CN108959165A - A kind of management system of GPU whole machine cabinet cluster - Google Patents
A kind of management system of GPU whole machine cabinet cluster Download PDFInfo
- Publication number
- CN108959165A CN108959165A CN201810690618.3A CN201810690618A CN108959165A CN 108959165 A CN108959165 A CN 108959165A CN 201810690618 A CN201810690618 A CN 201810690618A CN 108959165 A CN108959165 A CN 108959165A
- Authority
- CN
- China
- Prior art keywords
- gpu
- node
- management
- chip
- whole machine
- 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.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F15/00—Digital computers in general; Data processing equipment in general
- G06F15/16—Combinations of two or more digital computers each having at least an arithmetic unit, a program unit and a register, e.g. for a simultaneous processing of several programs
- G06F15/161—Computing infrastructure, e.g. computer clusters, blade chassis or hardware partitioning
Abstract
The invention discloses a kind of management systems of GPU whole machine cabinet cluster, comprising: is provided with the GPU node of managed devices and GPU card;The management node being connected with each GPU node, for being managed to corresponding managed devices in each GPU node and GPU card;The present invention passes through the setting for the management node being managed to corresponding managed devices and GPU card in each GPU node, originally the management module in each GPU node is independent from GPU node, individually designed management module is not needed on each GPU node, the management interface connecting with management node need to only be reserved, development cost is saved, the wasting of resources is avoided;And the compatibility of product can be enhanced by the Uniting of the management port of GPU node.
Description
Technical field
The present invention relates to power electronics field, in particular to a kind of management system of GPU whole machine cabinet cluster.
Background technique
With the rapid development of AI technology, it is surging that unprecedented research and development also occurs in the whole machine cabinet framework of high operational performance.And
The GPU card of the high calculated performance of (tall and handsome to reach) publication of NVIDIA at present, occupies leading status in calculated performance.It will be multiple
GPU (Graphics Processing Unit, graphics processor) card carries out parallel connection and forms a GPU node calculating collection
Group, collocation calculation server becomes the computing system that calculated performance is continuously improved in industry, and whole machine cabinet is with its high density convenience
Deployment way as entire computing system.
One whole machine cabinet can carry multiple GPU cards, carry out multi-layer to data link by the extended chip of PCIE
Extension, as soon as every increase layer extended chip, the increase of data path at double, the increase of the GPU card of carrying also at double.To these
GPU card and extended chip need a management module, carry out the collection of information and the monitoring of working condition.
In the prior art, GPU whole machine cabinet includes calculate node and GPU node, and often management module is arranged in GPU node
Interior, each GPU node devises the module for management, this to multiple GPU combination of nodes at large-scale cluster when, manage mould
The redundancy of block causes the waste of resource.Therefore, how large-scale GPU whole machine cabinet group system, avoid GPU node administration
The redundancy of module saves development cost, avoids the wasting of resources, is urgent problem now.
Summary of the invention
The object of the present invention is to provide a kind of management systems of GPU whole machine cabinet cluster, will manage mould using management node
Block is independent from GPU node, avoids the wasting of resources, enhances the compatibility of product.
In order to solve the above technical problems, the present invention provides a kind of management system of GPU whole machine cabinet cluster, comprising:
It is provided with the GPU node of managed devices and GPU card;
The management node being connected with each GPU node, for corresponding by pipe in each GPU node
Reason equipment and GPU card are managed.
Optionally, the management node, comprising:
Chip is controlled, for obtaining the information and working condition of corresponding GPU card in each GPU node;
First extended chip, for obtaining the information of corresponding managed devices in each GPU node;
Processor, it is each in each GPU node for obtaining by the control chip and first extended chip
Information, the information of working condition and managed devices of self-corresponding GPU card, and to respectively being corresponded in each GPU node
Managed devices are controlled.
Optionally, the processor by PCIE2.0 bus respectively with the control chip and first extended chip
It is connected.
Optionally, the management node, further includes:
The SATA interface being connected with the processor is that the processor installs operating system for connecting storage device.
