network interface card directly reads the required page in the destination node; the network interface card of the destination node returns a remote page fetching response message to the network interface card of the source node; the network interface card of the source node directly writes the page to the source node memory area. The network interface card adds frame header analysis logic, RDMA operation logic and virtual-real address conversion logic. In the present invention, the hardware on the network interface card is used to process most of the protocol overhead, and the user process and the network interface card can be directly accessed bidirectionally, and the remote page fetching operation can be realized without interrupting the current work of the remote node CPU.
Description
Technical field
The present invention relates to computer system, more particularly, relate to a kind of long-range page method of getting that is used for shared virtual memory system.
Background technology
Because share the easy programming of storage system (down with), the researchist considers to realize sharing the programming mode of storage on many machines that communicate with the message transmission, this is virtual shared storage.Virtual shared storage has easy programming simultaneously and because most of shared virtual memory system realizes possessing the good ratio of performance to price on a group of planes.Shared virtual memory system is many to be to share granularity with page or leaf.
The long-range page or leaf of getting refers to produce when source node skips leaf the signal that skips leaf, when finding that shared virtual address is not in the source node address scope, signal handler produces the long-range page or leaf request of getting, destination node is searched physical address corresponding according to virtual address after receiving and getting the page or leaf request, and the page or leaf with the physical address correspondence sends it back source node then.Traditional shared virtual memory system about long-range way of getting page or leaf is: have no progeny during source node is found to skip leaf and get the message of page or leaf request to corresponding process (rather than the network interface unit hardware) transmission of destination node, the operate as normal of the CPU of destination node will be got the page or leaf request and be interrupted, the content of pages that will read is encapsulated and sends message with form of message, source node receive return read page return messages, the CPU of source node is interrupted deblocking message once more and message body (content of pages) is delivered to corresponding page.
As mentioned above, traditional long-range page method of getting need produce distance host and interrupts, and has no progeny during distance host receives to stop ongoing calculating and carry out break in service, makes to calculate with communicate by letter can not be overlapping, has reduced the efficient of parallel computation.Simultaneously, the content of the page or leaf form with message is encapsulated and copies, increased extra expense.
Therefore need a kind of method, can reduce the long-range page expense of getting, make calculate with communicate by letter overlapping.
Summary of the invention
One of purpose of the present invention is to provide a kind of long-range page method of getting, and reduces long-range protocol processes expense of getting page or leaf; Two of purpose of the present invention is to provide a kind of long-range page method of getting, and does not interrupt the calculating of distance host when carrying out long-range get page, obtain to calculate with communicate by letter overlapping; Three of purpose of the present invention is to provide a kind of long-range page method of getting, and makes the long-range page or leaf energy fast return of getting.
To achieve these goals, the invention provides a kind of long-range page method of getting that is used for shared virtual memory system, described shared virtual memory system comprises with network interface unit and interconnected source node and the destination node of switch; Comprise step:
1) memory area on the network interface unit is mapped to the virtual address space of consumer process;
2) consumer process of source node is directly visited its network interface unit, produces and sends the long-range page request message of getting to the network interface unit of destination node;
3) network interface unit of destination node receives the long-range page request message of getting from source node;
4) network interface unit of destination node directly reads the required page or leaf in the destination node;
5) network interface unit of destination node is returned the long-range page response message of getting to the network interface unit of source node, comprises the page or leaf that reads in the step 4) in this message;
6) network interface unit of source node receives the described long-range page response message of getting;
7) network interface unit of source node is directly write described page or leaf the region of memory of source node.
In order to implement said method, the present invention also provides a kind of network interface unit that is used for shared virtual memory system, comprises that the pci interface controller transmits operation logic, control register logic, transmission FIFO and reception FIFO with message respectively by local bus and is electrically connected; The end of described transmission FIFO is electrically connected by sending logical and transmission driving; The end of described reception FIFO is by receive logic and receive the driving electrical connection; It is characterized in that, also comprise:
The frame head analysis logic is used to analyze long-range page request message and the long-range frame head of getting page response message got; Described frame head analysis logic one end is electrically connected with the pci interface controller by local bus, and the other end is electrically connected with reception FIFO;
The RDMA operation logic is used for control to the internal memory read-write operation, so that the page is directly read or writes from internal memory; Described RDMA (" RDMA " is that long-distance inner is directly visited Remote Direct Memory Access english abbreviation) operation logic is electrically connected with the pci interface controller by local bus, and is electrically connected with described frame head analysis logic and actual situation address translation logic respectively;
Actual situation address translation logic is used for the conversion of virtual address and physical address; Described actual situation address translation logic is electrically connected with the pci interface controller by local bus.
