CN109947581B - High concurrency data transfer method and data transfer method for switch - Google Patents
High concurrency data transfer method and data transfer method for switch Download PDFInfo
- Publication number
- CN109947581B CN109947581B CN201910249826.4A CN201910249826A CN109947581B CN 109947581 B CN109947581 B CN 109947581B CN 201910249826 A CN201910249826 A CN 201910249826A CN 109947581 B CN109947581 B CN 109947581B
- Authority
- CN
- China
- Prior art keywords
- data
- thread
- udp
- processed
- socket
- 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.)
- Active
Links
Images
Landscapes
- Data Exchanges In Wide-Area Networks (AREA)
Abstract
本发明公开了一种高并发的数据传送方法及用于交换机的数据传送方法，属于基于Linux内核的数据传送，要解决的技术问题为如何实现高并发数据的接收、处理和传送；该方法基于IO多路复用技术接收来自发送端的UDP数据，根据采集端的采集需求对UDP数据进行多线程处理，得到处理后数据，基于IO多路复用技术将处理后数据传送至采集端TCP端口。用于交换机的数据传送方法，包括通过高并发的数据传送方法，接收来自交换机的UDP数据，并对UDP数据进行多线程处理后，将处理后数据发送至采集端。
The invention discloses a high-concurrency data transmission method and a data transmission method for a switch, which belong to data transmission based on Linux kernel, and the technical problem to be solved is how to realize the reception, processing and transmission of high-concurrency data; the method is based on The IO multiplexing technology receives the UDP data from the sender, performs multi-thread processing on the UDP data according to the collection requirements of the collection end, and obtains the processed data. Based on the IO multiplexing technology, the processed data is transmitted to the collection end TCP port. The data transmission method used for the switch includes receiving UDP data from the switch through a highly concurrent data transmission method, and after multi-threaded processing of the UDP data, the processed data is sent to the acquisition terminal.
Description
Technical Field
The invention relates to the field of data transmission based on a Linux kernel, in particular to a high-concurrency data transmission method and a data transmission method for a switch.
Background
With the development of monitoring technology, traffic monitoring is more and more emphasized, the traditional scheme solves the problem of traffic monitoring at a switch or an application layer, the switch is not conscious in this respect for the requirement of counting results after data is calculated and processed, the data processing work needs to be handed to a server with stronger computing capacity for processing, and an application program on the server receives and processes the data of the switch and then needs to be sent to an acquisition end for filing. If the data collected by the switch is sent to the server and handed to the application program for processing, the following advantages exist: on one hand, the pressure of the switch can be reduced, on the other hand, the application of the flow statistics can have more flexibility, and the dimensionality of the statistics can be customized in a personalized way. But a problem is also extended: the number of data packets of the switch is very large, and if the data packets are directly sent to an application and processed, the data packets are lost, so that the statistical result is distorted.
In view of the above analysis, how to combine the switch application, the data transceiving application, the data processing application, and the acquisition application together by combining the IO multiplexing EPOLL technology with the multithread processing is a problem to be solved.
Disclosure of Invention
The technical task of the present invention is to provide a highly concurrent data transmission method and a data transmission method for a switch, so as to solve the problem of how to implement receiving, processing and transmitting highly concurrent data.
In a first aspect, the present invention provides a highly concurrent data transmission method, which receives UDP (User Datagram Protocol, chinese translation to User Datagram Protocol) data from a transmitting end based on an IO multiplexing technique, performs multithread processing on the UDP data according to an acquisition requirement of an acquisition end to obtain processed data, and transmits the processed data to a TCP port of the acquisition end based on the IO multiplexing technique.
Preferably, the highly concurrent data transfer method includes:
creating and initializing an IO multiplexing epoll instance, and monitoring a kernel event;
when a UDP data arrival signal is monitored, the UDP data is transferred to an application buffer area through a UDP data receiving thread;
when monitoring the acquisition notification, broadcasting condition variables of corresponding data types through a main thread, activating a work thread through the acquisition notification receiving thread, and performing multi-thread parallel processing on UDP data in corresponding application buffer areas through the work thread to obtain processed data;
and transmitting the processed data to the TCP port of the acquisition end through the processed data transmission thread.
epoll is poll improved by a Linux kernel for processing a large batch of file descriptors, is an enhanced version of a multiplex IO interface select/pol under Linux, and can remarkably improve the utilization rate of a system CPU (Central processing Unit) under the condition that a program is only slightly active in a large number of concurrent connections; when obtaining the event, it need not traverse the whole monitored descriptor set, only traverse the descriptor set asynchronously awakened by the kernel IO event and added into the Ready queue; besides providing Level trigger (Level trigger) of select/poll IO event, epoll also provides Edge trigger (Edge trigger), which makes it possible for user space program to buffer IO state, reduces epoll _ wait/epoll _ pwait calls, and improves application program efficiency.