Optionally, the management node, further includes:
The network interface being connected with the processor is transmitted for realizing the processor by the data that network carries out.
Optionally, the management node, further includes:
The USB interface being connected with the control chip, it is each in each GPU node to show for connecting display
The information and working condition of self-corresponding GPU card.
Optionally, the management node, further includes:
The RJ45 interface being connected with the control chip, for realizing the On-line Control of the control chip.
Optionally, the GPU node, comprising:
Second extended chip, for acquiring the information of each managed devices in the GPU node;
Third extended chip, for acquiring the information and working condition of each GPU card in the GPU node.
Optionally, first extended chip and the second extended chip in each GPU node are total by PCIE2.0
Line connects, and the second extended chip in the GPU node is all made of with each managed devices in the GPU node
The connection of PCIE2.0 bus.
Optionally, the control chip is connect with the third extended chip in each GPU node by I2C bus,
Third extended chip in the GPU node is all made of I2C bus with the GPU card in the GPU node and connect.
The management system of a kind of GPU whole machine cabinet cluster provided by the present invention, comprising: be provided with managed devices and GPU
The GPU node of card;The management node being connected with each GPU node, for being set to corresponding be managed in each GPU node
Standby and GPU card is managed;
As it can be seen that the present invention passes through the pipe that is managed to corresponding managed devices and GPU card in each GPU node
The setting for managing node, the management module in each GPU node originally is independent from GPU node, on each GPU node not
Individually designed management module is needed, only the management interface connecting with management node need to be reserved, save development cost, avoid money
Source waste;And the compatibility of product can be enhanced by the Uniting of the management port of GPU node.
Detailed description of the invention
In order to more clearly explain the embodiment of the invention or the technical proposal in the existing technology, to embodiment or will show below
There is attached drawing needed in technical description to be briefly described, it should be apparent that, the accompanying drawings in the following description is only this
The embodiment of invention for those of ordinary skill in the art without creative efforts, can also basis
The attached drawing of offer obtains other attached drawings.
Fig. 1 is a kind of structure chart of the management system of GPU whole machine cabinet cluster provided by the embodiment of the present invention;
Fig. 2 is the knot of management node in the management system of another kind GPU whole machine cabinet cluster provided by the embodiment of the present invention
Structure schematic diagram;
Fig. 3 is the structure of GPU node in the management system of another kind GPU whole machine cabinet cluster provided by the embodiment of the present invention
Schematic diagram.
Specific embodiment
In order to make the object, technical scheme and advantages of the embodiment of the invention clearer, below in conjunction with the embodiment of the present invention
In attached drawing, technical scheme in the embodiment of the invention is clearly and completely described, it is clear that described embodiment is
A part of the embodiment of the present invention, instead of all the embodiments.Based on the embodiments of the present invention, those of ordinary skill in the art
Every other embodiment obtained without making creative work, shall fall within the protection scope of the present invention.
Referring to FIG. 1, Fig. 1 is a kind of structure of the management system of GPU whole machine cabinet cluster provided by the embodiment of the present invention
Figure.The system may include:
It is provided with the GPU node 10 of managed devices and GPU card;
The management node 20 being connected with each GPU node 10, for being managed to corresponding in each GPU node 10
Equipment and GPU card are managed.
It is understood that the purpose of the present embodiment each GPU can will save originally for by the setting of management node 20
Point 10 in management module it is independent, management node 20 can the multiple GPU nodes 10 of carry, saved resource and cost;And
The reserved management interface connecting with management node 20 of GPU node 10 can use standard interface, be standardized definition, increase
The strong compatibility of product.
Wherein, pipe is carried out to corresponding managed devices in each GPU node 10 and GPU card for management node 20
The concrete mode of reason can such as use and existing skill by designer according to practical scene and user demand self-setting
The similar mode correspondence of management module in art in each GPU node 10 is configured, as shown in Fig. 2, management node 20 can be with
Pass through control chip 21 and processor 23 (CPU) realization and the control chip in each GPU node 10 and processing in the prior art
The way to manage of device.As long as management node 20 can to corresponding managed devices in each GPU node 10 and GPU card into
Row management, the present embodiment do not do any restrictions to this.
