WO2020180266A1

WO2020180266A1 - Data centers which function with active architecture for work continuity

Info

Publication number: WO2020180266A1
Application number: PCT/TR2019/050384
Authority: WO
Inventors: Furkan DUMAN; Onur ALTUN; Ragip Semih OCALANKOL
Original assignee: Kale Yazilim Sanayi Ve Ticaret Anonim Sirketi
Priority date: 2019-03-04
Filing date: 2019-05-28
Publication date: 2020-09-10
Also published as: TR201903183A2

Abstract

The present invention is a system comprising at least two data centers (100), namely, a data center A (101) and a data center B (102) connected to a communication network (400) and where said data center A (101) and said data center B (102) are connected to each other in a manner providing data exchange and having a memory unit (120) keeping the tables (123) including data, a processor unit (110) associated with said memory unit (120) in a manner reading data and writing data and comprising pluralities of client devices (200) sending data processing request to the table (123) by means of said communication network (400) to said data centers (100). Accordingly, the subject matter invention is characterized in that the processor unit (110) of the data center A (101) is configured to provide sending of a data synchronization message, including information related to data processing, to said data center B (102) for realizing the data processing in the related table (123) in its own memory unit (120) when the data processing request, sent by one of said client devices (200), is detected.

Description

DATA CENTERS WHICH FUNCTION WITH ACTIVE ARCHITECTURE FOR WORK

CONTINUITY

SPECIFICATION

TECHNICAL FIELD

The present invention relates to a system which functions with active architecture where said data center A and said data center B are connected to each other in a manner providing data exchange and having a memory unit keeping the tables including data, a processor unit associated with said memory unit in a manner reading data and writing data and comprising at least two data centers, namely, a data center A and a data center B connected to a communication network and comprising pluralities of client devices sending data processing request to the table by means of said communication network to said data centers.

PRIOR ART

Data centers provide storage of data in databases thereof and provide management of these data. The client devices, which send request to the data center from outside, can make changes in the permitted sections in the database or can add data.

Various methods and systems have been developed for preventing data loss as a result of a probable failure or probable disaster at data centers. One of these methods and systems is the traditional back up method. All of the data is backed up in a server. In case the data center is damaged, the data are recovered. In this case, a server or a data center stays passive and only realizes data storage function.

In traditional data center recovery solutions, the data are in general backed up in an asynchronous manner from a data center to another data center. When data is desired to be backed up in a synchronous manner, distance limitations occur due to high line capacity requirement and dependingly due to high costs. In case problems are faced in the active data center or in case maintenance is realized, service interruption and data loss may occur during the time until the other data center is activated.

In the application with number US2017317875A1 , a system is disclosed where work continuity is provided and where when one of the databases which functions active-active becomes inactive, the other database continues data processing service and where work continuity is provided.

As a result, because of all of the abovementioned problems, an improvement is required in the related technical field.

BRIEF DESCRIPTION OF THE INVENTION

The present invention relates to a system and a synchronization method, for eliminating the above mentioned disadvantages and for bringing new advantages to the related technical field.

An object of the present invention is to provide a system and a synchronization method which provides work continuity in storage of data.

Another object of the present invention is to provide a system and a synchronization method which provides prevention of data loss in case of a disaster.

In order to realize all of the abovementioned objects and the objects which are to be deducted from the detailed description below, the present invention is a system comprising at least two data centers, namely, a data center A and a data center B connected to a communication network and where said data center A and said data center B are connected to each other in a manner providing data exchange and having a memory unit keeping the tables including data, a processor unit associated with said memory unit in a manner reading data and writing data and comprising pluralities of client devices sending data processing request to the table by means of said communication network to said data centers. Accordingly, the subject matter invention is characterized in that the processor unit of the data center A is configured to provide sending of a data synchronization message, including information related to data processing, to said data center B for realizing the data processing in the related table in its own memory unit when the data processing request, sent by one of said client devices, is detected. Thus, each data center stays synchronous.