The present invention has following advantage:
1) in the present invention, by the memory area on the network interface unit being mapped to the virtual address space of consumer process, make that consumer process and network interface unit can two-way direct visits, can under the situation of the work at present that does not interrupt remote node CPU, realize getting page operations.
2) the present invention is with the most protocol overhead of the hardware handles on the network interface unit, reduced protocol processes expenses such as the reception message of consumer process and deblocking message, improved long-range efficient of getting page or leaf.
3) general DMA and RDMA operation all is directly to adopt physical address, and adopts virtual address in the present invention, for realizing long-rangely getting page or leaf and virtual shared storage provides very large facility.
Description of drawings
Fig. 1 is an exemplary virtual shared system;
Fig. 2 realizes long-range network interface unit of getting page method of the present invention;
Fig. 3 is a long-range method flow block diagram of getting page method of the present invention;
Fig. 4 realizes the long-range map addresses synoptic diagram of getting page method of the present invention.
Embodiment
Below in conjunction with the drawings and specific embodiments the present invention is described in further detail.
An exemplary virtual shared system as shown in Figure 1 comprises source node 10 and destination node 20, and they have network interface unit 30 and 30 ' respectively, and it is interconnected that network interface unit 30 and 30 ' is passed through switch 40.When concrete the application, switch 40 is realized by the fpga chip EP20K200EFC484-2X of Altera.
In the present invention, the long-range page operations of getting is mainly finished by the network interface unit on the node.Network interface unit 30 and 30 ' logical organization are as shown in Figure 2.In Fig. 2, network interface unit is connected with the pci bus of main frame with pci interface controller 301, frame head analysis logic 302, message are transmitted operation logic 303, RDMA operation logic 304, actual situation address translation logic 305, control register logic 306, reception FIFO307 and are sent FIFO308 and is connected by local bus, transmission logic 310 drives 312 with transmission and is connected with transmission FIFO308 in proper order, and receive logic 309 is connected with reception FIFO307 with reception driving 311 orders.Wherein, frame head analysis logic 302 is taked different actions according to the type of frame; Message is transmitted operation logic 303 and is responsible for common information receiving and transmitting operation; The 304 control memory read-write operations of RDMA operation logic; Actual situation address translation logic 305 is used for the conversion of virtual address and physical address, comprises TLB (Translation Lookaside Buffer) matching logic and an actual situation address translation table; The read-write operation of control register logic 306.When concrete the application, the logic function of network interface unit shown in Figure 2 is realized by the fpga chip EP1K100FC484-1 of Altera.
Fig. 3 is a long-range process flow diagram of getting page method of the present invention.As described below:
In step 101, for the storer that can allow on the direct accesses network interface card of consumer process, need in communication protocol, the memory area on the network interface unit (NIC) be mapped to the virtual address space of consumer process (User) by the method for map addresses.The map addresses is here finished by twice mapping as shown in Figure 4:
At first, the NIC register area is positioned at the PCI space, and on the PC based on the x86 architecture, the PCI space is consistent with the host memory physical address space, and the mapping in this stage does not need communication protocol to do any intervention;
Second, in the present invention, in order to make consumer process can directly visit this register area on the NIC, in communication protocol, the NIC register is mapped to again the virtual address space of consumer process in this part zone of physical address space, like this consumer process direct accesses network interface just.
In step 102, the consumer process of source node is directly visited its network interface unit, produces and sends the long-range page request message of getting to the network interface unit of destination node.Specifically comprise following substep:
Step 102a): the consumer process on the source node is inserted an available descriptor to the virtual address that page or leaf virtual address, data length (Ye size in other words), this page on destination node that will read return source node by communication pool;
Step 102b): write (descriptor is once write in transmission) on the register of network interface unit filling in the descriptor that finishes by communication thread;
Step 102c): the content that network interface unit is write according to consumer process is come framing, and puts a type code at frame head, is the long-range page request message of getting to show this frame, and this frame is sent to the network interface unit of destination node.As step 102a) as described in, this is long-range gets in the frame of page request message and also comprise: page or leaf virtual address territory, data length field, this page on destination node that read returns the virtual address territory of source node.This step is transmitted operation logic 303 by the message in the network interface unit shown in Figure 2 and is realized.