It should be noted that for the specific structure of management node 20, it can be by designer according to practical scene and use
Family demand self-setting, as shown in Fig. 2, management node 20 may include: control chip 21, for obtaining in each GPU node
The information and working condition of corresponding GPU card;First extended chip 22, it is respectively right in each GPU node 10 for obtaining
The information for the managed devices answered;Processor 23, for obtaining each GPU node by control chip and the first extended chip
In corresponding GPU card information, the information of working condition and managed devices, and to respectively being corresponded in each GPU node
Managed devices are controlled.That is, processor 23 can pass through the company with control chip 21 and the first extended chip 22
It connects, obtains the information of the information, working condition and managed devices of corresponding GPU card in each GPU node, and pass through the
One extended chip 22 is controlled to respectively corresponding to managed devices in each GPU node.For processor 23 and control chip
21 and first concrete mode that connects of extended chip 22, can be by designer's self-setting, processor 21 as shown in Figure 2 can be with
It is connected respectively with control chip and the first extended chip by PCIE2.0x1 bus, i.e., processor 21 can pass through PCIE link
To identify and distribute PCIE device resource;It can also be by other buses of such as PCIE3.0 bus, as long as processor may be implemented
Above-mentioned function, the present embodiment do not do any restrictions to this.
Corresponding, respective bus can be utilized respectively by management interface and connect by controlling chip 21 and the first extended chip 22
It is connected to the GPU card and managed devices of each GPU node 10, i.e., control chip 21 can be connected to each by management interface
The GPU card of GPU node 10, the first extended chip 22 can be connected to each being managed for GPU node 10 by management interface and set
It is standby.For control chip 21 and the corresponding bus of the first extended chip 22 selection, can by designer's self-setting,
As shown in Fig. 2, control chip 21 can obtain corresponding managed devices in each GPU node 10 by I2C bus
Information, the first extended chip 22 can obtain corresponding be managed in each GPU node 10 by PCIE2.0x1 bus and set
Standby information, i.e. the first extended chip 22 can be specially the PCIE extended chip for carrying out the extension of PCIE link.The present embodiment pair
This does not do any restrictions.
Optionally, as shown in Fig. 2, management node 20 can also include: the SATA interface being connected with processor 23, for connecting
Connecing storage device is that processor installs operating system;The network interface being connected with processor 23 passes through for realizing processor 23
The data transmission that network carries out;The USB interface being connected with control chip 21, for connecting display, to show each GPU node
The information and working condition of corresponding GPU card in 10;The RJ45 interface being connected with control chip 21, for realizing control core
The On-line Control of piece 21, the present embodiment do not do any restrictions to this.
Specifically, the specific device of control chip 21, the first extended chip 22 and processor 23 is selected, it can be by setting
Meter personnel are according to practical scene and user demand self-setting, as that can be carried out according to the quantity for the GPU node 10 to be connected
The selection of chip, when control chip 21 connects GPU node 10 with PCIE bus by I2C bus respectively with the first extended chip 22
When, to connect 12 GPU nodes 10, then to control chip 21 will support to be greater than 12 groups of I2C buses, and port number is greater than 13
A, control chip 21 can be AST2500, and PEX8625 can be used in the first extended chip 22, and control chip 21 is served only for pipe
Reason, does not need excessively high performance, and the AVOTON CPU C2350 of SOC can be used, and the Minisas of standard can be selected in connectivity port
The highly dense mouth of HD.
It should be noted that for the specific structure of GPU node 10, it can be by designer according to practical scene and user
Demand self-setting, as shown in figure 3, GPU node 10 may include: the second extended chip 11, for acquiring in GPU node 10
The information of each managed devices;Third extended chip 12, for acquiring the information and work of each GPU card in GPU node 10
Make state.That is, the information of each managed devices in GPU node 10 can be transferred to pipe by the second extended chip 11
The first extended chip 22 of node 20 is managed, and the control instruction of processor 23 is transferred to corresponding managed devices;Third expands
The information of each GPU card in GPU node 10 and working condition can be transferred to the control core of management node 20 by exhibition chip 12
Piece 21.
