Google Courts Enterprise For Cloud Platform
March 10, 2017 Jeffrey Burt
Google has always been a company that thinks big. After all, its mission since Day One was to organize and make accessible all of the world’s information.
The company is going to have to take that same expansive and aggressive approach as it looks to grow in a highly competitive public cloud market that includes a dominant player (Amazon Web Services) and a host of other vendors, including Microsoft, IBM, and Oracle. That’s going to mean expanding its customer base beyond smaller businesses and startups and convincing larger enterprises to store their data and run their workloads on its ever-growing infrastructure.
At its Google Next confab this week in San Francisco, Google officials took that to heart, emphasizing throughout the two-day event that despite its reputation as primarily a consumer company, Google Cloud is a business platform.
The search engine giant put a spotlight on such customers as Verizon, Colgate-Palmolive, and Home Depot, partnerships with such enterprise-focused tech firms as SAP, Intel, Rackspace Hosting, and Pivotal, expanded capabilities around everything from security and data preparation to, as we have noted, machine learning. Google also introduced new pricing models and said it was growing the number of datacenters around the world supporting Cloud Platform.
During their keynote addresses, Diane Greene, senior vice president of Google’s cloud efforts, and Urs Hölzle, the company’s senior vice president for cloud infrastructure, argued that Google’s infrastructure was more robust and reliable than its larger competitors (an allusion to Amazon’s recent S3 outage), and Eric Schmidt, executive chairman of Google’s parent company, Alphabet, made the big pitch, telling businesses and developers at the show about the tens of billions of dollars the company has spent over the past three years to build up its infrastructure, saying it is ready to take on their needs.
“We’ve put $30 billion into this platform,” Schmidt said. “Why try to replicate that? We’re here for real. This is an incredibly serious mission. The company has got both the money, the means, and the commitment to pull of a new platform of computation globally for everyone who needs it. Please don’t attempt to duplicate this. You have better uses of your money.”
Google will need all the business customers it can get if it hopes to gain ground on Amazon Web Services (AWS). According to the Synergy Research Group, AWS continues to hold 40 percent of the public cloud services market. By comparison, the next three contenders – Microsoft, Google, and IBM – have 23 percent, and while that percentage is growing – by 5 percent between the fourth quarter of 2015 and Q4 2016 – the growth is coming at the expense of smaller vendors. The market share of AWS remained unchanged during the year.
Among the efforts underway at Google is rapidly expanding the number of cloud data centers around the world. According to Hölzle, the company is adding three new facilities to its lineup – in California, Canada and the Netherlands – by the end of the year, bringing the total number of what Hölzle called “availability regions” to 17 and the number of “availability zones” to 50. The number of data centers (AWS has 16 regions, while Microsoft has 34 with Azure, and both are adding new sites) for cloud service providers is important not only for bringing the infrastructure and services closer to the business customer, but also for backup and disaster recovery. In addition, it ensures that the provider is complying with local regulations around data storage and transmission, security, and privacy.
Google also is targeting enterprises by talking about costs. According to Hölzle, businesses using other cloud service providers waste a lot of money by signing onto three-year leases to get the best pricing, but at the expense of being forced into fixed configurations that have to be paid for even if not all of the resources are used. Cloud Platform unveiled the beta of what he called “committed-use discounts,” a new pricing option for virtual machines. To get the discounts of up to 57 percent, businesses have to agree to one- or three-year commitments, but during that time, they have wide flexibility in the number of cores, amount of memory, and sizes of VMs they want.
“GCP – and only GCP – is a truly elastic cloud,” Hölzle said. “You only buy what you need, you only pay for it when you actually you use it, you get automatic discounts when you use it for extended periods of time, and we even alert you when you appear to be wasting resources.”
Company officials stressed that Google would rely on partnerships with top-tier tech vendors to help build out the cloud infrastructure and grow the services available to business customers. Intel has been optimizing its chips for Google for more than a decade, and in November expanded the partnership to enable the two companies to target the public cloud arena. It’s a partnership that makes a sense for both. For Intel, hyperscale vendors like Facebook, Amazon, Baidu, Tencent, Microsoft, and Google are becoming key buyers of servers and other datacenter hardware and are looking at solutions that can give them the right balance of performance and power and cost efficiency. In terms of chips, that also means finding alternatives to Intel’s X86 processors as a way of driving competition to gain better pricing and innovation. That means not only using Intel Xeon processors, but also looking into options such as GPU accelerators from Nvidia and AMD, ARM systems-on-a-chip (Microsoft this week announced plans to use ARM chips from Qualcomm in servers running in its Azure cloud) and IBM’s Power architecture (Google and Rackspace are collaborating on “Zaius,” a server based on IBM’s Power9 processor).
At Google Next, Google and Intel touted that Google would be the first cloud provider to offer Intel’s upcoming 14 nanometer “Skylake” Xeon processors to its customers, a move that was first announced by Hölzle last month. Google will incorporate the new Xeons in cloud servers in five regions – including parts of the United States, Western Europe, and Eastern Asia Pacific – ahead of rivals and before traditional server OEMs like Dell and Hewlett Packard Enterprise. In a blog post in February, Hölzle noted that “Skylake includes Intel Advanced Vector Extensions (AVX-512), which make it ideal for scientific modeling, genomic research, 3D rendering, data analytics and engineering simulations. When compared to previous generations, Skylake’s AVX-512 doubles the floating-point performance for the heaviest calculations.”
Another key partnership is with enterprise applications vendor SAP. The companies announced that SAP will provide SAP HANA, its in-memory database and analytics engine, on Google’s cloud, an important step for Google as it looks to highlight large enterprises as both customers and partners on its GCP. SAP has been pushing customers to embrace the cloud – the software maker runs its own cloud datacenters – and GCP represents a step into the public cloud arena through a partner. SAP also is offering a developer edition of SAP HANA on GCP, and Bernd Leukert, head of technology and innovation for SAP, said the two companies are partnering together on efforts around machine learning that will be announced at SAP’s user conference in May.
During the show, Google Cloud officials also touted other large enterprise customers, including Verizon and its plans to migrate 115,000 employees to Google’s G Suite of cloud applications and Colgate-Palmolive, which moved 28,000 employees to G Suite last year. Other big customers noted were Home Depot, eBay, and Disney, proof points for GCP officials that their cloud infrastructure is ready for large enterprise workloads.
Security has always been a key concern of businesses as they consider moving workloads and data to the cloud. However, those concerns in recent years have eased as cloud service providers have built up their security capabilities. Google officials put a focus in that area as well, with innovations around both its hardware and its APIs. On the hardware side, the company introduced Titan, a low-power microcontroller built into Google’s cloud servers that creates a security identify for the hardware and creates a root of trust between them and peripherals. Other security features an API to root out and protect sensitive data, software for managing symmetric encryption keys without a hardware security module and a way to use security keys for two-step authentication.
Google also wants to lure Windows customers and developers to its cloud. The company already has made steps in that direction by offering support for ASP.NET applications and pre-configured images on the Google Compute Engine for Microsoft SQL Server Enterprise. Now Google Cloud officials have created a Windows partner program that is designed to encourage and help enterprises migrate their Windows environments to GCP from their on-premises data centers or other cloud providers. In addition, Google officials said SQL Server Enterprise is now generally available on the Google cloud and introduced a beta of .NET for Google’s App Engine and Container Engine.