Google unveiled new database capabilities and partnerships at the recently-concluded Google Cloud Next conference. This included expanded beta access for Cloud Firestore, new functionality for Cloud Bigtable, and highly-requested features for Cloud Spanner. There were also announcements around expanded support for SAP and Oracle workloads.
Google unveiled new database capabilities and partnerships at the recently-concluded Google Cloud Next conference. This included expanded beta access for Cloud Firestore, new functionality for Cloud Bigtable, and highly-requested features for Cloud Spanner. There were also announcements around expanded support for SAP and Oracle workloads.
Google Cloud has a broad portfolio of first-party database services. Customers get managed MySQL and PostgreSQL databases via Cloud SQL. Google Cloud Spanner, released last year, is Google’s modern take on relational databases. NoSQL column storage is available with Cloud Bigtable, while Cloud Datastore offers a NoSQL document database. Cloud Firestore represents the next generation of Cloud Datastore. Developers use Firebase Realtime Database for a variety of mobile-centric use cases. And BigQuery—which got machine learning-infused updates at the conference—provides a managed data warehouse. Add all this to the ninety-plus third-party databases in the Google Cloud marketplace, and users aren’t short of options.
Cloud Firestore has been in limited beta until now. It’s billed as “the next major version of Cloud Datastore and a re- branding of the product.” In a blog post, Google Cloud’s VP of Engineering Brad Calder announced that Cloud Firestore, while still in beta, is coming to more users. He also shared that Cloud Firestore will offer a compatibility mode for existing Cloud Datastore users. With this new “Datastore Mode”, the Datastore user doesn’t need to change their apps to use the performance and consistency of Firestore. The conference talk ” Building Scalable Apps with Cloud Firestore ” shared extra Firestore news. The speakers revealed that Cloud Firestore is coming to new locations in the Americas, Europe, and Asia-Pacific. They also shared a 99.999% availability SLA for multi-region instances.
Bigtable development started in 2004, and it underpins many of the core Google services. It shipped as a publicly-accessible service on Google Cloud in 2015. Last week, Google announced the availability of regional replication, in beta since May 2018. Users can instantiate linked clusters when provisioning Bigtable instances, or, add a replicated cluster to an existing instance. Google replicates more than just data; new or deleted tables, added or removed column families, and changes to a column’s garbage-collection policy all get replicated too. In addition, Google added a new “key visualizer” function that’s automatically enabled for Cloud Bigtable instances. This feature shows a heat map of access patterns for use in troubleshooting performance issues. Finally, Google introduced client libraries for Cloud Bigtable. C# and Node.js are in beta, with Python, C++, Ruby, Java, and PHP on the way.
Cloud Spanner is the CAP theorem-busting relational database that promises strong consistency, high availability, and automatic replication. Google recently added import/export functionality aimed at simplifying disaster recovery, among other things. At the conference, Google previewed a data manipulation language (DML) for Cloud Spanner that makes it easier to use standard drivers and ORM tools. Google presenters also shared that query stats were on the way.
The Google Cloud team also teased further partner integrations. Google, Intel, and SAP are teaming up on higher performing and cheaper SAP HANA workloads.
This week we announced our collaboration with Intel and SAP to offer Compute Engine virtual machines backed by the upcoming Intel Optane DC Persistent Memory for SAP HANA workloads. Google Compute Engine VMs with this Intel Optane DC persistent memory will offer higher overall memory capacity and lower cost compared to instances with only dynamic random-access memory (DRAM). Google Cloud instances on Intel Optane DC Persistent Memory for SAP HANA and other in-memory database workloads will soon be available through an early access program.
Google shared plans for larger VM types to accomodate SAP HANA workloads. In addition to the 4TB machine types available today, they’re working on VMs with 12TB and 18TB of memory.
Google promised new support for Oracle workloads on Google Cloud. Calder says that Google is partnering with managed service providers that will offer managed Oracle workloads on Google Cloud. Customers can use their existing Oracle licenses, while getting the same response time for support as with other Google Cloud services. Details are light at this time. The service is planned for Fall of 2018.
