Computing and Data Storage: Cloud Services vs. On-Site Hardware
Utilizing cloud services for computer processing and storage needs for appropriate applications.
Item ID: 490
Technical Advisory Group: 2013 Information Technology TAG (#8)
Average TAG Rating: 2.85 out of 5
TAG Ranking Date: 10/25/2013
TAG Rating Commentary:
- Interesting but challenges around M&V; setting the baseline is difficult
- I don't believe energy savings is going to be a significant driver for companies to move to a cloud computing model. There are many considerations involved in a company’s decision of whether to move to a private or public cloud computing model and energy consumption is near the lower portion of that list.
- The cloud is presumed to be more efficient and probably will be because there is profit motive to operate efficiently but there is no guarantee. Perhaps moving to a cloud with known efficiency would be an option.
- But program design can't handle this.
- Lots of hidden costs in this area
- Not an ET. We only claim savings if cloud servers are in our service territory plus we have an enabler.
- This will only save energy/carbon if the Cloud provider is chosen based on their PUE, CUE and utilization rates
- Data security, locality, quality of connection, accountability, and peace of mind all play into small to medium enterprises maintaining their own datacenters.
"Cloud computing is a general term for anything that involves delivering hosted services over the Internet." (SearchCloudComputing, 2010)
Amazon first popularized the service and the term "cloud computing" with the introduction of Amazon Web Services in 2006. Since then, services are evolving and, as of 2013, have been developed into reliable and secure services that can be used by many enterprises to provide reserve computing capacity, backup services, or even to replace the data center altogether, relying on the cloud services provider to provide the necessary software and hardware to satisfy their data needs off site.
As this technology improves, and as IT managers gain trust in the technology and associated service in terms of speed, security, reliability, costs, and so on, more companies are beginning to use cloud services. This can save significant energy because cloud service providers often operate large service centers with very efficient servers and associated power distribution and cooling services, whereas most small company-owned server rooms or server closets are relatively inefficient, under-utilized, and inefficiently cooled. On measure of efficiency is the Power Usage Effectiveness, or PUE. An average data center is 1.91, and they can be up to 2.5 or more. A large, efficient data center can have a PUE of 1.05 or less. In addition, the IT equipment itself may be more efficient, providing the same computing services for 30-60% of the energy of a self-contained server room on site.
Baseline Description: Standalone data center or server room space
Baseline Energy Use: 1500 kWh per year per square foot
Determining energy usage and a baseline is particularly problematic for cloud computing. A reasonable way to approach this is to figure out a “typical” energy usage of a certain area of cloud computing services, then estimate how much more efficient this might be than what it would be replacing, and “reverse calculate” the baseline.
The basic notion is that cloud computing services are provided from large data centers by companies whose business it is to provide data services efficiently. With energy being one of their major operating expenses, it is in their interest to provide these services as energy-efficiently as practical. In addition, they are operating on a large enough scale that it makes sense for them to invest in efficient infrastructure and design that may not be affordable for the operator of a small data center or server room. Therefore, it is reasonable to assume that the computing energy usage of a typical custom-built data center would be noticeably higher than that of the typical cloud service provider. Likely cloud service providers will have more efficient servers and will run them more fully loaded. Let’s assume conservatively they use about 40% less power than a custom-built data center for the same computing power. Their Power Usage Effectiveness (PUE -- the ratio of total data center energy use to IT energy usage), for similar reasons, will also likely be superior. An average data center has a PUE of 1.91. A large, efficiently-run data center should be able to achieve a PUE of 1.2 or less. For this analysis, we will use the conservative figure of PUE = 1.2. With these factors in mind, it is reasonable to assume that the cloud services data center would use energy at about the rate of the most efficient modular data center units (see ET #485). The following estimated energy usage of a module is based on a Hewlett Packard (HP) "POD 240a" fully loaded and operating at an average of 20% of nameplate capacity 24/7 or 8760 hrs. per year. They are 21.5 ft. by 45 ft (968 s.f.), with a “redundant IT capacity” and a power rating of 1152 kW (assume the other 1152 kW of redundant capacity is not operating). The container holds up to 44 50U racks for a total U-space of 2200. This is assuming the module has a Power Usage Effectiveness (PUE) of 1.2 (HP claims 1.05). Thus annual energy usage of the module would be:
1152 kW x 20% load x 1.2 PUE x 8760 hr. / 968 sf = 2500 kWh per year per square foot.
In addition, a typical data center would not be as dense as the cloud service data center, so it will actually occupy more square feet, but for an apples-to-apples comparison, and to still be able to use the “per square foot” unit, we will assume that the baseline data center fits into a hypothetical space the same size as the module. With these simplifying assumptions, the baseline energy use would be:
2500 kWh per year per square foot x 150% x 1.91/1.2 = 5970 kWh per year per square foot
Manufacturer's Energy Savings Claims:
Currently no data available.
Best Estimate of Energy Savings:
"Typical" Savings: 50%
Energy Savings Reliability: 2 - Concept validated
Energy Use of Emerging Technology:
750 kWh per square foot per year
Energy Use of an Emerging Technology is based upon the following algorithm.
Baseline Energy Use - (Baseline Energy Use * Best Estimate of Energy Savings (either Typical savings OR the high range of savings.))
Units: square foot
Currently no data available.
Currently no data available.
Simple payback, new construction (years): N/A
Simple payback, retrofit (years): N/A
Cost Effectiveness is calculated using baseline energy use, best estimate of typical energy savings, and first cost. It does not account for factors such as impacts on O&M costs (which could be significant if product life is greatly extended) or savings of non-electric fuels such as natural gas. Actual overall cost effectiveness could be significantly different based on these other factors.
Reference and Citations:
Cloud computing saves energy, and here’s the data to show it
Report on the development of a new tool to model data center energy use called Cloud Energy and Emissions Research Model, or CLEER, developed by Lawrence Berkeley Nation Laboratory
Cloud Energy and Emissions Research Model
Lawrence Berkeley National Lab and Northwestern University
Web-based data center modeling tool developed by LBNL and Northwestern University to compare the energy use of a local data center to that of providing those same computing services via the cloud.
Working with Enterprise ALM: Mobile, cloud and more
Definition of Cloud and explanation of IaaS, PaaS, and SaaS
Developers, Developers, Developers: Engaging the Missing Link in IT Resource Efficiency
The Green Grid
The Green Grid Power Point presentation on the value of high utilization rates of IT equipment.
Cloud Computing -- The IT Solution for the 21st Century
Carbon Disclosure Project
Report of study conducted by Verdantix and sponsored by AT&T (cloud services provider) estimating that "cloud computing can help [large US] companies realize $12.3 billion in energy savings and 85.7 million metric tons of CO2 savings annually by 2020."