Google Cloud Memorystore
Key value databases
Database as a service (DBaaS) providers
Database software
NoSQL databases
- Features
- Ease of use
- Ease of management
- Quality of support
- Affordability
- Market presence
Take the quiz to check if Google Cloud Memorystore and its alternatives fit your requirements.
Pay-as-you-go
Free Trial
Free version unavailable
Small
Medium
Large
- Media and communications
- Arts, entertainment, and recreation
- Accommodation and food services
What is Google Cloud Memorystore
Google Cloud Memorystore is a managed in-memory data store service on Google Cloud that provides Redis and Memcached-compatible engines for caching, session storage, and low-latency key-value access. It targets application teams that want to offload provisioning, patching, and operational management of in-memory stores while integrating with Google Cloud networking and IAM. The service offers managed high availability options (engine-dependent) and supports common cache patterns used by web, mobile, and microservices applications.
Managed Redis and Memcached
The service provides managed deployments for Redis and Memcached, reducing the need to operate self-managed clusters. It handles common operational tasks such as instance provisioning, maintenance, and patching within the managed service model. This fits teams that need low-latency caching without adopting a general-purpose NoSQL database.
Low-latency in-memory access
Memorystore is designed for in-memory key-value access patterns such as caching, rate limiting, and session storage. It supports Redis data structures and Memcached semantics, enabling fast reads/writes compared with disk-based databases. This makes it suitable as a complementary layer alongside transactional or document databases rather than a system of record.
Google Cloud integration
Memorystore integrates with Google Cloud networking constructs (such as VPC connectivity) and operational tooling in the Google Cloud ecosystem. It aligns with Google Cloud identity and access patterns and can be monitored using Google Cloud’s observability services. This reduces integration effort for workloads already deployed on Google Cloud.
Not a primary database
Memorystore is primarily intended for caching and transient data rather than durable, long-term storage. While Redis can persist data, the service is commonly used as an acceleration layer, and application designs typically treat it as non-authoritative. Teams needing a system of record generally pair it with a separate database service.
Engine feature constraints
Capabilities depend on the selected engine and tier, and not all Redis ecosystem features are available in every managed configuration. Some advanced Redis deployment patterns (for example, certain clustering or module approaches) may not map directly to the managed service. This can limit portability for applications that rely on specific self-managed Redis extensions.
Cloud and service lock-in
Memorystore is delivered as a Google Cloud managed service, which ties operations, networking, and billing to Google Cloud. Migrating to another environment typically requires re-provisioning and data migration, and operational behaviors may differ from self-managed deployments. Organizations with multi-cloud mandates may need additional abstraction or alternative deployment approaches.
Plan & Pricing
Pricing model: Pay-as-you-go (consumption-based)
Memorystore for Redis (on-demand, per GiB-hour)
- Basic tier (per GiB-hour): M1 (1–4 GiB) $0.049; M2 (5–10 GiB) $0.027; M3 (11–35 GiB) $0.023; M4 (36–100 GiB) $0.019; M5 (>100 GiB) $0.016.
- Standard tier (per GiB-hour): M1 (1–4 GiB) $0.064; M2 (5–10 GiB) $0.054; M3 (11–35 GiB) $0.046; M4 (36–100 GiB) $0.035; M5 (>100 GiB) $0.03.
- Read-replica / Standard Tier instances are billed per-node (capacity × node count). Prices are shown per GiB-hour; billing is per-second (rounded up to nearest second). (See Redis pricing page.)
Memorystore for Redis Cluster / Valkey (node-based rates shown per node-hour)
- Example node prices (per node, per hour, default on Valkey page): shared-core-nano (1.4 GB) $0.0318 / hour; standard-small (6.5 GB) $0.1425 / hour; highmem-medium (13 GB) $0.1923 / hour; highmem-xlarge (58 GB) $0.8581 / hour.
- Backup / AOF persistence: AOF persistence charge (per GB-hour) and backup storage charges apply when enabled (example: AOF default $0.00054795 / GB-hour; backup storage default $0.00011111 / GB-hour). (See Valkey/Redis persistence & backups pricing.)
Memorystore for Memcached (on-demand, node composition)
- vCPU: $0.05 per vCPU per hour (example on-demand rate).
- Memory (per GiB-hour) for nodes: Node ≤ 4 GiB: $0.0044 / GiB-hour; Node > 4 GiB: $0.0089 / GiB-hour.
- Total instance hourly cost = (total vCPUs × vCPU rate) + (total memory GiB × memory rate). Billing is per-second.
Discounts / Commitments
- Spend-based Committed Use Discounts (CUDs) are available: typical example is 20% off for a 1-year commitment and 40% off for a 3-year commitment (applies to eligible Memorystore capacity). Some CUD eligibility/limits apply (e.g., Redis M1 tier <5 GB may be excluded for some CUDs). See the vendor CUD documentation for details.
Free tier / trial
- Free plan: No permanent product-specific free tier for Memorystore for Redis (official docs explicitly state Redis does not provide a free tier).
- Free trial: New Google Cloud customers can use the Google Cloud Free Trial (91 days, $300 in free credits) which can be applied to Google Cloud products, including Memorystore usage covered by the Free Trial. See Google Cloud Free Program docs.
Notes / keys
- Prices vary by region; pages provide region selector and per-region SKUs. Prices above are the USD default consumption model values shown on the official pages.
- Billing increments: charges are incurred in 1-second increments; instances start incurring charges when created (provisioned capacity billed whether used or not).
- Use the Google Cloud Pricing Calculator for estimates.
Seller details
Google LLC
Mountain View, CA, USA
1998
Subsidiary
https://cloud.google.com/deep-learning-vm
https://x.com/googlecloud
https://www.linkedin.com/company/google/
