Google announced updates to its cloud platform which should make significant reduction in network latency. What's more it should be faster and more secured, and with higher availability.
The first notable update, is part of Google's cloud core infrastructure. The internet tech giant said that it's updating its Andromeda software-defined network stack, to version 2.1. This update should bring about a 40 percent reduction in latency between Compute Engine virtual machines.
Launched in 2014, Andromeda was Google's big play to merge Android and Chrome OS to make a single unified operating system that would run across phones, tablets and laptops. So here it.serves as the point for Google's cloud, for provisioning, configuring and managing virtual networks.
It’s used to deliver networking services, for example, cloud platform firewalls, routing, and forwarding rules. All of these leverage the underlying internal Andromeda APIs and infrastructure.
In version 2.1, Google software engineer Jake Adriaens said that the release means that the company has managed to reduce network latency for Compute Engine VMs by a factor of eight since the stack was first released.
"This kind of network performance is especially important as more applications move into the cloud and are accessed via web browsers," said Adriaens. "While the headline metric is often bandwidth, network latency is frequently the more important determiner of application performance."
He added that lower latency should be useful for tasks such as financial transactions, advertisements, video, gaming, high-performance computing and in-memory databases.
To lower the latencies, Google bypassed Andromeda's hypervisor, which is a part of the stack that abstracts operating systems and applications away from the underlying hardware so virtual machines can share system resources such as processor cycles, memory space and network bandwidth.
On the previous version, hypervisor was used as a bridge between Andromeda and guest VMs. With the update, guest VMs and Andromeda can communicate directly.
In addition to the Andromeda update, Google also announced updates and general availability for its Cloud Dedicated Interconnect service. First introduced back in September, its Cloud Dedicated Interconnect enables customers to connect to its data centers through enterprise-grade connections with higher availability and lower latency than their existing internet connections.
n addition to that, the service has also been updated to support Cloud Router Global Routing, a feature that allows subnets in GCP to be accessible from any on-premises network through the Google network.
Initially, Google's Cloud Dedicated Interconnect is being offered from four new locations, including Atlanta, Mumbai, Munich and Montreal, in addition to the existing locations detailed on the product page.