The concrete mode and third extended chip 12 being connect for the second extended chip 11 with managed devices and GPU card
Concrete mode, can be by designer's self-setting, such as can be according to the company of the second extended chip 11 and the first extended chip 22
Connecing mode and third extended chip 12, the connection type of chip 21 is corresponding is configured with control, as shown in figure 3, extending when second
When chip 11 is connect with the first extended chip 22 by PCIE bus, the second extended chip 11 can be by PCIE bus and by pipe
Equipment connection is managed, i.e. the second extended chip 11 and the first extended chip 22 are PCIE extended chip;When third extended chip 12
When being connect with control chip 21 by I2C bus, third extended chip 12 can be connect by I2C bus with GPU card, i.e. third
Extended chip 12 is I2C extended chip.As long as above-mentioned function may be implemented in the second extended chip 11 and third extended chip 12, this
Embodiment does not do any restrictions to this.
Specifically, the specific device for the second extended chip 11 and third extended chip 12 selects, it can be by designer
Member such as can will respectively be connected according to practical scene and user demand self-setting with the second extended chip 11 and third extended chip 12
The selection that the quantity of the managed devices and GPU card that connect carries out chip is managed and sets by taking the GPU node 10 of 8GPU card as an example
Standby is 3 PEX9797, and the second extended chip 11 (PCIE extended chip) can select PEX8604, support the end of 4 PCIE2.0
Mouthful, PEX8608 can also be selected, the control of 7 managed devices can be carried out;GPU card is 8,12 (I2C of third extended chip
Extended chip) PCA9548APWR that 8 tunnels can be selected, more GPU cards are managed if necessary, can be connected in parallel multiple I2C and be expanded
Open up chip.The present embodiment does not do any restrictions to this.
It is understood that as shown in Figures 2 and 3, GPU node 10 can be utilized by management interface includes I2C bus
It is connected to the management interface of management node 20 with the cable of PCIE2.0x1 bus, realizes the interconnection of hardware;After hardware interconnection, GPU
After node 10 powers on, management node 20 is powered on, it can be seen that each GPU node 10 connected is managed under processor interface
Equipment, under control chip interface, it can be seen that the information and working condition of the GPU card of each GPU node 10 of connection.
In the present embodiment, the embodiment of the present invention passes through to corresponding managed devices and GPU in each GPU node 10
The setting for blocking the management node 20 being managed, the management module in each GPU node 10 originally is independent from GPU node 10
Out, individually designed management module is not needed on each GPU node 10, only need to reserves the management connecting with management node 20 and connects
Mouthful, development cost is saved, the wasting of resources is avoided;And it can be increased by the Uniting of the management port of GPU node 10
The compatibility of strong product.
The management system of GPU whole machine cabinet cluster provided by the present invention is described in detail above.It is used herein
A specific example illustrates the principle and implementation of the invention, and the above embodiments are only used to help understand
Method and its core concept of the invention.It should be pointed out that for those skilled in the art, not departing from this
, can be with several improvements and modifications are made to the present invention under the premise of inventive principle, these improvement and modification also fall into the present invention
In scope of protection of the claims.
It should also be noted that, in the present specification, such as first and second etc relational terms are used merely to one
A entity or operation are perhaps operated with another entity and are distinguished without necessarily requiring or implying these entities or behaviour
There are any actual relationship or orders between work.Moreover, the terms "include", "comprise" or its any variant are intended to
Cover non-exclusive inclusion, so that the process, method, article or equipment for including a series of element not only includes that
A little elements, but also other elements including being not explicitly listed either further include for this process, method, article or setting
Standby intrinsic element.In the absence of more restrictions, the element limited by sentence "including a ...", it is not excluded that
There is also other identical elements in the process, method, article or apparatus that includes the element.
Claims (10)
1. a kind of management system of GPU whole machine cabinet cluster characterized by comprising
It is provided with the GPU node of managed devices and GPU card;
The management node being connected with each GPU node, for being set to corresponding be managed in each GPU node
Standby and GPU card is managed.