In step 103, the network interface unit of destination node receives the long-range page request message of getting from source node.Destination node is taken out frame head data and is analyzed from the reception fifo buffer of its network interface unit, determine the long-range page read operation of getting, and carries out the frame head conversion---be converted to the frame head that returns.This step is realized by the frame head analysis logic in the network interface unit shown in Figure 2.
In step 104, the network interface unit of destination node directly reads the required page or leaf in the destination node.Specifically comprise following substep:
Step 104a): network interface unit log on interface local bus;
Step 104b): the actual situation address translation is carried out in the network interface unit application, is physical address with virtual address translation.Here, the actual situation address translation is finished by the actual situation address translation logic on the network interface unit shown in Figure 2.This actual situation address translation logic comprises an actual situation address translation buffer (TLB, Translation LookasideBuffer), wherein stores the coupling list item by virtual address and physical address.When carrying out the actual situation address translation, search item in the tlb entry on network interface unit with the virtual address coupling, then directly return physical address corresponding if find, do not interrupt if find then to produce, have no progeny in the NIC driver reception by the node main frame and carry out the actual situation address translation according to interrupt type, the physical address that obtains is write network interface unit, and actual situation address translation logic receives after the data tlb entry upgraded returns physical address simultaneously.
Step 104c): network interface unit will be returned frame head and write network interface transmission fifo buffer;
Step 104d): the region of memory of the physical address appointment that TLB is converted to is read the transmission fifo buffer with dma mode.This step is finished by the RDMA operation logic on the network interface unit shown in Figure 2.
In step 105, the network interface unit of destination node is returned the long-range page response message of getting to the network interface unit of source node, comprises the page or leaf that has read in this message.And in the long-range frame head of getting page response message, a type code is set, be the long-range page response message of getting to show this frame.
In step 106, the network interface unit receiving remote of source node is got a page response message.Source node takes out frame head data and analyzes from the reception fifo buffer of its network interface unit, according to type code, determine the long-range return of reading of getting page or leaf.
In step 107, the network interface unit of source node is directly write described page or leaf the region of memory of source node.Specifically comprise following substep:
Step 107a): the actual situation address translation is carried out in the network interface unit application, the concrete operations of this step and step 104b) basic identical; Because network interface unit can only be carried out work according to the pci bus address, and the PCI space is consistent with physical address space, as long as so know the pairing physical address of the process space that needs visit, network interface unit just can be by the corresponding region of memory of this address visit;
Step 107b): the network interface unit application obtains local bus;
Step 107c): the data that will receive in the fifo buffer are delivered to the specified region of memory of physical address that TLB is converted to.In order to realize this point, the empty page locking (pin) of source node consumer process is in its physical memory, and to prevent network interface when visiting this zone, operating system exchanges away the corresponding page.And, as shown in Figure 4, for the page that relates to RDMA (being the RDMA zone among Fig. 4), set up the mapping of consumer process virtual address and physical address, again this RDMA zone is mapped to the PCI space in the zone of physical address space, so, network interface is by this physical address calling party space reliably just;
Step 107d): the complement mark set of network interface unit corresponding descriptor in the source node host memory.
Claims (10)
1, a kind of long-range page method of getting that is used for shared virtual memory system, described shared virtual memory system comprise with network interface unit and interconnected source node and the destination node of switch; Comprise step:
1) memory area on the network interface unit is mapped to the virtual address space of consumer process;
2) consumer process of source node is directly visited its network interface unit, produces and sends the long-range page request message of getting to the network interface unit of destination node;
3) network interface unit of destination node receives the long-range page request message of getting from source node;
4) network interface unit of destination node directly reads the required page or leaf in the destination node;
5) network interface unit of destination node is returned the long-range page response message of getting to the network interface unit of source node, comprises the page or leaf that reads in the step 4) in this message;
6) network interface unit of source node receives the described long-range page response message of getting;
7) network interface unit of source node is directly write described page or leaf the region of memory of source node.
