This document introduces the relevant terms and concepts in the TencentDB for MongoDB help documentation to help you better understand the product features.
Instance
An instance is an independently running database environments in Tencent Cloud and are the basic units for purchasing the MongoDB service. Each instance is run as a separate process, and one database instance can contain multiple databases created by users. You can create, modify, and delete instances in the console. Instances are isolated from each other with dedicated resources, ensuring there is no CPU, memory, or I/O preemption across instances.
Region
A region refers to the geographical location where the servers of users' MongoDB instances reside. Each region is completely independent. When purchasing MongoDB instance resources, you need to specify the region closest to you to minimize access latency. Note that changing the region is not supported after the purchase. Additionally, TencentDB for MongoDB instances need to be combined with Cloud Virtual Machine (CVM) instances; it is required to ensure that they reside within the same region.
Availability Zone
An availability zone refers to physical data center within the same region provided by Tencent Cloud, but with separate power supplies and networks. The private networks between AZs are interconnected, the network latency between AZs is low, and fault isolation (excluding large-scale disasters or power outages) can be achieved between AZs. The isolation can prevent fault propagation, allowing users' businesses to provide a continuous online service.
Replica Set
A replica set is the core architecture for achieving MongoDB high availability. Each replica set consists of one primary node and multiple secondary nodes. The primary node handles all write operations and synchronizes data to secondary nodes through the asynchronous replication mechanism, achieving data redundancy across nodes. When the primary node fails, the cluster automatically triggers a second-level failover based on the election protocol, promoting a new primary node to ensure continuous service and business continuity. Secondary nodes can be configured as read-only nodes to share query load, enabling read-write separation.
Sharded Cluster
A sharded cluster is the horizontal scaling solution of MongoDB for handling massive data. Data is partitioned horizontally and stored in multiple shards through a shard key. The architecture consists of three layers: the routing layer (mongos), which acts as the unified access point that parses queries and routes them to the target shards; the configuration layer (Config Server), which achieves centralized management of shard rules and data distribution metadata; and the data layer (Shard), which is composed of multiple independent shards. Each shard is essentially a three-node replica set storing data subsets. This architecture supports dynamic shard addition for elastic scale-out and automatic balanced data distribution, overcoming stand-alone hardware limitations. It is suitable for large-scale scenarios such as the Internet of Things (IoT) and social platforms. For architecture details, see Cluster Architecture. mongod
Mongod is the core background service (daemon) process of the MongoDB system. It is the executable program responsible for running the database service and handling data storage and processing.
mongos
Mongos is the abbreviation for MongoDB shard and serves as the core routing service of a MongoDB sharded cluster. It acts as an intelligent gateway between the client and the cluster. Applications connect to mongos (instead of directly accessing shards), which centrally handles all read-write requests. Based on the shard key, mongos can automatically locate the shard where the data resides and route queries and write operations to the target node. It also merges the return results of multiple shards to provide a transparent and consistent query view for the client.
Shard
In a MongoDB sharded cluster, shards are independent subsets used for data storage. The cluster partitions data into multiple chunks based on the shard key and distributes them evenly across the shards.
Config Server
Config Server is the central metadata repository for a MongoDB sharded cluster. It is responsible for maintaining the routing rules, shard topology, and data distribution information of the cluster, which are automatically synchronized to all mongos routing nodes, ensuring accurate routing of query requests.
Tencent Cloud Console
The Tencent Cloud console is the unified web operation panel for Tencent Cloud resources. By logging in to the console, you can visually manage and monitor the full lifecycle of database instances.
Basic Network
A basic network is a public network space used by multiple users. It cannot be segmented. IP addresses are unique, randomly assigned, and cannot be modified.
Virtual Private Cloud (VPC) is the logically isolated network space customized by you on Tencent Cloud. You can create a logically isolated and user-defined network space for resources such as cloud servers and cloud databases to enhance the security of your cloud resources and meet the requirements of different use cases. You can learn more about VPC through the following video.
A security group is a virtual firewall that can filter the stateful data packets and set network access control for CVM, Cloud Load Balancer (CLB), and cloud database instances to control inbound and outbound traffic at the instance level. They serve as an important means of network security isolation.
Number of Connections
The number of connections refers to the number of TCP session channels established between the clients and the MongoDB instance, directly reflecting the concurrent access pressure. The upper limit of it is strictly constrained by the memory specification of the instance. Reusing the connection pool can significantly improve resource utilization.
Tag
A tag is used to classify and aggregate resources. If multiple types of TencentDB for MongoDB resources exist under your account with various correlations, tags allow you to classify and group resources that serve the same purpose or are associated with each other. This facilitates quick search of resources, batch operations, and efficient troubleshooting during routine Ops or issue location.
A project is the core logical unit of cloud resource management. By aggregating cloud products (such as databases, storage, and computing) into the same space, it enables resource grouping and isolation, unified permission management, and cost allocation accounting, providing infrastructure for team collaboration.
Tencent Cloud CVM is a scalable computing service provided by Tencent Cloud. Access to TencentDB for MongoDB is achieved through private network addresses automatically assigned by the CVM connections.
Cloud Access Management (CAM) is a user and permission management system provided by Tencent Cloud, which is designed to help customers securely and finely manage access to Tencent Cloud products and resources.
Data Transfer Service (DTS) is a comprehensive database data transfer service that integrates data migration, synchronization, and subscription. It enables users to complete database migration seamlessly without business interruption.
Slow Log
The MongoDB slow log is a diagnostic log that records all operations whose execution time exceeds a set threshold (such as 100 ms) and their details.
Rollback
Based on the latest full backup and subsequent Oplog (MongoDB) or binlog (MySQL) replay, this feature allows users to recover specified databases, tables, or the entire instance to any second within the past 7 days, minimizing losses caused by accidental database operations.
TencentDB for Redis® instances integrate the intelligent diagnosis and optimization feature of TencentDB for DBbrain (DBbrain). It provides real-time monitoring and diagnosis of the database instance exceptions, automatically generates health reports, and provides expert-level optimization suggestions. DBbrain helps you comprehensively master the current running status of your database, quickly troubleshoot issues, and optimize database performance promptly.