2. the management system of GPU whole machine cabinet cluster according to claim 1, which is characterized in that the management node, packet
It includes:
Chip is controlled, for obtaining the information and working condition of corresponding GPU card in each GPU node;
First extended chip, for obtaining the information of corresponding managed devices in each GPU node;
Processor, it is respectively right in each GPU node for obtaining by the control chip and first extended chip
Information, the information of working condition and managed devices for the GPU card answered, and to respectively corresponding by pipe in each GPU node
Reason equipment is controlled.
3. the management system of GPU whole machine cabinet cluster according to claim 2, which is characterized in that it is characterized in that, the place
Reason device is connected with the control chip and first extended chip respectively by PCIE2.0 bus.
4. the management system of GPU whole machine cabinet cluster according to claim 2, which is characterized in that the management node is also wrapped
It includes:
The SATA interface being connected with the processor is that the processor installs operating system for connecting storage device.
5. the management system of GPU whole machine cabinet cluster according to claim 2, which is characterized in that the management node is also wrapped
It includes:
The network interface being connected with the processor realizes what the processor was carried out by network for connecting external network
Data transmission.
6. the management system of GPU whole machine cabinet cluster according to claim 2, which is characterized in that the management node is also wrapped
It includes:
The USB interface being connected with the control chip is respectively right in each GPU node to show for connecting display
The information and working condition for the GPU card answered.
7. the management system of GPU whole machine cabinet cluster according to claim 2, which is characterized in that the management node is also wrapped
It includes:
The RJ45 interface being connected with the control chip, for realizing the On-line Control of the control chip.
8. according to the management system of the described in any item GPU whole machine cabinet clusters of claim 2 to 7, which is characterized in that the GPU
Node, comprising:
Second extended chip, for acquiring the information of each managed devices in the GPU node;
Third extended chip, for acquiring the information and working condition of each GPU card in the GPU node.
9. the management system of GPU whole machine cabinet cluster according to claim 8, which is characterized in that first extended chip
It is connect with the second extended chip in each GPU node by PCIE2.0 bus, the second extension in the GPU node
Chip is all made of PCIE2.0 bus with each managed devices in the GPU node and connect.
10. the management system of GPU whole machine cabinet cluster according to claim 8, which is characterized in that the control chip and every
Third extended chip in a GPU node is connected by I2C bus, third extended chip and institute in the GPU node
It states the GPU card in GPU node and is all made of the connection of I2C bus.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810690618.3A CN108959165A (en) | 2018-06-28 | 2018-06-28 | A kind of management system of GPU whole machine cabinet cluster |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810690618.3A CN108959165A (en) | 2018-06-28 | 2018-06-28 | A kind of management system of GPU whole machine cabinet cluster |
Publications (1)
Publication Number | Publication Date |
---|---|
CN108959165A true CN108959165A (en) | 2018-12-07 |
Family
ID=64487914
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201810690618.3A Pending CN108959165A (en) | 2018-06-28 | 2018-06-28 | A kind of management system of GPU whole machine cabinet cluster |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN108959165A (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109739560A (en) * | 2018-12-24 | 2019-05-10 | 郑州云海信息技术有限公司 | A kind of GPU card cluster configuration control system and method |
Citations (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103605404A (en) * | 2013-11-22 | 2014-02-26 | 曙光信息产业(北京)有限公司 | System with external expansion GPU (graphics processing unit) cards |
CN103780660A (en) * | 2012-10-25 | 2014-05-07 | 中国石油化工股份有限公司 | GPU cluster monitoring system and method for issuing monitoring alarm |