2, the long-range page method of getting that is used for shared virtual memory system according to claim 1, it is characterized in that, the described long-range page request message of getting comprises a message type indicator, and the network interface of described destination node is stuck in and receives this and long-rangely the type sign is judged after getting page request message; The described long-range page response message of getting comprises a message type indicator, and the network interface of described source node is stuck in and receives this and long-rangely the type sign is judged after getting page response message.
3, the long-range page method of getting that is used for shared virtual memory system according to claim 1, it is characterized in that, described long-range get to comprise in page request message to read the virtual address that virtual address, data length and this page on destination node of page or leaf returns source node.
4, the long-range page method of getting that is used for shared virtual memory system according to claim 1 is characterized in that, network interface unit reads required when page or leaf in step 4), also will carry out the actual situation address translation, so that be physical address with virtual address translation; The network interface unit of the destination node page or leaf that the physical address that obtains is pointed reads this network interface unit with dma mode then.
5, the long-range page method of getting that is used for shared virtual memory system according to claim 1 is characterized in that, described long-range getting also comprises the virtual address that the page or leaf that will read returns source node in page response message.
6, the long-range page method of getting that is used for shared virtual memory system according to claim 5, it is characterized in that, network interface unit also will be carried out the actual situation address translation when described region of memory is write page in step 7), so that be physical address with virtual address translation; Network interface unit is write page or leaf according to this physical address to described region of memory.
7, the long-range page method of getting that is used for shared virtual memory system according to claim 6 is characterized in that, in the network interface unit of source node when described region of memory is write page, with the empty page locking of source node consumer process in its physical memory.
8, according to a claim 4 or the 6 described long-range page methods of getting that are used for shared virtual memory system, it is characterized in that, described actual situation address translation is finished by the actual situation address translation buffer that provides on network interface unit, and described actual situation address translation buffer provides the coupling list item of virtual address and physical address.
9, the long-range page method of getting that is used for shared virtual memory system according to claim 8, it is characterized in that, when carrying out the actual situation address translation, if do not find the physical address that mates with virtual address at the actual situation address translation buffer, then produce and interrupt, carry out the actual situation address translation by the NIC driver of node main frame; NIC driver writes network interface unit with the physical address that obtains after finishing the actual situation address translation, and upgrades the coupling list item in the actual situation address translation buffer, returns physical address simultaneously.
10, a kind of network interface unit of implementing the described method of claim 1 comprises that the pci interface controller transmits operation logic, control register logic, transmission FIFO and reception FIFO with message respectively by local bus and is electrically connected; The end of described transmission FIFO is electrically connected by sending logical and transmission driving; The end of described reception FIFO is by receive logic and receive the driving electrical connection; It is characterized in that, also comprise:
The frame head analysis logic is used to analyze long-range page request message and the long-range frame head of getting page response message got; Described frame head analysis logic one end is electrically connected with the pci interface controller by local bus, and the other end is electrically connected with reception FIFO;
The RDMA operation logic is used for control to the internal memory read-write operation, so that the page is directly read or writes from internal memory; Described RDMA operation logic is electrically connected with the pci interface controller by local bus, and is electrically connected with described frame head analysis logic and actual situation address translation logic respectively;