Preferably, creating and initializing an IO multiplex epoll instance includes:
according to the classification of the flow of the sending end, initializing a socket of a UDP data receiving thread and the UDP data receiving thread, binding a port number for the UDP data receiving thread, and adding a specific socket to an event to an epoll instance;
initializing a socket of a TCP port of an acquisition end, binding the socket with the socket, monitoring the socket of the TCP port of the acquisition end, adding a read event to an epoll instance to the monitored socket, and initializing a processed data sending thread.
At least one pair of port numbers (sockets) is needed for establishing the network communication connection, the Socket is essentially A Programming Interface (API), and for encapsulating TCP/IP, the TCP/IP also provides an interface which can be used by programmers for network development, namely the Socket programming interface.
Preferably, the UDP data receiving thread includes an elephant stream UDP data receiving thread, an sflow UDP data receiving thread, and a delayed UDP data receiving thread; the processed data sending thread comprises a data sending thread processed by elephant flow, a data sending thread processed by sflow and a data sending thread processed by time delay.
The sFlow is a network monitoring technology jointly developed by InMon, HP and Foundation networks in 2001, adopts a data stream random sampling technology, can provide complete flow information from a second layer to a fourth layer, even in the whole network range, can adapt to flow analysis in an environment with super-large network flow (such as more than 10Gbit/s), and enables a user to analyze the performance, trend and existing problems of a network transmission flow in detail and in real time.
Preferably, the epoll example described above employs edge triggering.
Preferably, the kernel events include a readable event, an occurrence of a connection event, a writable event, and a disconnection event, and the listening for the kernel events includes:
under a first working condition, when a readable event is monitored, judging the source of the readable event through a main thread;
if the source switch sends a UDP data arrival signal to the UDP data receiving thread through the main thread;
if the TCP port of the source acquisition end, judging and broadcasting the condition variable of the corresponding data type through the main thread, and activating the work thread to process UDP data to be sent;
under a working condition II, when a connection event is monitored, the main thread receives an acquisition notice and an acquisition demand sent by a TCP port of an acquisition end, if the connection fails, a socket receiving the acquisition notice is added with a read event to an epoll instance, and a readable event on the socket receiving the acquisition notice is monitored continuously;
when a writable event is monitored, the kernel buffer area can be written, and a corresponding processed data sending thread is informed to send the processed data according to the condition variable of the corresponding data type;
and under the fourth working condition, when a disconnection event is monitored, closing the corresponding socket.
Preferably, the sending the processed data to the TCP port of the collection end by the processed data sending thread includes:
after the work thread processes the UDP data, releasing a data processing completion signal;
if the kernel buffer area is full, the processed data sending thread records the processed data which is sent completely, adds a socket corresponding to sending into the epoll instance, and monitors writable events;
and when the kernel buffer area can be written again, transmitting the processed data to be transmitted to the TCP port of the acquisition end through the corresponding processed data transmission thread.
In a second aspect, the present invention provides a data transfer method for a switch, including:
the switch sends UDP data;
according to the high-concurrency data transmission method, UDP data from the switch is received, multi-thread processing is carried out on the UDP data, and the processed data are sent to the acquisition end.
The highly concurrent data transfer method and the data transfer method for the switch of the invention have the following advantages:
1. monitoring the kernel event by an IO multiplexing epoll technology, simultaneously monitoring the receiving of UDP data and the sending of the processed data, and integrating switch sampling application and acquisition application by multithreading asynchronous processing, thereby solving the receiving bottleneck problem of frequently sent data by a switch;
2. multithreading asynchronous processing integrates switch sampling application and acquisition application, and dynamically allocates space in an application buffer area according to the size of received UDP data, so that data entering a kernel buffer area is guaranteed not to be lost and blocked;
3. the socket is in a non-blocking mode, the thread cannot block the socket to influence efficiency, data in a socket buffer which is not processed can be dynamically registered to enter a kernel for monitoring, and the data can be continuously processed when the next event is monitored.