KR101442643B1 (en) * | 2013-04-30 | 2014-09-19 | 전자부품연구원 | The Cooperation System and the Method between CPU and GPU |
CN107239346A (en) * | 2017-06-09 | 2017-10-10 | 郑州云海信息技术有限公司 | A kind of whole machine cabinet computing resource tank node and computing resource pond framework |
CN107659437A (en) * | 2017-09-15 | 2018-02-02 | 郑州云海信息技术有限公司 | A kind of whole machine cabinet computing resource Pooled resources automatic recognition system and method |
CN107682222A (en) * | 2017-09-25 | 2018-02-09 | 郑州云海信息技术有限公司 | A kind of information monitoring method, apparatus and system |
CN107831883A (en) * | 2017-11-24 | 2018-03-23 | 郑州云海信息技术有限公司 | A kind of GPU server power supplies abnormal protection system and method |
-
2018
- 2018-06-28 CN CN201810690618.3A patent/CN108959165A/en active Pending
Patent Citations (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103780660A (en) * | 2012-10-25 | 2014-05-07 | 中国石油化工股份有限公司 | GPU cluster monitoring system and method for issuing monitoring alarm |
KR101442643B1 (en) * | 2013-04-30 | 2014-09-19 | 전자부품연구원 | The Cooperation System and the Method between CPU and GPU |
CN103605404A (en) * | 2013-11-22 | 2014-02-26 | 曙光信息产业(北京)有限公司 | System with external expansion GPU (graphics processing unit) cards |
CN107239346A (en) * | 2017-06-09 | 2017-10-10 | 郑州云海信息技术有限公司 | A kind of whole machine cabinet computing resource tank node and computing resource pond framework |
CN107659437A (en) * | 2017-09-15 | 2018-02-02 | 郑州云海信息技术有限公司 | A kind of whole machine cabinet computing resource Pooled resources automatic recognition system and method |
CN107682222A (en) * | 2017-09-25 | 2018-02-09 | 郑州云海信息技术有限公司 | A kind of information monitoring method, apparatus and system |
CN107831883A (en) * | 2017-11-24 | 2018-03-23 | 郑州云海信息技术有限公司 | A kind of GPU server power supplies abnormal protection system and method |
Non-Patent Citations (1)
Title |
---|
刘淼: "《嵌入式系统接口设计与Linux驱动程序开发》", 31 May 2006, 北京航空航天大学出版社 * |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109739560A (en) * | 2018-12-24 | 2019-05-10 | 郑州云海信息技术有限公司 | A kind of GPU card cluster configuration control system and method |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
EP2260611B1 (en) | Automated power topology discovery | |
CN103501242B (en) | Method for managing resource and multiple-node cluster device | |
US20090187668A1 (en) | Protocol Independent Server Replacement and Replication in a Storage Area Network | |
CN106155959A (en) | Data transmission method and data transmission system | |
CN102346707B (en) | Server system and operation method thereof | |
CN105095553A (en) | Topology display method and device | |
CN211628241U (en) | PCIE bus structure switching topology through software | |
CN102622279A (en) | Redundant control system and method and management controllers | |
CN105874442A (en) | Computer system and method for accessing endpoint device in the computer system | |
CN109783151A (en) | The method and apparatus of rule change | |
CN107682411A (en) | A kind of extensive SDN controllers cluster and network system | |
CN109933557A (en) | A kind of generation method and device of I2C topological diagram | |
CN105786732B (en) | Data access method and device | |
CN106776425B (en) | A kind of working frequency setting method, main control card, service card and communication equipment | |
CN103092735A (en) | Method for updating node states | |
CN108959165A (en) | A kind of management system of GPU whole machine cabinet cluster | |
CN110276515A (en) | Method for allocating tasks, device, terminal and storage medium | |
CN108306964A (en) | Server node information centralized displaying method, system, equipment and storage medium | |
CN104391546A (en) | Elastic configuration-variable cloud server system architecture | |
CN109039955A (en) | A kind of storage system based on optical-fibre channel | |
CN104516852B (en) | The circuit of I/O links divides multiplexing | |
CN109213566A (en) | Virtual machine migration method, device and equipment | |
CN106294062A (en) | A kind of method by computing function in server with management function separate design | |
CN103905224B (en) | The method and system of Internet resources concentrate tube reason | |
CN109739560A (en) | A kind of GPU card cluster configuration control system and method |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20181207 |
|
RJ01 | Rejection of invention patent application after publication |