Google Cloud apps are getting a speed boost -- or more precisely, significantly reduced latency -- with an upgrade to the company's Andromeda software-defined networking (SDN) stack announced Thursday.

Andromeda connects all the parts of Google (Nasdaq: GOOG) Cloud Platform, and version 2.1, announced Thursday, reduces network latency between Compute Engine virtual machines by 40% over version 2.0, writes Jake Adriaens, Google Cloud software engineer, in a post on the Google Cloud Platform Blog.

Low latency is important as applications move to the cloud and are accessed through web browsers, Google says. "While the headline metric is often bandwidth, network latency is frequently the more important determiner of application performance. For example, low latency is essential for financial transactions, ad-tech, video, gaming, and retail, as well as workloads such as [high performance computing] applications, memcache and in-memory databases. Likewise, HTTP-based microservices will see significant improvement in responsiveness with reduced latency."

Google claims to have reduced latency eight-fold since it launched Andromeda in 2014.

Figure 1:

Keep up with the latest enterprise cloud news and insights. Sign up for the weekly Enterprise Cloud News newsletter.

The company explains technical details on its blog post.

Google had another networking announcement this week, launching Google Cloud Dedicated Internet into general availability. That service provides high-speed, reliable connections between enterprise on-premises networks and the Google Cloud Platform. (See Google Hybrid Cloud Networking Gets Production-Ready.)

And Google launched Network Service Tiers in August, connecting enterprise cloud applications to customers. (See Google Offering Tiered Networking for Cloud Customers.)

Related posts:

— Mitch Wagner Editor, Enterprise Cloud News