The public cloud is no longer a hard sell. According to research firm IDC, worldwide spending on public cloud computing recently approached $70 billion. As befits a market that size, channel partners now have options when it comes to public cloud providers for infrastructure- and platform-as-a-service environments. Among the top providers are Amazon Web Services (AWS), Google and its Google Compute Engine, Microsoft and its Azure platform, and open source organization OpenStack. It’s important to evaluate each service provider for the best fit. In doing so, partners need to go beyond price comparisons and instead look at the offerings in compute, storage, networking, and database capabilities.
Amazon Web Services
AWS is a leader in public cloud computing, having been founded 10 years ago. “AWS is most successful because it has first-mover advantage,” says David Hoff, CTO of Cloud Sherpas, a cloud services and advisory company in Atlanta, now part of Accenture. “With comprehensive features and functionality, AWS caters to traditional IT managers and systems admins.”
At last count, AWS offered 40-plus infrastructure and platform services ranging from compute, storage, and networking, to analytics, enterprise applications, and mobility. “With the immense number of offerings, AWS is readily available to satisfy a range of needs,” Hoff says. AWS can allow channel partners to “lift and shift a data center without dramatic architectural changes.”
AWS also gets high marks for ease of migration. “One of the biggest pros of AWS is the ease of getting up and running,” says Philip Flesher, founder and president of Ninbitsu, a boutique software firm in Austin, Texas. For example, Flesher says load balancing can be done with the click of a button. In addition, “AWS has excellent documentation and APIs,” he says.
As Flesher sees it, AWS is something of an 800-pound gorilla—a status that is both a blessing and a curse. “There is a temptation to get locked into AWS with all the services available,” he explains. “AWS is proprietary and is not backward-compatible; moving services off may require reengineering.”
When it comes to name recognition, Google is on par with AWS, thanks to its ubiquitous search engine. It is that search engine (as well as other applications) that provides the foundation of the Google Compute Engine (GCE), the company’s infrastructure-as-a-service (IaaS) offering. With GCE, virtual machines (VMs) can be spun up on demand and run in Google’s data centers. Flesher believes Google offers superior performance compared with AWS, and “you get more from your money.” After a minimum 10-minute charge, Google bills in minute increments, enabling a pay-as-you-use pricing model.
While AWS is way ahead in terms of a public cloud footprint, Google is making a big push. “The rate of features and functionality Google has added is pretty intense,” says Hoff, “making GCE best suited for organizations that have embraced the transformation to a digital enterprise.” Think of Google as ideal for Web-scale technologies, Android applications, and big data. The vendor has developed big data tools that are designed to handle massive volumes and petabyte-type workloads. “The BigQuery analytics service and [Cloud] Dataflow can be used to provide big data as a service,” Hoff says. This is just one example of how Google aims to capitalize on its data processing legacy.