GeminiDB Redis Instance Specifications
GeminiDB Redis API provides different product, storage, and instance types. Table 1 lists the specifications for them.
| Deployment Type | Product Type | DB Instance Type | Instance Creation Method | Specifications |
|---|---|---|---|---|
| Classic | Standard | Proxy cluster | Fast configure | Standard Cluster Instance Specifications in Fast Configuration |
| Standard configure | ||||
| Redis Cluster | Fast configure | Standard Cluster Instance Specifications in Fast Configuration | ||
| Standard configure | ||||
| Primary/Standby | Fast configure | Standard Primary/Standby Instance Specifications in Fast Configuration | ||
| Standard configure | ||||
| Capacity-oriented | Proxy cluster | Fast configure | Capacity-oriented Proxy Cluster Instance Specifications in Fast Configuration | |
| Standard configure | ||||
| Cloud native | Capacity-oriented | Proxy cluster | - | Specifications of Capacity-oriented Instances with Cloud Native Storage |
| Performance-enhanced | Proxy cluster | - | Shard Specifications of Performance-enhanced Instances with Cloud Native Storage |
Concepts
- Instance storage: GeminiDB Redis instances facilitate hot and cold data exchanges while offering storage that significantly exceeds the memory limit. Hot data is stored in the memory, and all data is stored in a high-performance storage pool. Shared storage is accessible to any node. The total storage determines the maximum volume of data that can be stored. Table 5 lists details about node memory.
- QPS reference: The QPS values provided are for reference only, based on simple GET/SET tests. For details, see Performance White Paper. QPS is affected by the command complexity, request size, and access model. Actual performance varies depending on the workload.
- Maximum connections: The maximum number of connections varies depending on the product type, instance architecture, and specifications. For details, see the maximum number of connections or total connections per shard in the instance specifications table.
- Node bandwidth: GeminiDB Redis instance nodes are built on ECSs. Node bandwidth scales with higher specifications.
- Redis Cluster instance bandwidth = Node bandwidth × Number of nodes
- Proxy cluster instance bandwidth = Min(ELB bandwidth, Node bandwidth × Number of nodes). The maximum ELB bandwidth is 10 Gbit/s by default.
The overall performance of a GeminiDB Redis cluster instance is related to its CPU. For example, the performance of a six-node instance with 2 vCPUs and 8 GB of memory is the same as that of a three-node instance with 4 vCPUs and 16 GB of memory. A preferential instance with 1 vCPU and 2 GB of memory is not recommended in the production environment due to limited resources, OOM, and node exceptions. The recommended specifications are 1 vCPU and 4 GB of memory or higher.
Instance Specifications in Fast Configuration
If you set Instance Creation Method to Fast configure, you can select the recommended node specifications, quantity, and storage.
The following specifications are suitable for both standard proxy cluster instances and Redis Cluster instances in fast configuration.
| Storage (GB) | Node Flavor | Node CPUs | Node Memory (GB) | Nodes | QPS Reference | Proxy Cluster Instance Bandwidth (Mbit/s) | Redis Cluster Instance Bandwidth (Mbit/s) | Max. Connections | Databases | Database Value Range | ACL Accounts |
