GeminiDB

Compatibility

Compatibility with the Redis, Cassandra, DynamoDB, and InfluxDB protocols

Scalability

Decoupled storage and compute means they can be scaled independently. Clusters can be scaled quickly, with no data migration required.

Reliability

Enterprise-class data consistency, fault reconstruction, and backup and restoration

Cost-effectiveness

Low storage costs with support for adaptive compression and tiered storage of hot and cold data

Why Huawei Cloud GeminiDB?

Scalability

Scalability

  • Shared-nothing and shared storage architecture: Stateless compute nodes share a stateful distributed storage engine.

  • High resource utilization: Built-in proxy clusters support dynamic data load balancing.

  • Elastic scaling: Compute nodes can be added in minutes and storage can be scaled up in seconds.

Reliability

Reliability

  • No data loss & Strong data consistency: Three copies of data are stored in a distributed shared storage pool, ensuring zero data loss and strong consistency for multi-point access.

  • High availability: Superlative fault tolerance (N-1 reliability) ensures that if a node is faulty, services can be taken over in seconds with little impact on services.

  • Stable performance: GeminiDB has stability designed in, from end to end, with distributed multithreading at the compute layer and remote direct memory access (RDMA) at the storage layer. The RocksDB storage engine is optimized, and cold and hot data separation accelerates hotspot data access.

Cost-effectiveness

Cost-effectiveness

  • Dedicated compute: With decoupled storage and compute, GeminiDB leverages dedicated compute nodes to boost computing performance and improve the compute resource utilization of standby servers that use traditional networks.

  • Low memory cost: GeminiDB Redis API has less demanding memory requirements than open-source Redis, which reduces hardware costs by as much as 75% to 90%.

  • Low storage cost: For time series data, with the support for adaptive compression and automated hot and cold tiered storage, GeminiDB Influx API requires you to pay only 1/10 of the storage cost of a relational database.

Comprehensive Solutions

Comprehensive Solutions

Key-Value
Data Characteristics

Key-value data and traditional Redis data structures like hashes, lists, sorted sets, sets, and streams

Mass data

Application Scenarios

Ads: To store feature profiles, AI services require low-latency and key-value (KV) databases that can hold lots of data. Media terminals also require low latency. GeminiDB Redis API is a great fit for advertising services using real-time advertising (RTA) and on the application platform, demand-side platform (DSP).

Gaming: GeminiDB Redis API helps game companies store player data, activities, rankings, friends, and more. It helps them respond quickly to data issues with Point-In-Time Recovery (PITR).

E-commerce: GeminiDB Redis API enables users to store massive volumes of inventory product data and provides excellent reliability.

Compatibility

GeminiDB Redis API is compatible with Redis APIs and supports primary/standby and cluster architectures.

GeminiDB Redis API is compatible with DynamoDB APIs.

Wide Tables
Data Characteristics

Wide table model, column fields with scalability, and structured and semi-structured data

Petabytes of data

Application Scenarios

Social networking: GeminiDB Cassandra API can store user data such as their followers, posts, and likes, and it can handle content distribution data for social networking applications. It can smoothly scale up to handle traffic surges when there are hot topics trending on applications such as Facebook, Discord, and MeeTime.

IoT: GeminiDB Cassandra API handles concurrent writes of massive data generated by IoT devices, including data about logistics, wearables, and smart city monitoring.

Data warehouse: GeminiDB Cassandra API can re-analyze massive data about advertisements, historical orders, feeds, messages, logs, and recommendations to form new user and behavior profiles.

Compatibility

GeminiDB Cassandra API is compatible with open-source Cassandra, with a development experience similar to that of MySQL and supporting cluster architectures.

GeminiDB Cassandra API is compatible with open-source HBase (in OBT).

GeminiDB Cassandra API is compatible with DynamoDB APIs and data can be imperceptibly migrated from DynamoDB to GeminiDB Cassandra API.

Time Series Data
Data Characteristics

Time series data with timestamps

Simplified data types including boolean, int, string, and float

A large volume of data which is mainly written and queried and seldom updated

Stable data traffic without obvious fluctuations between peak and off-peak hours

Structured data and data generated from a single data source

Data which is stored by time and deleted upon expiration

Application Scenarios

IoV: information such as vehicle status, location, and battery status

IoT: reported status data such as factory devices, terminals, and environments

Application O&M and monitoring: CPU, memory, and disk usage of VMs and containers

Business intelligence: aggregation calculating and analysis of data such as service records and user behaviors

Compatibility

GeminiDB Influx API is compatible with InfluxDB APIs and supports cluster architectures.

Continuous Innovation with Millions of Customers

Continuous Innovation with Millions of Customers

GeminiDB APIs

GeminiDB APIs