Intel takes 'Intel Inside' to the cloud
In 1991, Intel launched the “Intel Inside” program, making its microprocessors a selling point of PC purchases. Over 20 years later, Intel is bringing the same concept to the cloud.
Intel made the case Wednesday that businesses paying thousands of dollars to hosting providers and other cloud services were not actually aware of what processor architectures powered those services.
The “Intel Cloud Technology” program, similar to the stickers advertising Intel’s chips, will attempt to alleviate that shortcoming. Sixteen cloud providers agreed to participate, including Virtustream and Rackspace within the United States, following a similar program launched with Amazon last year.
“Cloud customers want to know what technology their applications are running on because it has direct impact to their business,” said Jason Waxman, vice president of Intel’s data center group, in a statement. “For the first time, users will have the transparency to select the technologies that are optimal for running their applications in the cloud.”
According to Patrick Moorhead, an analyst with Moor Insights, the main component of the initiative is “de-anonymizing” the traditional makeup of a server. Cloud service providers can be deliberately obtuse in describing their hardware, because they want to leverage the purchases they’ve made over a period of years. “I’ve even seen old tower servers at some of these guys,” he said.
Moorhead said that the Intel Cloud Technology program is primarily aimed at small businesses and home business users, who feel more comfortable choosing a chip from Intel versus a new breed of low-cost chip providers.
Intel’s apparent fear is that cloud providers will adopt low-cost, low-power chips from AMD and ARM, designed for specialized workloads. Intel has traditionally powered high-cost, high-power monolithic servers optimized for virtualization.
Increasingly, however, server makers are considering so-called heterogenous computing environments, where specialized low-power servers are used for specific purposes. Companies such as Facebook, for example, are building servers designed to strip out as much cost as possible using collaborative efforts like the Open Compute Project—which is scheduled to host its 2014 conference in San Jose, California, starting on January 28.
“That’s why Intel created the powered by Intel cloud technology program, so that leading cloud providers can assure customers that their workloads are migrating to servers based on Intel processors, just like the servers in their own datacenters,” Intel said in a promotional video accompanying the announcement.
But the real meat of the program is buried somewhat lower down. Intel said that it would “drive direct marketing campaigns and participate on co-marketing activities” with cloud-service-provider partners. That’s traditionally been code for supplying those providers with money that can be used for marketing and other customer acquisition techniques, with the caveat that the cloud providers adopt certain Intel technologies. An Intel spokesman declined to comment on the amount Intel dedicated to the marketing programs.
In all, it’s an easy way for businesses to justify making the “safe bet” and signing on to a cloud provider that uses the same Intel Xeon hardware they’ve always used. “When consumers compare two products, branded and unbranded, they’re far more likely to choose the one with the brand that they know,” Moorhead said.