A fully managed in-memory data store service for the Redis high-availability cloud data service that Google first announced in April is now accessible in beta for any organization that wants to try it out.
Google's new Cloud Memorystore is a highly scalable and highly available Redis service that is fully hosted and managed by the company. It is designed to provide fast, in-memory data caching for enterprises with Google cloud-hosted applications that need to process data in near real-time.
Google's hosted Redis service is fully compatible with open source Redis. So organizations running applications on the open source version of Redis can now migrate their applications to Google's Cloud Platform without any code changes, said Google product manager Gopal Ashok in a blog May 9.
"As more and more applications need to process data in real-time, you may want a caching layer in your infrastructure to reduce latency for your applications," Ashok wrote. With Google's managed Redis service, organizations can get fast in-memory caching and features like persistence, data replication and real-time messaging, he noted.
The open-source Redis is an in-memory data technology developed by Redis Labs that organizations can use as a database, data cache or message broker. Unlike traditional relational databases that write to disk, Redis offers a way to store or to cache data in memory, thereby enabling much faster data access for applications that need it.
Google's Cloud Memorystore is designed to give organizations a way to take advantage of Redis without having to worry about managing the technology. Organizations can use the service to build large application caches with sub-millisecond access times running on infrastructure totally managed by Google.
Unlike open source Redis, with Cloud Memorystore all Redis instances are protected via a private IP address and enterprises can enforce access management rules via Google's Cloud Identity and Access Management service, Ashok said. With Cloud Memorystore, enterprises also can use Google's Stackdriver cloud management and monitoring suite to monitor all Redis instances, he said.
Amazon offers a nearly identical service with its ElastiCache for Redis managed service. Like Google's Cloud Memorystore, Amazon's service is targeted at enterprises running data-intensive applications in the cloud. It provides companies with a way to store or cache application data in high-throughput, low latency in-memory data stores. Amazon has described ElastiCache as being popular among gaming companies, ad-tech firms, financial services companies and healthcare organizations.
Google is offering two service tiers with Cloud Memorystore—a Basic service tier and a Standard tier. The Basic tier is designed for organizations that want to use Redis for relatively simple data caching, while the Standard tier is meant for those that want high-availability guarantees around their Redis infrastructure.
Enterprises that sign up for either the Basic tier or the Standard tier can provision Redis instances of up to 300 GB and get network throughput of up to 12 Gbps, Ashok said. But those who sign up for the higher-end standard service also get high-availability features such as cross-zone replication and automatic failover capabilities. Standard tier customers of Cloud Memorystore also get a 99.9 percent service availability guarantee.
Like many of Google's cloud services, Google is offering an on-demand, per-second pricing model for Cloud Memorystore. There is no upfront cost associated with the service.