Intelligent Infrastructure

Mass data and next-gen workloads Intelligent Infrastructure

SoftLayer relies on Supermicro hardware for efficiency, performance

Supermicro is able to meet the diverse needs of one of the largest IT operations on the globe.

Cloud service providers have a wide range of factors to consider as they expand their data centers. While total cost of ownership is a prominent concern, this must be balanced with the unique needs of each vendor’s customer base. Some data center operators decide it is best to specialize in a specific service so that their cloud infrastructure can be tailored to fulfill that niche. However, many companies are finding that meeting a wider range of computing and storage needs can be a competitive advantage.

This is the challenge that cloud storage company SoftLayer has faced as it has expanded its hardware. As The Register reported, the company added 40,000 servers to its data center from 2011 to 2012 – hardware that largely comes from Supermicro, a Seagate Cloud Builder Alliance partner.

“[Supermicro knows] what our purchasing patterns are, they know what configurations have been purchased, and then they show us the best motherboard and system configuration options and we turn that into our standardized products,” SoftLayer COO Sam Fleitman told the news source.

The cloud provider’s 100,000 servers and 13 data centers make it “one of the largest IT operations in the world,” according to The Register. One of the company’s value propositions is the ability to meet the unique needs of its customers. For example, in its early days, SoftLayer’s infrastructure consisted largely of Intel Xeon and Advanced Micro Devices Opteron processors. While demand for AMD hardware has since shrunk, it is still able to provision it from Supermicro based on customer requests.

As the industry places more pressure on cloud providers to expand their service portfolios while keeping costs low, it will be critical to adopt hardware that can balance those seemingly contrasting needs. To this end, Supermicro’s hardware is a powerful asset.

A closer look at Supermicro’s hardware
The company’s FatTwin platform is built with a modular design, which allows for a high level of customization and makes it easier to maintain. The result is proven efficiency that is able to compete in highly demanding IT environments. For instance, comparison tests revealed that the front I/O FatTwin configuration of Supermicro’s 2U Twin server resulted in $184 per node savings over a four-year period.

Supermicro has since continued the trend of delivering high performance, high efficiency solutions that cloud providers can use to meet a wide range of needs. The company recently showcased its latest server and storage technology at the Computex conference in Taiwan.

“With energy costs and environmental impact rising worldwide, data centers and cloud service providers deserve the best server and storage architectures that offer greatest savings in terms of performance per watt, per dollar, per square foot,” said Supermicro CEO Charles Liang.

Liang further expanded on the benefits of his company’s FatTwin platform, noting that the latest incarnation offers 16 percent better energy efficiency and $500 in savings per node.

“Alongside our higher density 3U MicroCloud supporting twenty-four Intel Xeon hot-swap nodes and expanding range of energy efficient server and storage solutions, Supermicro is leading the industry with server-building block solutions that save energy, cost and in the long run, our one and only Mother Earth,” Liang said.

The company exhibited several new offerings at Computex, including the 4U FatTwin platform, which is built for high efficiency (94 percent or greater) and is configurable for high performance computing as well as big data. The big data configuration consists of 12 fixed 3.5″ hard disk drives and an optional 2 additional 2.5″ HDDs.