In a possible embodiment of the present invention, the processor unit of the data center B is configured to provide sending of data synchronization message, including information related to data processing, to said data center B, in order to realize the data processing in the related table in its own memory unit when the data processing request sent by one of said client devices is detected. In another possible embodiment of the present invention, a distributor device is provided between the client devices and the data centers; said distributor device is configured to detect a data traffic of the data centers and to send the data processing request to one of the data centers according to the detected data traffic when the processing request coming from at least one of the client devices is detected. Thus, all data centers operate in an active manner and the processes of the clients are realized in an instantaneous and consistent manner.

In another possible embodiment of the present invention, the data centers are configured to realize the data processing in the data processing request in a synchronous manner with the 2 phase commit (2pc) protocol.

In another possible embodiment of the present invention, the data centers are arranged to transfer the predetermined changes in the tables in a synchronous manner and to transfer the other changes in the tables in an asynchronous manner.

In another possible embodiment of the present invention, the processor unit of the data center A is configured to:

record the changes made in the tables in case it is detected that the data center B is deactivated,

provide sending of data synchronization messages, including the recorded changes, to the data center B in case it is detected that the data center B is in operating state in order to realize said changes in the tables in its own memory unit. Thus, in case of disaster, the protection of the data and re-synchronization are provided.

In another possible embodiment of the present invention, the processor unit is configured to bring the data center to the passive or closed state in case a predetermined fault is detected during data processing.

In another possible embodiment of the present invention, the data centers comprise an input/output unit and the processor unit is configured to bring the data center to active, closed or passive state according to an input received from the input/output unit.

The present invention is moreover a data synchronization method realized by a system comprising at least two data centers, namely, a data center A and a data center B connected to a communication network and where said data center A and said data center B are connected to each other in a manner providing data exchange and having a memory unit keeping the tables including data, a processor unit associated with said memory unit in a manner reading data and writing data and comprising pluralities of client devices sending data processing request to the table by means of said communication network to said data centers. Accordingly, the improvement of the present invention is that the following steps are realized by the processor unit of the data center A:

- detecting the data processing request sent by one of said client devices,

- providing sending data synchronization message, including information related to data processing, to said data center B in order to realize the data processing in the related table in its own memory unit.

BRIEF DESCRIPTION OF THE FIGURES

Figure 1 is a representative view of the database network.

Figure 2 is a more detailed representative view of a data center.

Figure 3 is a representative view of another possible embodiment of the system.

DETAILED DESCRIPTION OF THE INVENTION

In this detailed description, the subject matter is explained with references to examples without forming any restrictive effect only in order to make the subject more understandable.

The subject matter is a system having data centers (100) where the client devices (200) can realize data processing to the tables (123) in the memory unit (120). As will be described in detail below, instantaneous synchronization is provided between the data centers (100) as the triggerring units associated with the tables (123) are activated as data exchange is realized and as the same exchange is realized to the other data centers (100). Since the triggering units are activated at the instant when data change is made, the databases (122) continuously stay synchronous.

As also known in the art, the triggering units are functional modules formed by command lines which provide realization of various processes by being operated when data change is made in the associated table (123) as also known in the art.

In a possible embodiment, different triggering units can be provided for pluralities of tables. These triggering units can provide sending of the data in the tables by means of one of synchronous communication and asynchronous communication methods. The tables which shall be instantaneously consistent at each data center are selected to be sent in a synchronous manner and they are associated with the related triggering unit arranged to realize this. The tables, which do not have to be instantaneously consistent at each data center for the data, are selected to be sent in an asynchronous manner and they are associated with the related triggering unit arranged to realize this. Thus, the data loss of the tables which cannot bear data loss is prevented and the synchronization of the tables (for instance, the tables where the data like error records are kept) which cannot bear data loss is prevented from consuming excessive sources.

With reference to Figure 1 , the data center network comprises at least the data center A (101) connected to a communication network (400) and at least the data center B (102) connected to said communication network (400). Pluralities of client devices (200) are connected to the communication network (400) and message request can be sent to the data centers (100) related to data processing in the tables (123) in the memory units (120) of the data centers (100). A distributor device (300) is provided between the data centers (100) and the communication network (400). Said distributor device (300) distributes the requests received from the clients to the data centers (100). The distributor device (300) realizes said distribution process according to the present work load of the data centers (100) and/or according to the data traffic to the data centers (100). In other words, the distributor device (300) realizes load balancing for the data centers (100). Here, said communication network (400) is a wide area network, particularly internet.

