ra2 studio - Fotolia
Customers get new dual-region Google Cloud Storage option
Google updated its cloud storage with higher availability SLAs and a new dual-region option designed to give customers greater control over their data.
Google today updated its cloud object storage with new dual-regional options and higher availability service-level agreements. Google claims the new options will further differentiate its offerings from market leaders Amazon and Microsoft Azure.
The dual-regional Google Cloud Storage option gives customers greater control over where they store their data, while providing geographic redundancy in the event of an outage. The previous default for a Google Cloud Storage object was a larger multi-regional bucket. Google also offers customers the choice of a single-regional bucket.
Dual-region Google Cloud Storage option locations
The dual-regional Google Cloud Storage option is currently available in beta-test mode for U.S. data centers in Iowa and South Carolina, and European Union (EU)-based data centers in the Netherlands and Finland. Dominic Preuss, a Google director of product management, said the public cloud provider typically beta tests new capabilities for about two months. He said Google plans to add more dual-region pairs based on customer demand.
“If you wanted to do this on another public cloud vendor, you would have to put your data into a region, set up replication, and pay network traffic to replicate that data over,” Preuss said. “You wouldn’t be able to predict what the cost was going to be unless you understood all the network connectivity.”
Preuss said Google dual-region customers pay a single price because they do not need to set up replication and pay for storage in two locations plus network traffic fees. He expects the new dual-regional Google Cloud Storage option to be especially helpful with analytics and big data workloads. Those workloads run faster when data is close to compute resources. The company said the dual-regional option should also help customers that use a content delivery network.
Preuss said Twitter is already using the new dual-regional option as it moves more than 300 PB of Hadoop data into Google Cloud Storage. Twitter wanted to ensure the data is geographically redundant from an availability standpoint and also needed to run data processing jobs in the same region where the data is stored, Preuss said. He said Google previously stored copies of the data in multiple regions without telling customers the specific regions where the data was kept.
Availability SLA raised to 99.9%
Google Cloud Storage today also added a higher availability service-level agreement (SLA) for the Nearline and Coldline cloud storage in multi-regional locations in the U.S., EU or Asia. Google is boosting the SLA from 99.0% to 99.9% for its Nearline and Coldline storage. Nearline is for customers who access data less than once a month, and Coldline is for those expecting to access data less than once a year.
“Two nines of availability is not a very high level if you’re using mission-critical data. Getting the three nines of availability is significant for customers that rely on that data to run their businesses,” Preuss said.
He said the company is trying to simplify Google Cloud Storage options by asking customers to choose their desired level of redundancy (multi-regional, dual-regional or regional) and rate of access (Standard, Nearline or Coldline). That will enable them to better predict the cost of using cloud storage, he said.
Also today, Google launched a new C++ client library for use with its cloud storage. Preuss said Google was responding to customer demand, especially from the gaming and oil and gas industries that run large-scale jobs dependent on the C++ programming language. Google also supports Go, Java, .Net, Node.js, PHP, Python, and Ruby.
