Google Cloud Platform has rolled out a Beta version of Dedicated Interconnect – a networking service that connects customer on-premise networks directly to Google’s own network for the purposes of hybrid cloud computing.

Dedicated Interconnect offers private connections with high bandwidth and predictable performance, similar to Microsoft Azure ExpressRoute or AWS Direct Connect.

The service is aimed at Google’s largest corporate customers and is one of the more advanced networking products offered by the company.

More choice

Google is not the largest cloud provider in the world, but it runs the world’s largest cloud network. Dedicated Interconnect enables customers to use this network to link their on-premise IT resources to Google Compute Platform and their virtual private cloud network.

The service is available through one of Google’s Points of Presence – usually located in large colocation data centers in major connectivity hubs, like San Francisco, Dallas, London, Amsterdam and Singapore.

It promises better performance and could even help save on costs when compared to services running over public Internet, if the amount of data transferred is large enough.

Dedicated Interconnect is available in 10Gb/s increments – a single interconnect can include up to eight circuits, resulting in capacity of up to 80Gb/s. It can be configured to offer a 99.9 percent or a 99.99 percent uptime SLA, as long as the customer pays for additional redundant connections.

Google notes that the circuit between a customer’s network and Google’s own network is not encrypted by default, so those sending sensitive information would need to consider their own application-level encryption.

Last month, Google introduced a new, cheaper networking option for its cloud products, offering to deliver Google Cloud Platform over the public Internet. The service, currently in Alpha, was dubbed Standard Tier, while Google’s original high-performance network became Premium Tier.