Drawings
In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings needed to be used in the description of the embodiments will be briefly introduced below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and it is obvious for those skilled in the art to obtain other drawings based on these drawings without creative efforts.
The invention is further described below with reference to the accompanying drawings.
Fig. 1 is a flow chart of a highly concurrent data transfer method of embodiment 1.
Detailed Description
The present invention is further described in the following with reference to the drawings and the specific embodiments so that those skilled in the art can better understand the present invention and can implement the present invention, but the embodiments are not to be construed as limiting the present invention, and the embodiments and the technical features of the embodiments can be combined with each other without conflict.
The embodiment of the invention provides a high-concurrency data transmission method and a data transmission method for a switch, which are used for solving the technical problem of how to realize the receiving, processing and transmitting of high-concurrency data.
Example 1:
as shown in fig. 1, in the highly concurrent data transmission method of the present invention, UDP data from a transmitting end is received based on an IO multiplexing technique, multi-thread processing is performed on the UDP data according to an acquisition requirement of an acquisition end to obtain processed data, and the processed data is transmitted to a TCP port of the acquisition end based on the IO multiplexing technique.
The method comprises the following steps:
s100, creating and initializing an IO multiplexing epoll instance, and monitoring a kernel event;
s200, when a UDP data arrival signal is monitored, transferring the UDP data to an application buffer area through a UDP data receiving thread;
s300, when the acquisition notice is monitored, broadcasting condition variables of corresponding data types through the main thread, activating a work thread through the acquisition notice receiving thread, and performing multi-thread parallel processing on UDP data in corresponding application buffer areas through the work thread to obtain processed data;
and S400, transmitting the processed data to a TCP port of the acquisition end through the processed data transmission thread.
In step S100, creating and initializing an IO multiplex epoll instance, including the following sub-steps:
s110, initializing sockets of elephant flow, sflow and delay data according to the classification of flow of a sending end, binding ports, adding a reading event into an epoll example by using edge triggering, initializing udp data receiving threads of elephant flow, sflow, delay and the like, and respectively waiting for corresponding signals of arrival of udp data by using the udp data receiving threads;
s120, initializing a socket of a TCP port of an acquisition end, binding the socket with the socket, monitoring the socket of the TCP port of the acquisition end, adding a read event to an epoll instance by using edge triggering, initializing a data sending thread after elephant flow processing, a data sending thread after sflow processing and a data sending thread after delay processing, and respectively waiting for signals of different types of data after processing by the data sending thread after processing.
The kernel event comprises a readable event, a connection event, a writable event and a disconnection event, the kernel event is monitored, if the event corresponding to the kernel is triggered, the kernel goes to the corresponding processing logic, and the specific processing logic mainly comprises:
under a first working condition, when a readable event is monitored, judging the source of the readable event through a main thread;
if the source switch is (sfow/elephant stream/delayed data) UDP data, the main thread sends a UDP data arrival signal and informs a UDP data receiving thread (sfow/elephant stream/delayed data thread) to receive the UDP data;
if the TCP port of the source acquisition end judges the data type (sfow/elephant flow/delay data) through the main thread, broadcasting the condition variable corresponding to the data type, informing a work thread in a thread pool to receive the received UDP data, and processing the data to be sent to the acquisition end;
under a working condition II, when a connection event of a TCP port of an acquisition end is monitored, a main thread receives an acquisition notice and an acquisition demand sent by the TCP port of the acquisition end, if one-time connection fails, a socket receiving the acquisition notice is added with a read event to an epoll instance, and a readable event on the socket receiving the acquisition notice is continuously monitored;
under a third working condition, when a writable event is monitored, the kernel buffer area can be written, and a corresponding processed data sending thread is informed to send processed data according to a condition variable of a corresponding data type;
and under the fourth working condition, when a disconnection event is monitored, closing the corresponding socket.
In step S200, after receiving the UDP data arrival signal, the UDP data receiving thread (slow UDP thread/elephant stream UDP thread/delayed UDP thread) starts receiving the UDP data, and reads the UDP data from the kernel buffer to the application buffer.