The data centers (100) are associated with the communication network (400) by means of the communication unit (150) therein.

All of the data centers (100) function in an active manner. When required, one of the data centers (100) can be brought to passive state or closed state.

Here, the mentioned passive state describes the state where the synchronization of both databases (122) continues but where the passive data center (100) does not receive request from the client devices (200), and the closed state describes the state where the closed data center (100) does not receive request from the client devices and where the syncronization of the database (122) is not realized.

The data center A (101) and the data center B (102) are associated with each other so as to realize data exchange. With reference to Figure 2, each data center (100) comprises a memory unit (120). Said memory unit (120) can comprise memories which store data permanently and/or temporarily or can comprise suitable combinations of them.

A processor unit (110) is provided so as to write data to the memory unit (120) and so as to read data from the memory unit (120). The processor unit (110) can comprise pre programmed microprocessors.

The memory unit (120) can comprise a database (122) having tables (123) therein. The client devices (200) request for data processing on the tables (123) in this database. Here, the mentioned data processing describes the processes like erasing of the data in the areas of the tables (123), adding data and changing data. In order to be more descriptive and provided that no delimiting effect is created, for instance, a database can be a system where the records of civil registry, insurance, etc. of the citizens are kept, and the exemplary client devices (200) can be devices like computer, mobile telephone, tablet computer, etc. where the operators can realize change/update on these records.

The memory unit (120) can moreover comprise database management software (121) consisting of functional modules comprising command lines executed for realizing the processes to be made by the processor unit (110) in the database and for providing communication with the other database.

The data centers comprise a communication unit for connecting to the communication network (400) and for providing data exchange with each other. Said communication unit can comprise one or more than one network accommodating hardware. The data centers (100) can moreover comprise an input/output unit (160) for programming the memory unit (120) and/or the processor units (110) or for receiving output from the memory unit (120) and/or from the processor unit (110). Data input can be realized or data output can be received by means of the peripheral devices like keyboard, mouse, screen, etc. connected to the input/output unit (160).

Data centers can moreover comprise a data path (140) which provides data exchange in a suitable manner between the memory unit (120), the processor unit (110), the communication unit, the input/output unit (160) and the other components which have not been mentioned here but which are known in the art and in order to provide realization of data exchange of the other components, known in the art, in a suitable manner with each other. As an improvement of the invention, the database management software (121) comprises pluralities of triggering units associated with the tables (123). Said triggering units are activated when a change is realized in a line and provide sending of a data synchronization message, including the details related to this change, to the other database for realizing the change in the memory unit (120) of the other database. In other words, when the processor unit (110) of one of the data centers (100) realizes data processing in the table (123), it provides sending of a data synchronization message, including only the information related to said data processing, to the other database. In other words, when the triggering unit detects that data processing is realized in the table (123), a data synchronization message, including only the information related to this data processing, is sent to the other database (100). The other database makes change in tables (123) in the memory unit (120) according to this message. Thus, during synchronization, it is not necessary to change the whole memory and only the parts where change is made shall be changed and a synchronization method is provided where consistency and continuity are increased.

As another improvement of the invention, the instantaneous data change between the data centers (100) is realized in a synchronous manner. In other words, data processing is realized by means of 2 phase commit (2pc) protocol. Thus, it is assured that the data change is realized in each data center (100).

As another improvement of the invention, when one of the data centers (100) passes to closed condition because of undesired reasons, in other words, in case of a disaster, the processes are continued and data loss is prevented. Here, said disaster state can be power failure, connection failure, planned works within the scope of failure software and hardware maintenance works or natural disasters, etc. in the region where one of the data centers (100) exists.

