ºÝºÝߣ

ºÝºÝߣShare a Scribd company logo

H.brotherton student sustainabilitysummit

•Download as PPT, PDF•
•
Heather Brotherton
Heather Brotherton

The document discusses an experiment that measured the energy efficiency of different server operating systems. Quantitative data was collected on the watts consumed and threads run by servers using Ubuntu, Windows Server 2008 Core, and Windows Server 2008 with a graphical user interface (GUI). The findings showed the GUI version of Windows Server 2008 consumed 1.43 more watts on average than the hardware baseline, while the Core version only increased usage by 0.15 watts. Having a GUI required processing and rendering graphics, resulting in over 100 additional threads and approximately 1 more watt of power compared to non-GUI systems. Removing unnecessary components like graphics cards from servers could significantly reduce overall energy usage in data centers.

1 of 1
Download to read offline
Department of Computer Technology Energy Overhead of the Grapical User Interface in Server Operating Systems Introduction Experiment Operating systems reside on nearly every server, controlling systems resources (Vahdat, Lebeck, & Schlatter Ellis, 2000). Operating systems are continually designed to provide more features and these features increasingly include energy savings management. Focusing on server level operating system software allows us to take advantage of the Cascade Effect as described by Emerson. The Cascade Effect states that for every watt saved at the server level 2.84 watts are saved by the data center (Emerson). In addition if an operating system where chosen that required very little graphics and no sound additional savings could be achieved. This is because it would allow the removal or reduction of components. The chart labeled Watts lost per server provides a rough breakdown of the number of watts lost broken down by server component (Anderson, 2007). For example PCI cards, these would include video and audio cards. Exclusion of unnecessary PCI cards from servers could save 41 watts. Unnecessary components waste energy even when they are not used(Google). When these savings are multiplied into the average number of servers in a data center the savings become very significant. For example, if a PCI card such as video card were removed for a savings of 41watts from 500 servers in a data center, the cumulative watts saved would be 58220 watts per year. At an average of ten cents per kilowatt-hour this results in a savings of $51,035.65 per year. Watts used by servers are converted to heat, which is expressed in British Thermal Units(Anayochukwu Ani, Ndubueze Nzeako, & Chigbo Obianuko, 2012). Each watt consumed by a server translates into approximately 3.4129 BTUs per hour (Barielle, 2011). Waste heat must then be removed from the data center to avoid damaging the servers resulting in additional energy consumed by cooling units. This is the reasoning behind the Cascade Effect. This is also the reason that the primary focus of this study is on reducing the watts consumed at the server level. Quantitative data was collected to measure the efficiency of server operating systems. Experimentation and observation were employed at the server and software levels. The data collected includes observations of watts consumed by the server with different hardware and operating system software configurations. All energy readings were collected for a minimum of one hour using the Watts Up? Meter. The software used for the testing were the following x86 operating systems: Ubuntu 9.10 (Linux) Ubuntu 11.10 (Linux) Windows Server 2008 R2 Datacenter Core Windows Server 2008 R2 Datacenter GUI No operating system configuration changes were performed, all were installed using defaults. The systems were not connected to the Internet and no updates were performed on the operating systems. Linux based server operating systems ran the Top command during the observations. Top provides data for on-going processes. A sample was taken not running Top to serve as a baseline so that the load of running top can be determined. Top was configured via command line to take readings at intervals of one second. The results were sent to a file for possible analysis. The command used was: Type top -d 1 > /home/testOSName.txt The Windows based operating systems do not have a direct equivalent to the Linux Top program. The Windows command line tool called Typeperf was configured to provide much of the same information. The command used was: typeperf "MemoryAvailable bytes" "processor(*)% processor time" "Process(*)Thread Count" > testOSName.csv This tool was chosen in part because it would run with or without the standard GUI based Windows operating system. Another advantage to this tool is reduction in the possibility of creating the energy overhead that might come with a more sophisticated program. As this study’s focus is on differences in operating systems, adding another program would unnecessarily complicate the readings. Typeperf output to file proved to be valuable during the data analysis phase as the data gathered from the Typeperf tool provided insight into the differences in what was happening at the operating system level. Windows 2008 R2 Datacenter Core (Non-GUI) Time Threads Watts (Minutes) (Mean (Mean) Number) 3:16 263 17.1 Windows 2008 R2 Datacenter (GUI) Time Threads Watts (Minutes) (Mean (Mean) Number) 9:42 373 18.65 9:43 361 18.6 3:17 263 17 9:44 362 18.65 9:45 360 18.65 3:18 3:19 260 256 17 17.05 9:46 360 18.7 3:20 259 17.05 9:47 368 18.7 3:21 255 17.15 9:48 9:49 364 363 18.6 18.7 3:22 255 17.25 3:23 252 17.1 9:50 362 18.65 3:24 250 17.2 9:51 364 18.7 3:25 254 17.2 9:52 376 18.65 3:26 250 17.25 Findings The greatest overall variation is the Windows 2008 R2 Datacenter graphical user interface operating system that varied from 15.30 watts to 24.90 watts with a difference of 9.60 watts. The data collected provides some insight into the operating system-power consumption relationship. The data collected indicates a correlation between increased energy consumption and the presence of a graphical user interface. The Comparison of watts consumed by operating system scatter chart shows that the operating systems that do not run a graphical user interface (GUI) use roughly 17.5 to 17.6 watts. The two graphical user interface (GUI) based operating systems tested consumed 18.1 to 18.9 watts roughly. It could be extrapolated that not using a GUI would save .6 to 1.3 watts per server. The data collected confirmed the expected difference between GUI based operating systems and non-GUI. One would assume that there is more energy overhead in a system that is processing and rendering graphics than one that does not. The tables above compare the Windows 2008 R2 datacenter thread and watt consumption for a period of ten minutes The mean number of threads the GUI version of the OS is running is 365; the non-GUI mean was 256. This difference of approximately 109 threads may account for the overall difference in energy consumption. This also indicates that a reduction of 100 threads can save roughly one watt at the server level. The server hardware as configured was found to have a mean energy consumption of 17.42 watts. The Core (non-GUI) installation of Windows 2008 datacenter increases the power load by .15 watts from the baseline hardware load. The standard Windows 2008 R2 datacenter installation increases the power load from the baseline by 1.43 watts. This may not seem like much but this represents an increase of nearly ten times the load added by the non-GUI version. The watt readings from the micro server tests combined with the typeperf data provide insight into why the GUI operating system consumes more energy. The GUI version of the Windows Server 2008 R2 datacenter runs over 100 threads more than the non-GUI Core version of Windows Server 2008 R2 datacenter. This 100-thread overhead represents approximately one watt of additional energy consumption at the server level. Allowing for the cascade effect this would be approximately three watts per server. References Anayochukwu Ani, V., Ndubueze Nzeako, A., & Chigbo Obianuko, J. (2012). Energy Optimization at Datacenters in Two Different Locations of Nigeria. International Journal of Energy Engineering, 151-164. Anderson, N. (2007, 27 2). New industry group looks to cut server room power consumption. Retrieved 11 21, 12, from Ars Technica: http://arstechnica.com/uncategorized/2007/02/8932/ Barielle, S. (2011, 11). Calculating TCO for Energy. Retrieved from IBM Systems Magazine: http://www.ibmsystemsmag.com/mainframe/Business-Strategy/ROI/energy_estimating/ Emerson. (n.d.). Energy Logic: Calculating and Prioritizing your data center IT Efficency Actions. Retrieved 3 13, 2012, from Efficient Data Centers.com: http://www.efficientdatacenters.com/edc/docs/EnergyLogicMetricPaper.pdf Google. (n.d.). Efficiency: How we do it. Retrieved 11 21, 12, from Google Data centers: http://www.google.com/about/datacenters/efficiency/internal/#servers ©Heather Brotherton 2011-2013