Actual situation address translation logic is used for the conversion of virtual address and physical address; Described actual situation address translation logic is electrically connected with the pci interface controller by local bus.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CNB2004100427427A CN100349142C (en) | 2004-05-25 | 2004-05-25 | Remote page access method for use in shared virtual memory system and network interface card |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CNB2004100427427A CN100349142C (en) | 2004-05-25 | 2004-05-25 | Remote page access method for use in shared virtual memory system and network interface card |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN1705269A CN1705269A (en) | 2005-12-07 |
| CN100349142C true CN100349142C (en) | 2007-11-14 |
Family
ID=35577718
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CNB2004100427427A Expired - Lifetime CN100349142C (en) | 2004-05-25 | 2004-05-25 | Remote page access method for use in shared virtual memory system and network interface card |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN100349142C (en) |
Families Citing this family (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102651690B (en) * | 2011-02-28 | 2015-11-25 | 国际商业机器公司 | To the method and apparatus that shared drive controls |
| CN108845877B (en) * | 2013-05-17 | 2021-09-17 | 华为技术有限公司 | Method, device and system for managing memory |
| CN104202391B (en) * | 2014-08-28 | 2018-09-25 | 浪潮(北京)电子信息产业有限公司 | RDMA communication means between the no tight coupling system of shared system address space |
| CN104881330B (en) * | 2015-05-22 | 2018-08-21 | 大唐移动通信设备有限公司 | A kind of method and apparatus of multi-process shared data |
| WO2020118650A1 (en) * | 2018-12-14 | 2020-06-18 | 华为技术有限公司 | Method for quickly sending write data preparation completion message, and device, and system for quickly sending write data preparation completion message |
| CN112395220B (en) * | 2020-11-18 | 2023-02-28 | 海光信息技术股份有限公司 | Processing method, device and system of shared storage controller and storage controller |
| CN114285676B (en) * | 2021-11-24 | 2023-10-20 | 中科驭数（北京）科技有限公司 | Intelligent network card, network storage method and medium of intelligent network card |
| CN113868155B (en) * | 2021-11-30 | 2022-03-08 | 苏州浪潮智能科技有限公司 | Memory space expansion method and device, electronic equipment and storage medium |
| CN117908988B (en) * | 2024-01-18 | 2025-03-04 | 中南大学 | Method and system for loading and running program content from server memory on demand |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US5592625A (en) * | 1992-03-27 | 1997-01-07 | Panasonic Technologies, Inc. | Apparatus for providing shared virtual memory among interconnected computer nodes with minimal processor involvement |
| CN1421789A (en) * | 2001-11-27 | 2003-06-04 | 华为技术有限公司 | Realizing method of double-channel shared memory |
| CN1451114A (en) * | 2000-01-05 | 2003-10-22 | 英特尔公司 | Shared between processing threads |
| CN1474296A (en) * | 2003-04-11 | 2004-02-11 | 大唐移动通信设备有限公司 | Data interacting method and device between multiple processors based on shared storage |
-
2004
- 2004-05-25 CN CNB2004100427427A patent/CN100349142C/en not_active Expired - Lifetime
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US5592625A (en) * | 1992-03-27 | 1997-01-07 | Panasonic Technologies, Inc. | Apparatus for providing shared virtual memory among interconnected computer nodes with minimal processor involvement |
| CN1451114A (en) * | 2000-01-05 | 2003-10-22 | 英特尔公司 | Shared between processing threads |
| CN1421789A (en) * | 2001-11-27 | 2003-06-04 | 华为技术有限公司 | Realizing method of double-channel shared memory |
| CN1474296A (en) * | 2003-04-11 | 2004-02-11 | 大唐移动通信设备有限公司 | Data interacting method and device between multiple processors based on shared storage |
Non-Patent Citations (1)
| Title |
|---|
| VIA(Virtual Interface Architecture)上的软件DSM系统实现和性能 史岗，尹宏达，胡明昌，胡伟武.计算机学报，第卷号：26卷第期号：12期 2003 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN1705269A (en) | 2005-12-07 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN102498478B (en) | Iommu using two-level address translation for i/o and computation offload devices on a peripheral interconnect | |
| US6611883B1 (en) | Method and apparatus for implementing PCI DMA speculative prefetching in a message passing queue oriented bus system | |
| US7702826B2 (en) | Method and apparatus by utilizing platform support for direct memory access remapping by remote DMA (“RDMA”)-capable devices | |
| CN1307570C (en) | Method and apparatus forcontrolling flow of data between data processing systems via a memory | |
| US6385705B1 (en) | Circuit and method for maintaining order of memory access requests initiated by devices in a multiprocessor system | |
| US9274940B2 (en) | Method and apparatus for allocating memory space with write-combine attribute | |
| US8484307B2 (en) | Host fabric interface (HFI) to perform global shared memory (GSM) operations | |
| TW552515B (en) | Input/output (I/O) address translation in a bridge proximate to a local I/O bus | |