This improvement of the invention is basically as follows: The processor unit (110) of the data center A (101) detects that data center B (102) is in disaster state, in other words the data center B (102) is in closed state. The processor unit (110) realizes the coming data processing requests and records each process. When the procesor unit (110) of the data center A (101) detects that the data center B (102) arrives at the active state, it sends data synchronization messages including the recorded processes in order for the data center B (102) to realize the same changes in the databases in its own memory unit (120). The data center B (102) realizes changes in the tables (123) in the database according to the data synchronization messages and provides synchronization. After synchronization is provided, data center B can be activated. The mentioned processor unit (110) can comprise pluralities of processors. These processors can be positioned at different servers. In a possible embodiment of the present invention shown in Figure 3, the servers where the processors of this processor unit are distributed are given. A synchronization server (520) can also be provided in a separate application server (510) and the database (122) in the memory unit (120) is controlled by an application server (510). The synchronization server (520) provides data synchronization and moreover, the synchronization server (520) checks the health condition of the related data center (100) and according to the health information coming from the synchronization server (520) in the other data center (100), the operation mode (active-active, active-passive, active-closed, etc.) of the system can be determined.

In another embodiment, the database (122) is controlled by a database server (not illustrated in the figure) and an application server interacts with the client devices (200) and provides making change in the database (122), and the processor unit (110) which provides realization of the synchronization and which provides determination of the operation mode of the data centers are provided in the synchronization server (520).

The synchronization servers (520) transfer the health condition information to the load distributor device (300) and determine whether requests arrive at the data centers where said synchronization servers (520) exist.

In a possible embodiment of the present invention, the synchronization server (520) brings the related data center to closed or passive state when fault or faults, having predetermined characteristic related to the database (122) or related to the application where the clients access from the application server (510), is/are received. Said characteristic can be the fault type, the fault receiving frequency, the source of the fault, etc. In a possible embodiment, pluralities of application servers (510) provides pluralities of databases (122) and these databases (122) and these application servers (510) can be associated with each other. For instance, one of the application servers (510) can make process in pluralities of databases (122). Application servers which are N in number can be associated with databases which are in M in number in various combinations. For instance, in case application servers which are X in number among the application servers which are N in number do not work, the synchronization server (520) can bring the data center to the passive state.

When the synchronization server (520) detects fault with the predetermined characteristic related to one of the related application servers (510) and the databases (122), it can turn off only the database (122) or the application server (510) where error is taken and which is associated with the one where error is taken.

In another possible embodiment of the present invention, the data center (100) where the input/output unit exists can be brought to passive or closed state by means of an input/output unit (160) in accordance with the request for maintenace, etc. and it can also be brought to active state after operation.

The operation of a possible embodiment of the present invention whose details are given above is as follows: One of the client devices (200) requests for data processing for realizing change in an area of a line of one of the tables (123) in the database. The distributor device (300) receives this data processing request and selects one of the data centers (100) according to another parameter like the process density of the data centers (100) or like the geographical closeness of the data centers (100) to the client device (200). The distributor device (300) transfers this request to the selected database (100) (in this exemplary embodiment, the selected data center (100) is the data center A (101)). The selected data center (100) arranges the table (123) according to the data processing request. When the processor unit (110) detects data change, it activates the triggering units associated with the related table (123) and the triggering units send the realized changes to the database at the other data center (100), in other words, the triggering units send the data synchronization message comprising data processing to the database in the other data center (100) (in this exemplary embodiment, the other data center (100) is the data center B (102)). The processor unit (110) of the database in the data center B (102) provides making change in the tables (123) in the database according to the data synchronization message. Thus, the databases (122) in each data center (100) are the same.

In an exemplary condition where the data center B (102) is in disaster state, in other words where the data center B (102) is in closed state, the data center A (101) detects this state. The distributor device (300) guides all of the data processing requests coming from the client devices (200) to the data center A (101). Data center A (101) changes the database (122) according to the data processing requests and moreover records each change made. When the data center A detects that the data center B (102) becomes active again, the data center A sends the recorded changes with the data synchronization messages and provides synchronization of the database of the data center B (102) and the database of the data center A (101). Data center B (102) realizes the data processing provided in the data synchronization messages received from the data center A (101) in its own database (122). The protection scope of the present invention is set forth in the annexed claims and cannot be restricted to the illustrative disclosures given above, under the detailed description. It is because a person skilled in the relevant art can obviously produce similar embodiments under the light of the foregoing disclosures, without departing from the main principles of the present invention.

