Supercharge your data the open-source way: Memorystore for Valkey is now GA

Supercharge your data the open-source way: Memorystore for Valkey is now GA

Editor’s note: Ping Xie is a Valkey maintainer on the Valkey Technical Steering Committee (TSC).


Memorystore, Google Cloud’s fully managed in-memory service for Valkey, Redis and Memcached, plays an increasingly important role in our customers’ deployments — in fact, over 90% of the top 100 Google Cloud customers use Memorystore. Today, we’re excited that the Memorystore for Valkey service is now generally available, a significant step forward for open-source in-memory data management on the cloud. With the GA, you can now run your production workloads on Memorystore for Valkey backed by a 99.99% availability SLA along with features such as Private Service Connect, multi-VPC access, cross-region replication, persistence, and many more.

When we launched the preview of Memorystore for Valkey in August 2024, hundreds of Google Cloud customers like Major League Baseball (MLB) and Bandai Namco Studios Inc. jumped in and deployed the service. In the last few months, they’ve provided us with invaluable feedback that has shaped the service we’re announcing today:

“At Major League Baseball, our use of Memorystore has been a key part in optimizing how we bring data to our fans. We are excited about the general availability of Memorystore for Valkey, a truly open-source alternative. We believe its inherent flexibility and the power of community-driven development will further enhance our speed, scalability, and real-time data processing capabilities, allowing us to better serve our fans, players, and operations.” – Rob Engel, Vice President of Software Engineering, Major League Baseball

“Bandai Namco Studios uses Memorystore to power the low-latency and high-scale performance essential for many of our titles. We’re excited about the GA launch of Memorystore for Valkey. Its speed, features, and truly open-source nature will empower us to enhance real-time gameplay and scale for our global player base. We look forward to leveraging Memorystore for Valkey’s capabilities to continue pushing the boundaries of gaming innovation.” – Motoo Fukuda, Technical Director at Bandai Namco Studios Inc.

aside_block
<ListValue: [StructValue([('title', '$300 in free credit to try Google Cloud databases'), ('body', <wagtail.rich_text.RichText object at 0x3e85aab91790>), ('btn_text', 'Start building for free'), ('href', 'http://console.cloud.google.com/freetrial?redirectPath=/products?#databases'), ('image', None)])]>

What’s new at GA

At GA, Memorystore for Valkey is backed by a 99.99% SLA powered by Google’s advanced high availability and zonal placement algorithms, and ships with a comprehensive suite of enterprise-grade features such as:

  • Support for Private Service Connect: Memorystore for Valkey is built on top of Private Service Connect, which allows customers to connect to up to 250 shards using just two IP addresses. Memorystore’s discovery endpoint being highly available ensures no single point of failure for your cluster.

  • Zero-downtime scaling: Memorystore for Valkey offers zero downtime scaling (in and out) so your cluster can grow with your application’s needs, and so it’s cost-optimized for your  workloads. It supports cluster sizes ranging from 1 to 250 nodes.

  • Integrated Google-built vector similarity search: Memorystore for Valkey supports ultra-low latency, in-memory vector search, and can perform vector search at single-digit millisecond latency on over a billion vectors, with greater than 99% recall.

    This performance is powered by Google’s vector search module, the official search module for the Valkey OSS project, which is integrated into Memorystore for Valkey. The module enables modern AI applications for gen AI use cases such as retrieval-augmented generation (RAG), recommendation systems, and semantic search. With hybrid search support, users can achieve more accurate and contextually relevant search results, leading to improved application performance and a better user experience.

  • Managed backups: Access to built-in managed backups enables both automated and on-demand backups for migrations, disaster recovery, and compliance.

  • Cross-region replication (CRR): Using CRR, you can achieve disaster recovery prepared-ness and low-latency reads across regions. At this time, in addition to the primary region, we support up to two secondary regions with clusters that in turn can have varying numbers of replicas. Memorystore for Valkey ensures both the data plane and control plane remain in sync across regions.

  • Multi-VPC access: Memorystore for Valkey supports multiple client-side VPCs to connect to one Private Service Connection endpoint on the Valkey cluster. Using this technology, you can securely connect clients across multiple projects and VPCs.

  • Persistence: Memorystore for Valkey offers both RDB-snapshot and AOF-logging based persistence to meet varying data durability requirements.

Memorystore for Valkey supports both Valkey 7.2, and our engine of choice, Valkey 8.0, which offers many enhancements over its predecessors:

  • Exceptional performance: With asynchronous I/O improvements, Memorystore for Valkey 8.0 delivers better throughput and achieves up to 2x Queries Per Second(QPS) of Memorystore for Redis Cluster at microseconds latency, helping applications handle demanding internet-scale workloads with ease.

    While priced in-line with Memorystore for Redis Cluster, Memorystore for Valkey’s performance optimizations can lead to substantial cost savings by potentially requiring fewer nodes to handle the same workload.

  • Optimized memory efficiency: Valkey 8.0’s optimized memory management delivers improved memory savings, reducing operational costs across various workloads.

  • Enhanced reliability: Valkey 8.0 offers significantly more reliable scaling with Google-contributed features like automatic failover for empty shards and highly available migration states. Additionally, we also introduced migration states auto-reparing to further strengthen system resilience.

In addition, Memorystore for Valkey also provides other capabilities, such as  maintenance windows, single zone clusters, single shard clusters, no-cost inter-zone replication, etc. 

Our commitment to open source and customer trust

Following licensing updates to Redis OSS by Redis Inc. in March 2024, the open-source community established Valkey OSS as an alternative that’s supported by organizations including Google, Amazon, Snap and others.

We deeply value the trust you place in us. To ensure you continue to have access to powerful, open technology, we launched Memorystore for Valkey on Google Cloud. Unlike Redis, the Valkey OSS project is under the BSD 3-clause license and backed by the Linux Foundation. The momentum behind Valkey has been exhilarating. 

In addition to Memorystore for Valkey, we are also committed to supporting and delivering new capabilities for Memorystore for Redis Cluster and Memorystore for Redis. And when Memorystore for Redis customers are ready to adopt Valkey — for its price-performance, reliability and open-source nature — we offer full migration support. Memorystore for Valkey is fully compatible with Redis OSS 7.2 APIs and your favorite clients, making it easy to switch to open source. Further, you can reuse your Memorystore for Redis and Memorystore for Redis cluster committed use discounts (CUDs), smoothing the transition.

Try Memorystore for Valkey today

The best way to experience the power of Memorystore for Valkey is to try it out. Get started with the documentation or deploy your first Valkey instance. Don’t let having to self-manage Redis hold you back. Experience the simplicity and speed of Memorystore for Valkey today and see how it can power your applications, so you can focus on what matters: innovating and creating impactful applications for your business!