|---|---|---|---|---|---|---|---|---|---|---|---|
| 4 | geminidb.redis.medium.2 | 1 | 2 | 2 | 20,000 | 1,600 | 1,600 | 20,000 | 1,000 | 0–999 | 200 |
| 8 | geminidb.redis.medium.4 | 1 | 4 | 2 | 20,000 | 1,600 | 1,600 | 20,000 | 1,000 | 0–999 | 200 |
| 16 | geminidb.redis.large.4 | 2 | 8 | 2 | 40,000 | 2,400 | 2,400 | 20,000 | 1,000 | 0–999 | 200 |
| 24 | geminidb.redis.large.4 | 2 | 8 | 3 | 60,000 | 3,600 | 3,600 | 30,000 | 1,000 | 0–999 | 200 |
| 32 | geminidb.redis.large.4 | 2 | 8 | 4 | 80,000 | 4,800 | 4,800 | 40,000 | 1,000 | 0–999 | 200 |
| 48 | geminidb.redis.xlarge.4 | 4 | 16 | 3 | 120,000 | 7,500 | 7,500 | 30,000 | 1,000 | 0–999 | 200 |
| 64 | geminidb.redis.xlarge.4 | 4 | 16 | 4 | 160,000 | 10,000 | 10,000 | 40,000 | 1,000 | 0–999 | 200 |
| 96 | geminidb.redis.2xlarge.4 | 8 | 32 | 3 | 240,000 | 10,240 | 15,000 | 30,000 | 1,000 | 0–999 | 200 |
| 128 | geminidb.redis.2xlarge.4 | 8 | 32 | 4 | 320,000 | 10,240 | 20,000 | 40,000 | 1,000 | 0–999 | 200 |
| 192 | geminidb.redis.2xlarge.4 | 8 | 32 | 6 | 480,000 | 10,240 | 30,000 | 60,000 | 1,000 | 0–999 | 200 |
| 256 | geminidb.redis.2xlarge.4 | 8 | 32 | 8 | 640,000 | 10,240 | 40,000 | 80,000 | 1,000 | 0–999 | 200 |
| 384 | geminidb.redis.2xlarge.4 | 8 | 32 | 10 | 800,000 | 10,240 | 50,000 | 100,000 | 1,000 | 0–999 | 200 |
| 512 | geminidb.redis.4xlarge.4 | 16 | 64 | 6 | 960,000 | 10,240 | 54,000 | 60,000 | 1,000 | 0–999 | 200 |
| 768 | geminidb.redis.4xlarge.4 | 16 | 64 | 9 | 1,440,000 | 10,240 | 81,000 | 90,000 | 1,000 | 0–999 | 200 |
| 1,024 | geminidb.redis.4xlarge.4 | 16 | 64 | 12 | 1,920,000 | 10,240 | 108,000 | 120,000 | 1,000 | 0–999 | 200 |
| 2,048 | geminidb.redis.4xlarge.4 | 16 | 64 | 22 | 3,520,000 | 10,240 | 198,000 | 220,000 | 1,000 | 0–999 | 200 |
| 4,096 | geminidb.redis.8xlarge.4 | 32 | 128 | 24 | 7,680,000 | 10,240 | 432,000 | 240,000 | 1,000 | 0–999 | 200 |
| 8,192 | geminidb.redis.8xlarge.4 | 32 | 128 | 36 | 11,520,000 | 10,240 | 648,000 | 360,000 | 1,000 | 0–999 | 200 |
The following specifications are suitable for standard primary/standby instances in fast configuration.
| Storage (GB) | Node Flavor | Node CPUs | Node Memory (GB) | Nodes | Shards | QPS Reference | Bandwidth (Mbit/s) | Max. Connections | Databases | Database Value Range | ACL Accounts |
|---|---|---|---|---|---|---|---|---|---|---|---|
| 4 | geminidb.redis.medium.2 | 1 | 2 | 2 | 1 | 8,000 | 1,600 | 10,000 | 1,000 | 0–999 | 200 |
| 8 | geminidb.redis.medium.2 | 1 | 2 | 2 | 1 | 8,000 | 1,600 | 10,000 | 1,000 | 0–999 | 200 |
| 16 | geminidb.redis.medium.4 | 1 | 4 | 2 | 1 | 10,000 | 1,600 | 10,000 | 1,000 | 0–999 | 200 |
| 24 | geminidb.redis.large.4 | 2 | 8 | 2 | 1 | 20,000 | 2,400 | 10,000 | 1,000 | 0–999 | 200 |
| 32 | geminidb.redis.large.4 | 2 | 8 | 2 | 1 | 20,000 | 2,400 | 10,000 | 1,000 | 0–999 | 200 |
| 48 | geminidb.redis.xlarge.4 | 4 | 16 | 2 | 1 | 40,000 | 5,000 | 20,000 | 1,000 | 0–999 | 200 |
| 64 | geminidb.redis.xlarge.4 | 4 | 16 | 2 | 1 | 40,000 | 5,000 | 20,000 | 1,000 | 0–999 | 200 |
| 96 | geminidb.redis.2xlarge.4 | 8 | 32 | 2 | 1 | 80,000 | 10,000 | 20,000 | 1,000 | 0–999 | 200 |
| 128 | geminidb.redis.4xlarge.4 | 16 | 64 | 2 | 1 | 160,000 | 10,000 | 20,000 | 1,000 | 0–999 | 200 |
The following specifications are suitable for capacity-oriented proxy cluster instances in fast configuration.