In step S300, after receiving the acquisition notification and the acquisition requirement, the main thread determines a corresponding data type variable (slow condition variable/elephant flow condition variable/delay condition variable), and broadcasts a condition variable (slow condition variable/elephant flow condition variable/delay condition variable) of the corresponding data type, and the work thread in the thread pool is activated, and can process data in the corresponding application buffer in a multi-thread manner.
In step S400, the processed data is sent to the TCP port of the collection end through the processed data sending thread, which includes the following sub-steps:
s410, after the work thread finishes processing the UDP data, releasing a data processing finishing signal;
s420, if the kernel buffer area is full, the processed data sending thread records the sent processed data, adds a corresponding socket to the epoll instance and monitors writable events;
and S330, when the kernel buffer area can be written again, transmitting the processed data to be transmitted to the TCP port of the acquisition end through the corresponding processed data transmission thread.
Example 2:
the invention provides a data transmission method for an exchanger, which receives UDP (user Datagram protocol) data sent by the exchanger through the high-concurrency data transmission method disclosed by the embodiment 1, concurrently processes the data through a work thread, and transmits the processed data to an acquisition end.
The above-mentioned embodiments are merely preferred embodiments for fully illustrating the present invention, and the scope of the present invention is not limited thereto. The equivalent substitution or change made by the technical personnel in the technical field on the basis of the invention is all within the protection scope of the invention. The protection scope of the invention is subject to the claims.
Claims (6)
1. The high-concurrency data transmission method is characterized in that UDP data from a sending end is received based on an IO multiplexing technology, multithreading processing is carried out on the UDP data according to the acquisition requirement of an acquisition end to obtain processed data, and the processed data are transmitted to a TCP port of the acquisition end based on the IO multiplexing technology;
the method comprises the following steps:
creating and initializing an IO multiplexing epoll instance, and monitoring a kernel event;
when a UDP data arrival signal is monitored, the UDP data is transferred to an application buffer area through a UDP data receiving thread;
when monitoring the acquisition notification, broadcasting condition variables of corresponding data types through a main thread, activating a work thread through the acquisition notification receiving thread, and performing multi-thread parallel processing on UDP data in corresponding application buffer areas through the work thread to obtain processed data;
the processed data are sent to a TCP port of an acquisition end through a processed data sending thread;
the kernel events comprise readable events, connection events, writable events and disconnection events, and the monitoring of the kernel events comprises the following steps:
under a first working condition, when a readable event is monitored, judging the source of the readable event through a main thread;
if the source switch sends a UDP data arrival signal to the UDP data receiving thread through the main thread;
if the TCP port of the source acquisition end, judging and broadcasting the condition variable of the corresponding data type through the main thread, and activating the work thread to process UDP data to be sent;
under a working condition II, when a connection event is monitored, the main thread receives an acquisition notice and an acquisition demand sent by a TCP port of an acquisition end, if the connection fails, a socket receiving the acquisition notice is added with a read event to an epoll instance, and a readable event on the socket receiving the acquisition notice is monitored continuously;
when a writable event is monitored, the kernel buffer area can be written, and a corresponding processed data sending thread is informed to send the processed data according to the condition variable of the corresponding data type;
and under the fourth working condition, when a disconnection event is monitored, closing the corresponding socket.
2. The method of claim 1, wherein creating and initializing an IO multiplex epoll instance comprises:
according to the classification of the flow of the sending end, initializing a socket of a UDP data receiving thread and the UDP data receiving thread, binding a port number for the UDP data receiving thread, and adding a specific socket to an event to an epoll instance;
initializing a socket of a TCP port of an acquisition end, binding the socket with the socket, monitoring the socket of the TCP port of the acquisition end, adding a read event to an epoll instance to the monitored socket, and initializing a processed data sending thread.
3. The method according to claim 1, wherein the UDP data receiving thread includes an elephant stream UDP data receiving thread, an sflow UDP data receiving thread, and a delayed UDP data receiving thread;
the processed data sending thread comprises a data sending thread processed by elephant flow, a data sending thread processed by sflow and a data sending thread processed by time delay.