| US7707383B2 (en) | Address translation performance in virtualized environments | |
| CN102326156B (en) | Opportunistic improvement of MMIO request handling based on target reporting of space requirements | |
| JP5681782B2 (en) | On-die system fabric block control | |
| US20050144422A1 (en) | Virtual to physical address translation | |
| US8631184B2 (en) | Interconnection method and device, for example for systems-on-chip | |
| CN109308283B (en) | SoC (system on chip) system and peripheral bus switching method thereof | |
| US20240264768A1 (en) | Request Processing Method, Apparatus, and System | |
| WO2015180598A1 (en) | Method, apparatus and system for processing access information of storage device | |
| CN100349142C (en) | Remote page access method for use in shared virtual memory system and network interface card | |
| US7657724B1 (en) | Addressing device resources in variable page size environments | |
| CN107924343B (en) | Low latency inter-node communication | |
| CN101290605A (en) | Data processing method, communication system and related equipment | |
| US20090199200A1 (en) | Mechanisms to Order Global Shared Memory Operations | |
| CN113722247A (en) | Physical memory protection unit, physical memory authority control method and processor | |
| CN114201268A (en) | A data processing method, apparatus, device and readable storage medium | |
| WO2022032990A1 (en) | Command information transmission method, system, and apparatus, and readable storage medium | |
| CN116186793B (en) | RISC-V based security chip architecture and working 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 | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant | ||
| EE01 | Entry into force of recordation of patent licensing contract |
Assignee: Beijing Loongson Technology Service Center Co.,Ltd. Assignor: Institute of Computing Technology, Chinese Academy of Sciences Contract fulfillment period: 2009.12.16 to 2028.12.31 Contract record no.: 2010990000062 Denomination of invention: Remote page access method for use in shared virtual memory system and network interface card Granted publication date: 20071114 License type: exclusive license Record date: 20100128 |
|
| LIC | Patent licence contract for exploitation submitted for record |
Free format text: EXCLUSIVE LICENSE; TIME LIMIT OF IMPLEMENTING CONTACT: 2009.12.16 TO 2028.12.31; CHANGE OF CONTRACT Name of requester: BEIJING LOONGSON TECHNOLOGY SERVICE CENTER CO., LT Effective date: 20100128 |
|
| EC01 | Cancellation of recordation of patent licensing contract |
Assignee: LOONGSON TECHNOLOGY Corp.,Ltd. Assignor: Institute of Computing Technology, Chinese Academy of Sciences Contract record no.: 2010990000062 Date of cancellation: 20141231 |
|
| EM01 | Change of recordation of patent licensing contract |
Change date: 20141231 Contract record no.: 2010990000062 Assignee after: LOONGSON TECHNOLOGY Corp.,Ltd. Assignee before: Beijing Loongson Technology Service Center Co.,Ltd. |
|
| LICC | Enforcement, change and cancellation of record of contracts on the licence for exploitation of a patent or utility model | ||
| EE01 | Entry into force of recordation of patent licensing contract |
Application publication date: 20051207 Assignee: LOONGSON TECHNOLOGY Corp.,Ltd. Assignor: Institute of Computing Technology, Chinese Academy of Sciences Contract record no.: 2015990000066 Denomination of invention: Remote page access method for use in shared virtual memory system and network interface card Granted publication date: 20071114 License type: Common License Record date: 20150211 |
|
| TR01 | Transfer of patent right |
Effective date of registration: 20200819 Address after: 100095, Beijing, Zhongguancun Haidian District environmental science and technology demonstration park, Liuzhou Industrial Park, No. 2 building Patentee after: LOONGSON TECHNOLOGY Corp.,Ltd. Address before: 100080 Haidian District, Zhongguancun Academy of Sciences, South Road, No. 6, No. Patentee before: Institute of Computing Technology, Chinese Academy of Sciences |
|
| TR01 | Transfer of patent right | ||
| EC01 | Cancellation of recordation of patent licensing contract |
Assignee: LOONGSON TECHNOLOGY Corp.,Ltd. Assignor: Institute of Computing Technology, Chinese Academy of Sciences Contract record no.: 2015990000066 Date of cancellation: 20200928 |
|
| EC01 | Cancellation of recordation of patent licensing contract | ||
| CP01 | Change in the name or title of a patent holder |
Address after: 100095 Building 2, Longxin Industrial Park, Zhongguancun environmental protection technology demonstration park, Haidian District, Beijing Patentee after: Loongson Zhongke Technology Co.,Ltd. Address before: 100095 Building 2, Longxin Industrial Park, Zhongguancun environmental protection technology demonstration park, Haidian District, Beijing Patentee before: LOONGSON TECHNOLOGY Corp.,Ltd. |
|
| CP01 | Change in the name or title of a patent holder | ||
| CX01 | Expiry of patent term |
Granted publication date: 20071114 |
|
| CX01 | Expiry of patent term |