More Related Content

H.brotherton student sustainabilitysummit

  • 1. Department of Computer Technology Energy Overhead of the Grapical User Interface in Server Operating Systems Introduction Experiment Operating systems reside on nearly every server, controlling systems resources (Vahdat, Lebeck, & Schlatter Ellis, 2000). Operating systems are continually designed to provide more features and these features increasingly include energy savings management. Focusing on server level operating system software allows us to take advantage of the Cascade Effect as described by Emerson. The Cascade Effect states that for every watt saved at the server level 2.84 watts are saved by the data center (Emerson). In addition if an operating system where chosen that required very little graphics and no sound additional savings could be achieved. This is because it would allow the removal or reduction of components. The chart labeled Watts lost per server provides a rough breakdown of the number of watts lost broken down by server component (Anderson, 2007). For example PCI cards, these would include video and audio cards. Exclusion of unnecessary PCI cards from servers could save 41 watts. Unnecessary components waste energy even when they are not used(Google). When these savings are multiplied into the average number of servers in a data center the savings become very significant. For example, if a PCI card such as video card were removed for a savings of 41watts from 500 servers in a data center, the cumulative watts saved would be 58220 watts per year. At an average of ten cents per kilowatt-hour this results in a savings of $51,035.65 per year. Watts used by servers are converted to heat, which is expressed in British Thermal Units(Anayochukwu Ani, Ndubueze Nzeako, & Chigbo Obianuko, 2012). Each watt consumed by a server translates into approximately 3.4129 BTUs per hour (Barielle, 2011). Waste heat must then be removed from the data center to avoid damaging the servers resulting in additional energy consumed by cooling units. This is the reasoning behind the Cascade Effect. This is also the reason that the primary focus of this study is on reducing the watts consumed at the server level. Quantitative data was collected to measure the efficiency of server operating systems. Experimentation and observation were employed at the server and software levels. The data collected includes observations of watts consumed by the server with different hardware and operating system software configurations. All energy readings were collected for a minimum of one hour using the Watts Up? Meter. The software used for the testing were the following x86 operating systems: Ubuntu 9.10 (Linux) Ubuntu 11.10 (Linux) Windows Server 2008 R2 Datacenter Core Windows Server 2008 R2 Datacenter GUI No operating system configuration changes were performed, all were installed using defaults. The systems were not connected to the Internet and no updates were performed on the operating systems. Linux based server operating systems ran the Top command during the observations. Top provides data for on-going processes. A sample was taken not running Top to serve as a baseline so that the load of running top can be determined. Top was configured via command line to take readings at intervals of one second. The results were sent to a file for possible analysis. The command used was: Type top -d 1 > /home/testOSName.txt The Windows based operating systems do not have a direct equivalent to the Linux Top program. The Windows command line tool called Typeperf was configured to provide much of the same information. The command used was: typeperf "MemoryAvailable bytes" "processor(*)% processor time" "Process(*)Thread Count" > testOSName.csv This tool was chosen in part because it would run with or without the standard GUI based Windows operating system. Another advantage to this tool is reduction in the possibility of creating the energy overhead that might come with a more sophisticated program. As this study’s focus is on differences in operating systems, adding another program would unnecessarily complicate the readings. Typeperf output to file proved to be valuable during the data analysis phase as the data gathered from the Typeperf tool provided insight into the differences in what was happening at the operating system level. Windows 2008 R2 Datacenter Core (Non-GUI) Time Threads Watts (Minutes) (Mean (Mean) Number) 3:16 263 17.1 Windows 2008 R2 Datacenter (GUI) Time Threads Watts (Minutes) (Mean (Mean) Number) 9:42 373 18.65 9:43 361 18.6 3:17 263 17 9:44 362 18.65 9:45 360 18.65 3:18 3:19 260 256 17 17.05 9:46 360 18.7 3:20 259 17.05 9:47 368 18.7 3:21 255 17.15 9:48 9:49 364 363 18.6 18.7 3:22 255 17.25 3:23 252 17.1 9:50 362 18.65 3:24 250 17.2 9:51 364 18.7 3:25 254 17.2 9:52 376 18.65 3:26 250 17.25 Findings The greatest overall variation is the Windows 2008 R2 Datacenter graphical user interface operating system that varied from 15.30 watts to 24.90 watts with a difference of 9.60 watts. The data collected provides some insight into the operating system-power consumption relationship. The data collected indicates a correlation between increased energy consumption and the presence of a graphical user interface. The Comparison of watts consumed by operating system scatter chart shows that the operating systems that do not run a graphical user interface (GUI) use roughly 17.5 to 17.6 watts. The two graphical user interface (GUI) based operating systems tested consumed 18.1 to 18.9 watts roughly. It could be extrapolated that not using a GUI would save .6 to 1.3 watts per server. The data collected confirmed the expected difference between GUI based operating systems and non-GUI. One would assume that there is more energy overhead in a system that is processing and rendering graphics than one that does not. The tables above compare the Windows 2008 R2 datacenter thread and watt consumption for a period of ten minutes The mean number of threads the GUI version of the OS is running is 365; the non-GUI mean was 256. This difference of approximately 109 threads may account for the overall difference in energy consumption. This also indicates that a reduction of 100 threads can save roughly one watt at the server level. The server hardware as configured was found to have a mean energy consumption of 17.42 watts. The Core (non-GUI) installation of Windows 2008 datacenter increases the power load by .15 watts from the baseline hardware load. The standard Windows 2008 R2 datacenter installation increases the power load from the baseline by 1.43 watts. This may not seem like much but this represents an increase of nearly ten times the load added by the non-GUI version. The watt readings from the micro server tests combined with the typeperf data provide insight into why the GUI operating system consumes more energy. The GUI version of the Windows Server 2008 R2 datacenter runs over 100 threads more than the non-GUI Core version of Windows Server 2008 R2 datacenter. This 100-thread overhead represents approximately one watt of additional energy consumption at the server level. Allowing for the cascade effect this would be approximately three watts per server. References Anayochukwu Ani, V., Ndubueze Nzeako, A., & Chigbo Obianuko, J. (2012). Energy Optimization at Datacenters in Two Different Locations of Nigeria. International Journal of Energy Engineering, 151-164. Anderson, N. (2007, 27 2). New industry group looks to cut server room power consumption. Retrieved 11 21, 12, from Ars Technica: http://arstechnica.com/uncategorized/2007/02/8932/ Barielle, S. (2011, 11). Calculating TCO for Energy. Retrieved from IBM Systems Magazine: http://www.ibmsystemsmag.com/mainframe/Business-Strategy/ROI/energy_estimating/ Emerson. (n.d.). Energy Logic: Calculating and Prioritizing your data center IT Efficency Actions. Retrieved 3 13, 2012, from Efficient Data Centers.com: http://www.efficientdatacenters.com/edc/docs/EnergyLogicMetricPaper.pdf Google. (n.d.). Efficiency: How we do it. Retrieved 11 21, 12, from Google Data centers: http://www.google.com/about/datacenters/efficiency/internal/#servers ©Heather Brotherton 2011-2013