4. The method of claim 1, wherein the epoll instance uses edge triggering.
5. The method according to claim 1, 2, 3 or 4, wherein the sending the processed data to the TCP port of the collecting end by the processed data sending thread includes:
after the Work thread processes the UDP data, releasing a data processing completion signal;
if the kernel buffer area is full, the processed data sending thread records the processed data which is sent completely, adds a socket corresponding to sending into the epoll instance, and monitors writable events;
and when the kernel buffer area can be written again, transmitting the processed data to be transmitted to the TCP port of the acquisition end through the corresponding processed data transmission thread.
6. Data transfer method for a switch, characterized by comprising:
the switch sends UDP data;
by the highly concurrent data transfer method according to any one of claims 1 to 5, receiving UDP data from a switch, and after performing multi-thread processing on the UDP data, sending the processed data to a collection side.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910249826.4A CN109947581B (en) | 2019-03-29 | 2019-03-29 | High concurrency data transfer method and data transfer method for switch |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910249826.4A CN109947581B (en) | 2019-03-29 | 2019-03-29 | High concurrency data transfer method and data transfer method for switch |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109947581A CN109947581A (en) | 2019-06-28 |
| CN109947581B true CN109947581B (en) | 2021-08-27 |
Family
ID=67012972
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910249826.4A Active CN109947581B (en) | 2019-03-29 | 2019-03-29 | High concurrency data transfer method and data transfer method for switch |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109947581B (en) |
Families Citing this family (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112631788B (en) * | 2021-01-06 | 2023-11-28 | 上海哔哩哔哩科技有限公司 | Data transmission method and data transmission server |
| CN112954373A (en) * | 2021-01-26 | 2021-06-11 | 四川天翼网络服务有限公司 | High-concurrency multi-protocol-supported audio and video on demand system and method |
| CN113468081B (en) * | 2021-07-01 | 2024-05-28 | 福建信息职业技术学院 | Device and method for converting serial port into udp based on ebi bus |
| CN113342554B (en) * | 2021-08-02 | 2022-01-04 | 阿里云计算有限公司 | IO multiplexing method, medium, device and operating system |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103164256A (en) * | 2011-12-08 | 2013-06-19 | 深圳市快播科技有限公司 | Processing method and system capable of achieving one machine supporting high concurrency |
| CN103780517A (en) * | 2014-01-20 | 2014-05-07 | 江苏省未来网络创新研究院 | Driving system and driving method for cooperative work of controller and exchanger under different versions of OpenFlow |
| CN106411777A (en) * | 2016-09-27 | 2017-02-15 | 福建中金在线信息科技有限公司 | Method and system for processing high concurrent data |
| CN108459915A (en) * | 2018-01-31 | 2018-08-28 | 中电福富信息科技有限公司 | A kind of TCP channel suitable for high concurrent scene manages system and method |
| CN109298943A (en) * | 2018-10-11 | 2019-02-01 | 四川长虹电器股份有限公司 | Concurrent processing method for UDP server |
Family Cites Families (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US9384071B2 (en) * | 2011-03-31 | 2016-07-05 | Solarflare Communications, Inc. | Epoll optimisations |
| CN102316116B (en) * | 2011-09-29 | 2014-05-21 | 高新兴科技集团股份有限公司 | System for supporting multiple platform network communication processing and method thereof |
| CN105049162A (en) * | 2015-05-21 | 2015-11-11 | 北京科东电力控制系统有限责任公司 | Public network information high-efficiency transmission method based on Epoll model |
| CN107484227B (en) * | 2017-09-01 | 2021-01-01 | 天津赞普科技股份有限公司 | Wifi networking multi-hot-spot control communication method |
-
2019
- 2019-03-29 CN CN201910249826.4A patent/CN109947581B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103164256A (en) * | 2011-12-08 | 2013-06-19 | 深圳市快播科技有限公司 | Processing method and system capable of achieving one machine supporting high concurrency |
| CN103780517A (en) * | 2014-01-20 | 2014-05-07 | 江苏省未来网络创新研究院 | Driving system and driving method for cooperative work of controller and exchanger under different versions of OpenFlow |
