Skip to main content

Configuration

This guide covers all configuration options available when creating and managing Redis cache clusters on Flash. Understanding these settings helps you optimize your cache's performance, resource usage, and availability to match your specific requirements.

General Settings​

Display Label: The human-readable name shown throughout the Viduli console interface. This can be updated at any time to reflect changes in your cache's purpose or project context.

Resource Name: A unique identifier for your Redis cache cluster within the project. Once created, this name cannot be changed as it's used internally for resource management and networking. Choose a descriptive name that clearly identifies your cache's function.

Cache Configuration​

Engine: The cache engine type, which is Redis for this resource. This setting is read-only and cannot be modified after cluster creation.

Version: The Redis engine version deployed for your cache cluster. This setting is read-only and is selected during cluster creation. Currently available version: 7.2.7.

Compute Resources​

All compute resources are dedicated allocations, ensuring consistent performance and predictable resource availability for your cache operations.

CPU Allocation: The amount of CPU resources allocated to your cache cluster. Higher CPU allocations provide better performance for high-throughput operations and complex data structure manipulations.

Note: CPU can be allocated in fractional units called millicpu, where 1000 millicpu equals 1 full vCPU core.

Memory Allocation: The amount of RAM allocated to your cache cluster. Memory is specified in megabytes and is the most critical resource for Redis performance since all data is stored in memory. Adequate memory allocation ensures optimal cache hit rates and prevents data eviction under load.

Disk Storage: The amount of disk space (in GB) allocated for your cache persistence and backup operations. This includes space for RDB snapshots, AOF logs, and temporary files. While Redis primarily operates in memory, disk storage is essential for data durability and recovery operations.

tip

Cache Optimization: Redis performance is heavily dependent on available memory since all data is stored in RAM. Consider your dataset size and access patterns - ensure sufficient memory to avoid eviction policies triggering under normal load conditions.

Scaling Configuration​

Region Mode: The scaling mode for your cache deployment. This setting is read-only and determines whether your cache operates within a single region or across multiple regions. During the beta period, only single-region mode is available.

Deployment Region: The geographic region where your Redis cache cluster is deployed. This setting is read-only after cluster creation and affects latency for your applications and data residency requirements.

Replication Mode: The replication configuration for your cache cluster. Choose from the following options:

  • Standalone: Single Redis instance with no replication (minimum 1 node)
  • Active-Passive: Primary-replica setup with automatic failover (minimum 2 nodes) Not released yet
  • Cluster: Distributed Redis cluster for horizontal scaling (minimum 3 nodes)

Nodes: Define the number of Redis nodes in your cluster. The minimum and maximum values depend on your selected replication mode:

  • Standalone mode: 1 node (no additional nodes available)
  • Active-Passive mode: 2-6 nodes (1 primary + 1-5 replicas)
  • Cluster mode: 3-12 nodes (distributed across multiple shards)
info

Replication Benefits: Active-Passive replication provides high availability through automatic failover, while Cluster mode enables horizontal scaling and higher throughput by distributing data across multiple shards. Choose based on your availability and performance requirements.

Cost Estimation​

Monthly Cost: Displays the estimated monthly cost based on your current configuration settings. This calculation includes compute resources (CPU and memory), storage allocation, and the number of nodes. Costs are calculated based on dedicated resource allocation and provide predictable billing.

The cost estimation updates automatically as you modify configuration settings, helping you understand the financial impact of your choices before applying changes to your cache cluster.