{"id":5563,"date":"2014-07-31T11:34:28","date_gmt":"2014-07-31T09:34:28","guid":{"rendered":"http:\/\/blog.zhaw.ch\/icclab\/?p=5563"},"modified":"2014-08-14T10:41:51","modified_gmt":"2014-08-14T08:41:51","slug":"understanding-how-server-energy-consumption-varies-with-cpu-bound-workload","status":"publish","type":"post","link":"https:\/\/blog.zhaw.ch\/icclab\/understanding-how-server-energy-consumption-varies-with-cpu-bound-workload\/","title":{"rendered":"Understanding how server energy consumption varies with CPU-bound workload"},"content":{"rendered":"<p>In some of our project work, we need to understand the energy consumption of our servers. \u00a0One basic test we performed was to determine how server power consumption increases with load. In this blog post we document the results we obtained.<\/p>\n<p><!--more--><\/p>\n<p>We focused on increasing the compute load on the machine by increasing the amount of simultaneous \u03c0 calculations; the idea being to load up all the individual cores in the machine and monitor how energy consumption increases.<\/p>\n<p>The machines we were using were <a href=\"http:\/\/www-03.ibm.com\/systems\/x\/hardware\/rack\/x3550m4\/\">IBM System x3550 M4 Servers<\/a>. These had:<\/p>\n<ul>\n<li>2 * Intel(R) Xeon(R) CPU E5-2640 @ 2.50GHz<\/li>\n<li>192 GB RAM<\/li>\n<li>1.5 TB HDD<\/li>\n<\/ul>\n<p>These machines provide access to energy consumption information through the <a href=\"http:\/\/www-01.ibm.com\/support\/knowledgecenter\/linuxonibm\/liaai.energymgmt\/liaaienergyloadmod.htm?lang=en\">IBM AEM<\/a> interface; this exposes how many Joules have been consumed by the machine since some start time and this can easily be converted to Watts over some timeframe. We find that these machines typically consume about 140W when idling.<\/p>\n<p>Each experiment was designed such that a number of <a href=\"https:\/\/github.com\/icclab\/Arcus-Energy-Monitor\/blob\/juno-branch\/Extras\/picalc.py\">calculations of \u03c0<\/a> are performed simultaneously. As the number of \u03c0 calculations is increased, the number of active cores should increase and we can understand how loading up the cores increases energy consumption. In each case, the amount of energy consumed during the \u03c0 calculations is obtained (by determining the difference in total Joules consumed between the start and the end of the experiment). The total runtime is recorded also, although generally this did not change so much as we only kept the number of calculations down to the number of available cores. \u00a0We ran three variants of the tests which calculate \u03c0\u00a0to different levels of precision. The first test calculates one hundred thousand numbers after the decimal point of the value of \u03c0, the second 200,000 and third 800,000. The results obtained are shown in the graphs below.<\/p>\n<p><a href=\"http:\/\/blog.zhaw.ch\/icclab\/files\/2014\/07\/pical1.jpg\"><img loading=\"lazy\" decoding=\"async\" class=\"size-medium wp-image-5564 aligncenter\" src=\"http:\/\/blog.zhaw.ch\/icclab\/files\/2014\/07\/pical1-300x206.jpg\" alt=\"pical1\" width=\"300\" height=\"206\" srcset=\"https:\/\/blog.zhaw.ch\/icclab\/files\/2014\/07\/pical1-300x206.jpg 300w, https:\/\/blog.zhaw.ch\/icclab\/files\/2014\/07\/pical1-435x300.jpg 435w, https:\/\/blog.zhaw.ch\/icclab\/files\/2014\/07\/pical1.jpg 481w\" sizes=\"auto, (max-width: 300px) 100vw, 300px\" \/><\/a><\/p>\n<p style=\"text-align: center\"><em>Figure 1: Variation in Server Energy Consumption with\u00a0number of simultaneous\u00a0\u03c0 calculations &#8211; low precision<\/em><\/p>\n<p style=\"text-align: center\">\u00a0<a href=\"http:\/\/blog.zhaw.ch\/icclab\/files\/2014\/07\/pical3.jpg\"><img loading=\"lazy\" decoding=\"async\" class=\"alignnone size-medium wp-image-5566\" src=\"http:\/\/blog.zhaw.ch\/icclab\/files\/2014\/07\/pical3-300x202.jpg\" alt=\"pical3\" width=\"300\" height=\"202\" srcset=\"https:\/\/blog.zhaw.ch\/icclab\/files\/2014\/07\/pical3-300x202.jpg 