IBM Launches "Mainframe Gas Gauge" in Environmental Push

An Industry-First, New Environmental Figures Demonstrate the Mainframe’s Energy Usage Versus x86 Servers and Other Distributed Computing Platforms



ARMONK, NY .- In an extension of the company’s Project Big Green, IBM (NYSE: IBM) today launched a program that allows mainframe customers to monitor their systems’ precise energy consumption in real-time.

IBM will also begin publishing typical energy consumption data for the IBM System z9 mainframe. The data is derived from actual field measurements of approximately 1,000 customer machines, determining average watts/hour consumed which can be used to calculate watts per unit -- similar to automobile miles per gallon estimates and appliance kilowatt per year ratings.

The data collected for August and September determines that typical energy use can be normally 60% of the “label” or maximum rating for the model of mainframe measures.

With this news, IBM becomes the first company to embrace recommendations from a recent EPA report that encourages server vendors to publish typical energy consumption figures for servers.

“The mainframe’s high utilization rates and extreme virtualization capability may help make it a more energy-efficient choice for large enterprises,” said David Anderson PE, IBM green consultant. “A single mainframe running Linux may be able to perform the same amount of work as approximately 250 x86 processors while using as little as two to ten percent of the amount of energy. Customers can now measure the energy advantages of IBM System z.”

Here’s how the metering system works: the new IBM solution monitors a mainframe’s actual energy and cooling statistics (collected by internal sensors); and presents them in real time on the System Activity Display. With this system, a user can now correlate the energy consumed with work actually performed. When the machine reports its maintenance health on a weekly basis, the power statistics may be used. These statistics can be observed real time or also summarized for project or trend analysis. Energy consumption statistics are used for demonstrating cost savings toward electric rebates and programs to reduce data center energy consumption.

A Power Estimator Tool is also available for future planning. It calculates how changes in system configurations and workloads can affect the entire energy “envelope” -- including the power needed to both run and cool the machines. For example, a customer adding a single mainframe processor for Linux applications could project the amount of additional energy required before and when the feature is turned on. Normally less than approximately 20 watts are added when an Integrated Facility for Linux (IFL) feature is turned on. Typically, a single mainframe processor with zVM virtualization can perform the work of multiple x86 processors, because of the mainframe’s design point for running many mixed workloads at high utilization rates. A single processing chip executing hundreds of workloads efficiently is the key to consuming much less energy than many x86 servers which have many more power consuming components This translates into a simplified infrastructure and cost savings.

Typical Energy Consumption Figures

The metering system is being launched in tandem with a new IBM program to publish consolidated real-world consumption figures by model for System z9. These typical use figures will be vital to data center planning as they will give customers an idea of how much energy a particular mainframe consumes.

IBM has summarized the field population data for each month since August 2, 2007, when the EPA published the report to Congress on Data Center and Server Energy Efficiency. The EPA encouraged server vendors to publish typical energy usage numbers to enable purchasers of servers to make informed decisions based on energy efficiency.