How to Reduce Cloud Computing Costs
You went out and brokered cloud services until you proved you had a market. Then you set about building your own cloud infrastructure to deliver those services instead of subcontracting them out to someone else. In doing so, you carved a big chunk of cost out of your sales and increased your profits.
It becomes a matter of degrees.
Once you’ve got your own data center up and running, there are really only two ways to make it more profitable:
- Sell more services that use it.
- Reduce the cost of operating it.
We’ll save No. 1 for another post. But when you’re looking to reduce the cost of operating your own cloud data center infrastructure, the “secret sauce” lies in doing many small things well. Making small gains everywhere adds up into a significant reduction in operating expenses.
Power down on consumption.
According to an IDC white paper sponsored by Dell™ & Intel®, “Depending on the type of data center, energy costs are between 30% and 40% of an organization's operational costs in the region, and they are slated to grow further if customers continue to add more server, storage and networking equipment without consciously embarking on an energy-saving strategy.”
Reducing power consumption is a sure way to lower your operating costs. In an April 2015 post on the Schneider Electric Data Center blog, Steven Carlini, senior director of data center global solutions, tells us that, “The role that power use plays around data center performance and cost is enormous — efficiency is naturally found by reducing power consumption and improving power usage effectiveness (PUE).”
To provide a sense of how substantial the waste is, Carlini adds, “Worldwide, data center power and cooling infrastructure wastes more than 60,000,000 megawatt-hours per year of electricity — energy not used to power IT equipment. This is a colossal financial burden for our industry, and now a key public policy issue.”
Suggesting that a systematic effort to conserve and improve energy management will reduce electrical consumption by up to 90%, he points out that, “The latest UPS systems perform exponentially well at lighter loads. Actual loss can be reduced by 65%, since power must not be used to cool wasted heat.”
Even at the component level, smart design can yield ongoing cost savings. According to a Micron® Technical Marketing Brief, “Data center floor space and power are premium commodities. Data centers may fill up or become power-constrained, and the option of building out more space or power capacity may not be practical or affordable. Devices and components designed with greater density and power efficiency help address these constraints.”
Micron designs and manufactures a family of enterprise-class SSDs for data center infrastructure needs, from SATA and SAS hard disk drive (HDD) replacements to ultra, low-latency PCIe Flash storage solutions that deliver performance, capacity, durability and power advantages for the data center.
Dell also addresses this issue in “The Efficient Data Center Lowers Costs, Risks Profile and Environmental Impacts,” in which they predict, “Looking ahead, digital content will only increase — doubling every 18 months, according to market research firm IDC — and power consumption will continue to be a hot-button issue as nonrenewable energy resources dwindle and electricity costs rise.”
Dell’s approach is called the Efficient Data Center, and according to Dell, “it can help you free up some 50% of your IT budget while also lowering your carbon footprint. Built on virtualization, automation and consolidation, this strategy yields open, robust and cost-effective solutions that help you optimize your current center, virtualize in a time frame that makes sense for your business and leverage cloud technologies where appropriate.”
Their free Data Center Capacity Planner provides power, cooling and airflow estimates for server centers. You can also calculate potential energy savings using Dell's Client Energy Savings Calculator. Consultants from the Dell Data Center Practice can help you determine how to get the most out of your power efficiency and data center architectural choices.
Manage storage resources.
Leveraging scale-out architectures, Dell also helps you grow storage capacity on demand and avoid the need to overprovision. Since capacity and demand can be closely matched, this is a highly efficient way to manage resources, including powering and cooling. A broad array of services and products — including virtualized EqualLogic™ storage area networks (SANs) with built-in automation, cost-effective and capable PowerVault™ products and flexible Dell/EMC arrays with advanced data management features — can help you make smart choices about where, how and at what cost to store your data.
Dell offers a new set of services called Intelligent Data Management (IDM), designed to encompass growth planning and analysis, data protection, archiving, discovery and compliance, and optimization of server and storage resources in the data center.
Selecting the right storage solutions can also help reduce electrical power costs. As Fujifilm VP of Marketing Rich Gadomski points out, “One reason for the heavy energy consumption by data centers is that they rely heavily on spinning hard disk drive technology to store their data. Often the response to increasing data growth has been to add more disk arrays to solve the problem. A hard disk drive platter spinning 24/7/365 at 7,000 or 10,000 RPMs requires power to not only spin it, but to cool it as well. Otherwise, the heat generated by the constant spinning would corrupt and eventually destroy the data.”
Micron recommends the use of Solid State Drives (SSDs) to replace these heat-generating spinning disks. “By using solid state media instead of physically spinning disks, SSDs deliver random I/O performance that is orders of magnitude faster than HDDs — at a lower cost per IOPS. Superior performance, better performance-to-power characteristics, predictable lifespan, standard form factors and design/technology maturity have secured the SSD’s place in the modern, high-performance data center.”
Consolidate physical servers with virtualized ones.
IDC advises, “One of the biggest problems with most data centers is over-provisioning. They are planned for peak periods that come once in a while, and for the rest of the period the excess capacity is grossly underutilized. In the past few years, organizations have addressed this redundancy with the deployment of virtualization. However, one problem is at risk of being replaced with another as customers are again creating too many servers — only of the virtual type instead. The problem is addressed with better management and automation tools that can help improve utilization, create more capacity for scale and help improve the provisioning of IT.”
By consolidating many physical servers into fewer virtualized servers hosting multiple virtual machines (VMs), data centers can reduce electrical power, cooling and maintenance costs significantly. When the system finds the opportunity to move all VMs off a given host — such as by automating the management of virtual clusters with tools like Microsoft® System Center — it can also power down that host until it is needed again, creating further power savings.
Capitalize on cloud services.
With the ability to connect Active Directory for Windows Server to Active Directory for Azure, many users are leveraging cloud infrastructure services to accommodate short-term scaling requirements rather than invest in new server and storage hardware that will sit idle most of the time. Active Directory enables system managers to view and control the entire hybrid infrastructure as one contiguous environment.
Partner with Insight.
While any one of these strategies alone may not produce sufficient savings, taken together they can be used to realize significant cost reductions in the ongoing operation of your cloud data center. For more information, or to discuss your specific needs, contact Insight Service Provider sales today.