Google Cloud Launches E2 Compute Instances to Cut Costs by 31 Percent
Google Cloud introduces the E2 family of general-purpose virtual machines, promising significant savings over older models. The new instances use a custom scheduler to maximize hardware efficiency without throttling performance.
Google Cloud unveils the E2 family of general-purpose compute instances, delivering cost savings of about 31 percent compared to the existing N1 instances. The new virtual machines run on standard Intel and AMD processors while utilizing a custom CPU scheduler that dynamically maps virtual resources to physical hardware to maximize utilization.
This custom scheduler provides sub-microsecond wake-up latencies and faster context switching, giving Google efficiency gains that it passes directly to users. Unlike competing cloud providers, Google promises that E2 VMs sustain high CPU loads without artificial throttling or complicated pricing structures.
The E2 family offers predefined configurations ranging from 2 vCPUs with 8 GB of memory up to 16 vCPUs with 128 GB of memory. Google also introduces smaller E2-based instances similar to its existing f1-micro and g1-small machine types for lightweight workloads that require minimal baseline CPU performance.