Google today announced the General Availability (GA) of Compute Engine. Launched during the Google I/O in 2012, Google Compute Engine (GCE) has been in beta for more than year. During initial days, I was skeptical of Google’s commitment and its ability to rapidly innovate to meet the pace of AWS. Today GCE is fully evolved to become a viable alternative to Amazon EC2.
Some of the highlights of GCE include the following –
- Initial reports suggested that GCE performs better than Amazon EC2 and Windows Azure. Google focused on making sure that the CPU, disk I/O and network I/O are optimized to deliver the best possible performance.
- GCE offers the widest choice of instance types. Across 4 families, customers can choose from 15 different instance types that vary by the number of CPUs, memory and storage.
- Networking is a first class citizen of GCE. It is possible to create virtual networks and subnets that are independent of VMs. Google leverages its worldwide network to offer better throughout and lower latency rates.
- GCE’s load balancer doesn’t require pre-warming like Amazon’s ELB. Recently Google demonstrated that they could scale the load balancer to support over a million requests.
- GCE supports all the popular distributions of Linux including Red Hat Enterprise Linux. This will attract many enterprise LAMP customers.
- Google’s Live Migration technology minimizes the maintenance window of the VMs. It can automatically restart faulty VMs to restore the health.
- GCE’s persistent disks are designed to deliver faster IOPS at a cheaper price than the competition.
- Google is partnering with the ecosystem through the support of Right Scale, Scalr, Qubole , Wowza and others.
Google’s official entry into the IaaS market marks a milestone for the customers. Looking back at the progress that Google made with GCE in the last few months, it is clear that they want to compete with AWS and Microsoft to get their stake in the cloud services.