| Storage (GB) | Node Flavor | Node CPUs | Node Memory (GB) | Nodes | QPS | Proxy Cluster Instance Bandwidth (Mbit/s) | Redis Cluster Instance Bandwidth (Mbit/s) | Max. Connections | Databases | Database Value Range | ACL Accounts |
|---|---|---|---|---|---|---|---|---|---|---|---|
| 256 | geminidb.redis.xlarge.4 | 4 | 8 | 3 | 96,000 | 7,500 | 7,500 | 15,000 | 16 | 0–15 | 200 |
| 512 | geminidb.redis.xlarge.4 | 4 | 8 | 6 | 192,000 | 10,240 | 15,000 | 30,000 | 16 | 0–15 | 200 |
| 1,024 | geminidb.redis.xlarge.4 | 8 | 16 | 6 | 384,000 | 10,240 | 15,000 | 30,000 | 16 | 0–15 | 200 |
| 2,048 | geminidb.redis.2xlarge.4 | 8 | 16 | 12 | 768,000 | 10,240 | 60,000 | 60,000 | 16 | 0–15 | 200 |
| 4,096 | geminidb.redis.4xlarge.4 | 16 | 32 | 12 | 1,536,000 | 10,240 | 108,000 | 60,000 | 16 | 0–15 | 200 |
| 8,192 | geminidb.redis.4xlarge.4 | 16 | 32 | 20 | 2,560,000 | 10,240 | 180,000 | 100,000 | 16 | 0–15 | 200 |
| 12,288 | geminidb.redis.4xlarge.4 | 16 | 32 | 26 | 3,328,000 | 10,240 | 234,000 | 130,000 | 16 | 0–15 | 200 |
Node Specifications of Instances with Classic Storage
| Node Flavor | vCPUs | Memory (GB) | Max. Persistent Storage per Node (GB) | Maximum Connections per Node | Assured Bandwidth (Mbit/s) |
|---|---|---|---|---|---|
| geminidb.redis.medium.2 | 1 | 2 | 4 | 10,000 | 800 |
| geminidb.redis.large.2 | 2 | 4 | 8 | 10,000 | 1,200 |
| geminidb.redis.xlarge.2 | 4 | 8 | 16 | 10,000 | 2,500 |
| geminidb.redis.2xlarge.2 | 8 | 16 | 32 | 10,000 | 5,000 |
| geminidb.redis.4xlarge.2 | 16 | 32 | 64 | 10,000 | 9,000 |
| geminidb.redis.8xlarge.2 | 32 | 64 | 128 | 10,000 | 18,000 |
| geminidb.redis.medium.4 | 1 | 4 | 8 | 10,000 | 800 |
| geminidb.redis.large.4 | 2 | 8 | 16 | 10,000 | 1,200 |
| geminidb.redis.xlarge.4 | 4 | 16 | 32 | 10,000 | 2,500 |
| geminidb.redis.2xlarge.4 | 8 | 32 | 64 | 10,000 | 5,000 |
| geminidb.redis.4xlarge.4 | 16 | 64 | 128 | 10,000 | 9,000 |
| geminidb.redis.8xlarge.4 | 32 | 128 | 256 | 10,000 | 18,000 |
| geminidb.redis.medium.8 | 1 | 8 | 16 | 10,000 | 800 |
| geminidb.redis.large.8 | 2 | 16 | 32 | 10,000 | 1,200 |
| geminidb.redis.xlarge.8 | 4 | 32 | 64 | 10,000 | 2,500 |
| geminidb.redis.2xlarge.8 | 8 | 64 | 128 | 10,000 | 5,000 |
| geminidb.redis.4xlarge.8 | 16 | 128 | 256 | 10,000 | 9,000 |
| geminidb.redis.8xlarge.8 | 32 | 256 | 512 | 10,000 | 18,000 |
Specifications of Capacity-oriented Instances with Cloud Native Storage
| Flavor | vCPUs | Memory (GB) | Maximum Connections per Node | Databases | ACL Accounts |
|---|---|---|---|---|---|
| geminidb.redis-geminifs.large.4 | 2 | 8 | 10,000 | 1,000 | 200 |
| geminidb.redis-geminifs.xlarge.4 | 4 | 16 | 10,000 | 1,000 | 200 |
| geminidb.redis-geminifs.2xlarge.4 | 8 | 32 | 10,000 | 1,000 | 200 |
| geminidb.redis-geminifs.4xlarge.4 | 16 | 64 | 10,000 | 1,000 | 200 |
| geminidb.redis-geminifs.8xlarge.4 | 32 | 128 | 10,000 | 1,000 | 200 |
For specific existing instances, the per-node bandwidth is 768 Mbit/s, which matches other Redis cloud services. If existing instances experience high throughput, you can submit a service ticket on the console to consult on the bandwidth usage.
Shard Specifications of Performance-enhanced Instances with Cloud Native Storage
The following specifications are only available to performance-enhanced instances.
| Shard Specifications | Memory (GB) | Connections Per Shard | Databases | Database Value Range | ACL Accounts |
|---|---|---|---|---|---|
| 8 GB | 8 | 5,000 | 1,024 | 0–1023 | 200 |
| 16 GB | 16 | 5,000 | 1,024 | 0–1023 | 200 |
| 32 GB | 32 | 5,000 | 1,024 | 0–1023 | 200 |
| 64 GB | 64 | 5,000 | 1,024 | 0–1023 | 200 |
| 128 GB | 128 | 5,000 | 1,024 | 0–1023 | 200 |
Feedback
Was this page helpful?
Provide feedbackThank you very much for your feedback. We will continue working to improve the documentation.See the reply and handling status in My Cloud VOC.
For any further questions, feel free to contact us through the chatbot.
Chatbot