| CN106411777A (en) * | 2016-09-27 | 2017-02-15 | 福建中金在线信息科技有限公司 | Method and system for processing high concurrent data |
| CN108459915A (en) * | 2018-01-31 | 2018-08-28 | 中电福富信息科技有限公司 | A kind of TCP channel suitable for high concurrent scene manages system and method |
| CN109298943A (en) * | 2018-10-11 | 2019-02-01 | 四川长虹电器股份有限公司 | Concurrent processing method for UDP server |
Non-Patent Citations (2)
| Title |
|---|
| 基于TCP、UDP的多路多数据流融合网络系统设计;龙玉军;《电子设计工程》;20111105;第19卷(第21期);第69-72页 * |
| 采用epoll模型设计多路udp接收程序;dong_beijing;《https://blog.csdn.net/dong_beijing/article/details/60876591》;20170308;第1-5页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109947581A (en) | 2019-06-28 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109947581B (en) | High concurrency data transfer method and data transfer method for switch | |
| US9143467B2 (en) | Network interface controller with circular receive buffer | |
| Buonadonna et al. | An implementation and analysis of the virtual interface architecture | |
| US6594701B1 (en) | Credit-based methods and systems for controlling data flow between a sender and a receiver with reduced copying of data | |
| US9112894B2 (en) | Real time distributed network monitoring and security monitoring platform (RTD-NMS) | |
| García-Dorado et al. | High-performance network traffic processing systems using commodity hardware | |
| CN105141603B (en) | Communication data transmission method and system | |
| Welsh et al. | Atm and fast ethernet network interfaces for user-level communication | |
| US11444890B2 (en) | Method and device for improving bandwidth utilization in a communication network | |
| WO2014082562A1 (en) | Method, device, and system for information processing based on distributed buses | |
| CN111522663B (en) | Data transmission method, device and system based on distributed storage system | |
| CN113067810B (en) | Network packet capturing method, device, equipment and medium | |
| CN106789099B (en) | PCIE-based high-speed network isolation method and terminal | |
| CN107479966B (en) | Signaling acquisition method based on multi-core CPU | |
| CN109800035B (en) | Algorithm integrated service framework system | |
| US20070288646A1 (en) | Communication interface device and communication method | |
| GB2458952A (en) | A device queues packets and stores a state parameter for each packet, indicating a state of an entity associated with the packet, e.g. reset state of a server | |
| MacArthur et al. | An efficient method for stream semantics over rdma | |
| CN111131081A (en) | Method and device for supporting multi-process high-performance unidirectional transmission | |
| CN117421122A (en) | Data transmission method, network card and CPU | |
| CN116610530A (en) | Processing method and device of network data, computer equipment and storage medium | |
| CN1306772C (en) | Access method of short packet data | |
| Baker et al. | Via communication performance on a gigabit ethernet cluster | |
| US20250139023A1 (en) | ALLOWING NON-VOLATILE MEMORY EXPRESS (NVMe) OVER FABRIC (NVMe-oF) TRAFFIC OVER INTERFACES USING A SCALABLE END POINT (SEP) ADDRESSING MECHANISM | |
| KR100534618B1 (en) | apparatus method of exchange packet in socket communication processor |
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 | ||
| CB03 | Change of inventor or designer information | ||
| CB03 | Change of inventor or designer information |
Inventor after: Zhu Naiguo Inventor after: Li Chao Inventor after: Yu Hao Inventor before: Li Chao Inventor before: Yu Hao |
|
| CB02 | Change of applicant information | ||
| CB02 | Change of applicant information |
Address after: 250100 No. 1036 Tidal Road, Jinan High-tech Zone, Shandong Province, S01 Building, Tidal Science Park Applicant after: Inspur cloud Information Technology Co.,Ltd. Address before: 250100 No. 1036 Tidal Road, Jinan High-tech Zone, Shandong Province, S01 Building, Tidal Science Park Applicant before: Tidal Cloud Information Technology Co.,Ltd. Address after: 250100 No. 1036 Tidal Road, Jinan High-tech Zone, Shandong Province, S01 Building, Tidal Science Park Applicant after: Tidal Cloud Information Technology Co.,Ltd. Address before: 250100 S06 tower, 1036, Chao Lu Road, hi tech Zone, Ji'nan, Shandong. Applicant before: SHANDONG INSPUR CLOUD INFORMATION TECHNOLOGY Co.,Ltd. |
|
| GR01 | Patent grant | ||
| GR01 | Patent grant |