REFERENCE NUMBERS

100 Data center

110 Processor unit

120 Memory unit

121 Database management software

122 Database

123 Table

140 Data path

150 Communication unit

160 Input/output unit

101 Data center A

102 Data center B

200 Client device

300 Distributor device

400 Communication network

510 Application server

520 Synchronization server

Claims

1. A system comprising at least two data centers (100), namely, a data center A (101) and a data center B (102) connected to a communication network (400) and where said data center A (101) and said data center B (102) are connected to each other in a manner providing data exchange and having a memory unit (120) keeping the tables (123) including data, a processor unit (110) associated with said memory unit (120) in a manner reading data and writing data and comprising pluralities of client devices (200) sending data processing request to the table (123) by means of said communication network (400) to said data centers (100), characterized in that the processor unit (110) of the data center A (101) is configured to provide sending of a data synchronization message, including information related to data processing, to said data center B (102) for realizing the data processing in the related table (123) in its own memory unit (120) when the data processing request, sent by one of said client devices (200), is detected.

2. The system according to claim 1 , wherein the processor unit (110) of the data center B (101) is configured to provide sending of data synchronization message, including information related to data processing, to said data center B (102), in order to realize the data processing in the related table (123) in its own memory unit (120) when the data processing request sent by one of said client devices (200) is detected.

3. The system according to claim 1 , wherein a distributor device (300) is provided between the client devices (200) and the data centers (100); said distributor device (300) is configured to detect a data traffic of the data centers (100) and to send the data processing request to one of the data centers (100) according to the detected data traffic when the processing request coming from at least one of the client devices (200) is detected.

4. The system according to claim 1 , wherein the data centers (100) are configured to realize the data processing in the data processing request in a synchronous manner with the 2 phase commit (2pc) protocol.

5. The system according to claim 1 , wherein the data centers (100) are arranged to transfer the predetermined changes in the tables in a synchronous manner and to transfer the other changes in the tables in an asynchronous manner.

6. The system according to claim 1 , wherein the processor unit (110) of the data center A (101) is configured to:

record the changes made in the tables (123) in case it is detected that the data center B (100) is deactivated,

provide sending of data synchronization messages, including the recorded changes, to the data center B (102) in case it is detected that the data center B (100) is in operating state in order to realize said changes in the tables (123) in its own memory unit (120).

7. The system according to claim 1 , wherein the processor unit (110) is configured to bring the data center (100) to the passive or closed state in case a predetermined fault is detected during data processing.

8. The system according to claim 1 , wherein the data centers comprise an input/output unit (160) and the processor unit (110) is configured to bring the data center to active, closed or passive state according to an input received from the input/output unit (160).

9. A data synchronization method realized by a system comprising at least two data centers (100), namely, a data center A (101) and a data center B (102) connected to a communication network (400) and where said data center A (101) and said data center B (102) are connected to each other in a manner providing data exchange and having a memory unit (120) keeping the tables (123) including data, a processor unit (110) associated with said memory unit (120) in a manner reading data and writing data and comprising pluralities of client devices (200) sending data processing request to the table (123) by means of said communication network (400) to said data centers (100), wherein the following steps are realized by the processor unit (110) of the data center A (101):

detecting the data processing request sent by one of said client devices (200), providing sending data synchronization message, including information related to data processing, to said data center B (102) in order to realize the data processing in the related table (123) in its own memory unit (120).

10. The synchronization method according to claim 9, wherein the data center B (102) comprises the following steps realized by the processor unit (110):

detecting the data processing request sent by one of said client devices (200), providing sending of data synchronization message, including information related to data processing, to the data center A (101) in order to realize the data processing in the related table (123) in its own memory unit (120).

11. The synchronization method according to claim 9, wherein the data center A (101) comprises the following steps realized by the processor unit (110):

recording the changes made in the tables (123) in case it is detected that data center B (100) is deactivated,

providing sending of data synchronization message, including the changes recorded in case it is detected that the data center B (100) is in operating condition, to the data center B (102) in order to realize said changes in the tables (123) in its own memory unit (120).