More than 90% of the top 100 Google Cloud customers utilize Memorystore, Google Cloud’s fully managed in-memory solution for Valkey, Redis, and Memcached. Memorystore is becoming important in client installations. Google Cloud announced that the Memorystore for Valkey service is now widely accessible, marking a major advancement in open-source cloud in-memory data management. The GA now supports Memorystore for Valkey for production workloads, including Private Service Connect, multi-VPC access, cross-region replication, persistence, and more. It also comes with a 99.99% availability SLA.
Hundreds of Google Cloud customers, including Major League Baseball (MLB) and Bandai Namco Studios Inc., stepped in and implemented Memorystore for Valkey as soon it was released the preview in August 2024. They have given us priceless input over the past few months, which has influenced the service they are launching:
Using Memorystore has been crucial to Major League Baseball’s efforts to optimise data delivery to its fans. Google Cloud is that Memorystore for Valkey, a really open-source substitute, is now generally accessible. In order to better serve players, fans, and operations, they think that its built-in flexibility and the strength of community-driven development will further improve speed, scalability, and real-time data processing capabilities.
Memorystore powers Bandai Namco Studios’ high-scale and low-latency performance, which is crucial for many of games. The GA debut of Memorystore for Valkey excites them. Its features, speed, and open-source nature will enable us to expand for worldwide player base and improve real-time gameplay. It’s excited to use Memorystore for Valkey’s capabilities to keep innovating in the gaming industry.
What’s new at GA
At GA, Memorystore for Valkey comes with an extensive set of enterprise-grade capabilities including the following and is supported by a 99.99% SLA driven by Google’s sophisticated high availability and zonal placement algorithms:
Support for Private Service Connect: Since Valkey’s Memorystore uses Private Service Connect, customers may connect to 250 shards with two IP addresses. Because Memorystore’s discovery endpoint is highly available, your cluster won’t have a single point of failure.
Zero-downtime scaling: Memorystore for Valkey offers cost-optimized functionality and zero-downtime scalability to let your cluster grow to meet application demands. Cluster sizes between 1 and 250 nodes are supported.
Integrated Google-built vector similarity search: Memorystore for Valkey supports ultra-low latency, in-memory vector search, and over a billion vectors with 99% recall and single-digit millisecond latency.
The official search module for the Valkey OSS project, Google’s vector search module, which is incorporated into Memorystore for Valkey, is responsible for this speed. The module enables RAG, recommendation systems, and semantic search, among other Artificial Intelligence use cases. Users may obtain more precise and contextually relevant search results with hybrid search capability, which enhances user experience and application performance.
Managed backups: Integrated managed backups automate migration, compliance, and catastrophe recovery.
Cross-region replication (CRR): You may attain low-latency readings across locations and disaster recovery readiness using CRR. Google Cloud now support up to two additional regions with clusters, each of which may have a different number of replicas, in addition to the primary region. Valkey’s memorystore makes sure the control plane and data plane stay in sync across geographical boundaries.
Multi-VPC access: Memorystore for Valkey connects several client-side VPCs to a single Valkey cluster Private Service Connection endpoint. You may safely link customers across several projects and VPCs using this technology.
Persistence: To accommodate different needs for data durability, Memorystore for Valkey provides persistence based on both RDB-snapshot and AOF-logging.
Both Valkey 7.2 and preferred engine, Valkey 8.0, which has several improvements over its predecessors, are supported by Memorystore for Valkey:
Exceptional performance: Applications can easily manage demanding internet-scale workloads with Memorystore for Valkey 8.0’s asynchronous I/O advancements, which boost throughput and reach up to 2x Queries Per Second (QPS) of Memorystore for Redis Cluster at microsecond latency.
Memorystore for Valkey’s speed optimisations might result in significant cost savings by perhaps having fewer nodes to handle the same workload, even if its pricing is comparable to that of Memorystore for Redis Cluster.
Optimized memory efficiency: Optimised memory management in Valkey 8.0 results in better memory savings and lowers operating expenses for a variety of workloads.
Enhanced reliability: With Google-contributed features like automated failover for empty shards and highly available migration states, Valkey 8.0 provides noticeably more dependable scalability. To increase system resilience even more, they also included migration states auto-reparing.
Other features offered by Memorystore for Valkey include maintenance windows, single zone clusters, single shard clusters, free inter-zone replication, and more.
Google Cloud’s commitment to open source and customer trust
After Redis updated Redis OSS’s licensing in March 2024, the open-source community created Valkey OSS as a substitute that is backed by companies like Google, Amazon, Snap, and others.
Your confidence in them is much appreciated. They released Memorystore for Valkey on Google Cloud to make sure you always have access to robust, open technologies. The Linux Foundation supports the Valkey OSS project, which, in contrast to Redis, is licensed under the BSD 3-clause license. It has been to watch Valkey gain momentum.
It focusses on supporting and adding functionality to Memorystore for Redis Cluster, Redis, and Valkey. Additionally, Google Cloud provide complete migration help when Memorystore for Redis clients are prepared to switch to Valkey due to its open-source nature, dependability, and price-performance ratio. The migration to open source is made simple by Memorystore for Valkey’s complete compatibility with Redis OSS 7.2 APIs and your preferred clients. Additionally, you may facilitate the transition by reusing your Memorystore for Redis and Memorystore for Redis cluster committed use discounts (CUDs).
Try Valkey’s Memorystore now
Trying it out is the greatest approach to discover Memorystore’s power for Valkey. Launch your first Valkey instance or begin working through the documentation. Redis self-management shouldn’t be a hindrance. Discover how Memorystore for Valkey can power your apps and how easy and quick it is to use. This will allow you to concentrate on what really matters creating innovative and powerful applications for your company.