This document contains the current resource capacity limits and per-second API request quotas for Memorystore for Redis.
Instance limits
Memorystore for Redis enforces the following usage limits:
| Limit | Value1 | 
|---|---|
| Databases per instance | 16 | 
| Connected clients per instance for Basic Tier | 65,000 | 
| Connected clients per instance for Standard Tier | 65,000 | 
| Connected clients per instance for Standard Tier with read replicas | 65,000 for each node including the primary and read replicas | 
1 For instances that use both in-transit encryption and Redis version 6.x or earlier, connection limits can be lower. For more information, see Connection limits for in-transit encryption.
Regional capacity quota
The default per-project quota for Memorystore for Redis allows you to provision 1 TB of Redis capacity in each region where the service is available.
| Quota | Value | 
|---|---|
| Data provisioned per region | 1 TB | 
For example, if you provision three instances in your project with 300 GB
each in us-central1, you will have consumed 900 GB out of the 1 
TB quota for us-central1. With the default quota, you can provision an
additional 100 GB in us-central1. To request a higher quota for a region,
see Regional capacity quota increases.
Regional capacity quota increases
To request a higher regional capacity for your project, follow these steps:
- Go to the Quotas page in the Google Cloud console.
   
Quotas - From the Services drop down, select Memorystore for Redis API.
 - From the list of quotas, click the checkbox next to the quota labeled Total Redis capacity (GB) for the desired region.
 - Click the Edit Quotas button.
 - In the edit pane, enter your contact information, then click Next.
 - Enter your desired quota, then click Done.
 - Click Submit request.
 
Per-second API requests quota
The per-second quota for Memorystore for Redis API requests/operations is as follows, and is subject to change:
| API Requests Quota | Value | 
|---|---|
| Create Redis instance requests per project per minute | 30 | 
| API Requests per minute | 3,000 | 
Create Redis instance requests per project per minute includes create operations (gcloud redis instances create).
Memorystore for Redis API operations include:
- Get operations (
gcloud redis instances describe) - List operations (
gcloud redis instances list) - Create operations (
gcloud redis instances create) - Delete operations (
gcloud redis instances delete) - Import and export operations (
gcloud redis instances importandgcloud redis instances export) - Update operations (
gcloud redis instances update) - Upgrade operations (
gcloud redis instances upgrade) - Failover operations (
gcloud redis instances failover) 
This quota is for redis.googleapis.com requests/operations, not OSS Redis protocol/commands. These limits apply to each Google Cloud console project and are shared across all applications and IP addresses using that project.