Due to the possibility of accidents such as machine failure, the high availability of databases has always been a concern. As a NoSQL database, TCA PlusDB provides scalability and high availability across data center scenarios. The purpose of this document is to introduce the TCAPLUSDB scheme for the deployment of dual data centers in the same city.

The principle of high availability technology of TcaplusDB

TCAPlus service modules already support high availability deployments

Here are the details:

  1. MySQL: Use a native MySQL Cluster Cluster or cloud provider mySQLDB. It is recommended to configure the domain name access, behind the multiple RS mode

    1. (Deployment method: refer to the deployment method on the Internet)

  2. OMS: Support to build multiple OMS management terminals, which are equal to each other and do not influence each other

    1. (Deployment mode: To add new machines after Mysql privileges, install xampp, loose the OMS package, execute. / install_web. Sh was introduced into the Mysql connection information as a parameter, once installed delete/data/tcaplus/tcapoms/WWW/app/install. PHP initial page, Can be used normally)

  3. Tcapcenter: Each cluster (set) has a ServingCenter, which supports the deployment of StandbyCenter and monitors the status of multiple ServingCenters in a one-to-many way. Once the ServingCenter is confirmed to be out of service, the role switch can be completed within minutes. Take over the subordinate processes of the original ServingCenter

    1. (Deployment: After adding MySQL privileges to the new machine, register the new cluster from “Cluster Maintenance” and then use the “Install StanbyCenter” document.)

  4. TCAPDIR: Each cluster (set) deploys at least 2 TCAPDIR, each TCAPDIR knows the complete list of directory servers, when the connected TCAPDIR fails, the client will automatically select other surviving TCAPDIR from the list for connection, so as to facilitate the daily process capacity expansion

    1. Mysql > assign Mysql privileges to TCAPDIR, and then initialize TCAPDIR into the pool.
    2. , fill in the login information, and put the machine into the pool. Refer to the chapter of Machine Shelving for details
    3. Click on directory service
    5. Click “Add directory service”
    7. Click Submit, and after installation, TCAPDIR has completed capacity expansion

  5. TCAPSVR: Storage layer with one master and one slave (hereinafter referred to as: In general, Master undertakes DB read and write requests. When Master fails, Proxy will quickly know the problem (internal statistics shows that within a certain period of time, no response was received after the request was sent to Master by Proxy). The Center is then notified to make a unified decision. When the switching conditions are met, the Slave will be notified to switch its role to Master and adjust the routing table. After the switching is completed, the TCA proxy will be notified to update the Master information

    1. (Deployment mode: refer to the storage layer scaling capacity)

  6. Tcaproxy: Deploy at least two TCA proxies in each game zone to achieve request balance at the client, and expand the capacity of TCA Proxy to enhance throughput capacity. When the process of TCA Proxy fails abnormally and the Center senses heartbeat interruption, it will notify the client to update the list of Proxy through TCAPDIR (eliminate abnormal process).

    1. (Deployment mode: refer to access layer scaling capacity)

  7. TCAPDB: Deploy at least two TCAPDBs for each service (APP). TCAPSVR will realize request balancing from multiple TCAPDBs when forwarding and caching data, and expand the capacity of TCAPDB to enhance throughput capacity

    1. (Deployment mode: refer to the scaling capacity of caching layer)

  8. Cold standby center (Gluster) : The storage cluster is in RAID-1 deployment mode. When a machine fails, the service will not be affected. After the machine is replaced or the hard disk is repaired, automatic data repair can be launched

    1. (Deployment: Refer to the chapter “Data Backup” where Gluster’s Operations Management mentions node expansion.)

TCAPLUSDB is a distributed NoSQL database produced by Tencent. The storage and scheduling codes are completely self-developed. It has features such as cache + floor-to-ground fusion architecture, petabyte storage, millisecond delay, lossless horizontal scaling and complex data structure. At the same time, it has the characteristics of rich ecology, convenient migration, extremely low cost of operation and maintenance and five nine high availability. Customers cover the fields of games, Internet, government affairs, finance, manufacturing and Internet of Things.