300w, https:\/\/blog.zhaw.ch\/icclab\/files\/2014\/07\/pical3-445x300.jpg 445w, https:\/\/blog.zhaw.ch\/icclab\/files\/2014\/07\/pical3.jpg 499w\" sizes=\"auto, (max-width: 300px) 100vw, 300px\" \/><\/a><\/p>\n<p style=\"text-align: center\"><em>Figure 2: Variation in Server Energy Consumption with\u00a0number of simultaneous\u00a0\u03c0 calculations &#8211; moderate\u00a0precision<\/em><\/p>\n<p style=\"text-align: center\"><a href=\"http:\/\/blog.zhaw.ch\/icclab\/files\/2014\/07\/pical2.jpg\"><img loading=\"lazy\" decoding=\"async\" class=\"alignnone size-medium wp-image-5565\" src=\"http:\/\/blog.zhaw.ch\/icclab\/files\/2014\/07\/pical2-300x203.jpg\" alt=\"pical2\" width=\"300\" height=\"203\" srcset=\"https:\/\/blog.zhaw.ch\/icclab\/files\/2014\/07\/pical2-300x203.jpg 300w, https:\/\/blog.zhaw.ch\/icclab\/files\/2014\/07\/pical2-442x300.jpg 442w, https:\/\/blog.zhaw.ch\/icclab\/files\/2014\/07\/pical2.jpg 501w\" sizes=\"auto, (max-width: 300px) 100vw, 300px\" \/><\/a><\/p>\n<p style=\"text-align: center\"><em>Figure 3: Variation in Server Energy Consumption with\u00a0number of simultaneous\u00a0\u03c0 calculations &#8211; high\u00a0precision<\/em><\/p>\n<p>The three different experiments essentially show the same results: the power consumption varies from about 170W for one active core up to about 260 W for all cores active. Also, it can be seen that the energy consumption increases quite linearly as the number of calculations varies from about 2-10 and then flattens out a little. It is interesting to note that the baseline energy consumption &#8211; energy consumption in the idle state &#8211; accounts for about 50% of the total energy consumption when the machine is fully loaded.<\/p>\n<p>The above tests were performed for a standard CentOS installation in which the processors are configured to deliver good performance. Modern OSs and processors support different configurations in which performance and energy consumption can be traded off &#8211; this is done via <a href=\"https:\/\/wiki.archlinux.org\/index.php\/CPU_frequency_scaling\">CPU frequency scaling<\/a> which gives the OS control over how the CPU changes between states that have different power\/performance characteristics. This is done using CPU <a href=\"https:\/\/www.kernel.org\/doc\/Documentation\/cpu-freq\/governors.txt\">governors<\/a> which are essentially policies that govern these transitions. Examples of these policies include ondemand and powersave (ondemand essentially increases the CPU frequency if the processor load exceeds 95%, while powersave attempts to run the processor using the minimum available frequency). The concrete linux packages required to leverage this frequency scaling are <i>cpupowerutils <\/i>and the <i>acpi-cpufreq<\/i> kernel <a href=\"https:\/\/wiki.archlinux.org\/index.php\/CPU_frequency_scaling#CPU_frequency_driver\">module<\/a>.<\/p>\n<p>To understand the impact of the CPU governor, we performed the same tests again, this time using the powersave governor rather than the ondemand governor (which is enabled by default). The results obtained are shown below.<\/p>\n<p style=\"text-align: center\"><a href=\"http:\/\/blog.zhaw.ch\/icclab\/files\/2014\/07\/picalcsv1.jpg\"><img loading=\"lazy\" decoding=\"async\" class=\"alignnone size-medium wp-image-5567\" src=\"http:\/\/blog.zhaw.ch\/icclab\/files\/2014\/07\/picalcsv1-300x204.jpg\" alt=\"picalcsv1\" width=\"300\" height=\"204\" srcset=\"https:\/\/blog.zhaw.ch\/icclab\/files\/2014\/07\/picalcsv1-300x204.jpg 300w, https:\/\/blog.zhaw.ch\/icclab\/files\/2014\/07\/picalcsv1-440x300.jpg 440w, https:\/\/blog.zhaw.ch\/icclab\/files\/2014\/07\/picalcsv1.jpg 501w\" sizes=\"auto, (max-width: 300px) 100vw, 300px\" \/><\/a><\/p>\n<p style=\"text-align: center\"><em>Figure 4: Variation in Server Energy Consumption with\u00a0number of simultaneous\u00a0\u03c0 calculations &#8211; low\u00a0precision, powersave governor<\/em><\/p>\n<p style=\"text-align: center\"><a href=\"http:\/\/blog.zhaw.ch\/icclab\/files\/2014\/07\/picalcsv2.jpg\"><img loading=\"lazy\" decoding=\"async\" class=\"alignnone size-medium wp-image-5568\" src=\"http:\/\/blog.zhaw.ch\/icclab\/files\/2014\/07\/picalcsv2-300x207.jpg\" alt=\"picalcsv2\" width=\"300\" height=\"207\" srcset=\"https:\/\/blog.zhaw.ch\/icclab\/files\/2014\/07\/picalcsv2-300x207.jpg 300w, https:\/\/blog.zhaw.ch\/icclab\/files\/2014\/07\/picalcsv2-434x300.jpg 434w, https:\/\/blog.zhaw.ch\/icclab\/files\/2014\/07\/picalcsv2.jpg 492w\" sizes=\"auto, (max-width: 300px) 100vw, 300px\" \/><\/a><\/p>\n<p style=\"text-align: center\"><em>Figure 5: Variation in Server Energy Consumption with\u00a0number of simultaneous\u00a0\u03c0 calculations &#8211;\u00a0moderate\u00a0precision,\u00a0<em>powersave governor<\/em><\/em><\/p>\n<p style=\"text-align: center\"><a href=\"http:\/\/blog.zhaw.ch\/icclab\/files\/2014\/07\/picalcsv3.jpg\"><img loading=\"lazy\" decoding=\"async\" class=\"alignnone size-medium wp-image-5569\" src=\"http:\/\/blog.zhaw.ch\/icclab\/files\/2014\/07\/picalcsv3-300x193.jpg\" alt=\"picalcsv3\" width=\"300\" height=\"193\" srcset=\"https:\/\/blog.zhaw.ch\/icclab\/files\/2014\/07\/picalcsv3-300x193.jpg 300w, https:\/\/blog.zhaw.ch\/icclab\/files\/2014\/07\/picalcsv3-464x300.jpg 464w, https:\/\/blog.zhaw.ch\/icclab\/files\/2014\/07\/picalcsv3.jpg 499w\" sizes=\"auto, (max-width: 300px) 100vw, 300px\" \/><\/a><\/p>\n<p style=\"text-align: center\"><em>Figure 6: Variation in Server Energy Consumption with\u00a0number of simultaneous\u00a0\u03c0 calculations &#8211; high\u00a0precision,\u00a0<em>powersave governor<\/em><\/em><\/p>\n<p>As with the first set of tests, the three sets of experiments showed essentially the same result. Clearly, the powersave governor has a very significant impact on the amount of energy consumed, with only very modest increases in energy consumption over the idle state (30-40W). This is in contrast to the ondemand case in which the difference could reach about 130W. Interestingly, in the powersave case, the idle power consumption accounts for a very large percentage of the total energy consumption. While the powersave governor results in a significant reduction in energy consumption; it also results in an increase in the amount of time taken to perform the calculations &#8211; generally, the time taken to perform the calculations almost doubled.<\/p>\n<p>The main finding from the above experiments is that the baseline energy consumption accounts for a very significant percentage of the total energy consumption. This does give rise to questions around how significant the energy savings can be by performing software optimizations on today\u2019s hardware. We will explore this in future work.<\/p>\n<p><em>(Thanks to <a href=\"http:\/\/about.me\/brunograz\">Bruno<\/a> for generating the results documented here).<\/em><\/p>\n<div class=\"pt-sm\">Schlagw\u00f6rter: <a href=\"https:\/\/blog.zhaw.ch\/icclab\/tag\/cpu-bound-workload\/\">cpu bound workload<\/a>, <a href=\"https:\/\/blog.zhaw.ch\/icclab\/tag\/server-energy-consumption\/\">server energy consumption<\/a><br><\/div>","protected":false},"excerpt":{"rendered":"<p>In some of our project work, we need to understand the energy consumption of our servers. \u00a0One basic test we performed was to determine how server power consumption increases with load. In this blog post we document the results we obtained. Schlagw\u00f6rter: cpu bound workload, server energy consumption<\/p>\n","protected":false},"author":92,"featured_media":0,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"ngg_post_thumbnail":0,"footnotes":""},"categories":[5],"tags":[435,434],"features":[],"class_list":["post-5563","post","type-post","status-publish","format-standard","hentry","category-articles","tag-cpu-bound-workload","tag-server-energy-consumption"],"yoast_head":"<!-- This site is optimized with the Yoast SEO Premium plugin v27.2 (Yoast SEO v27.2) - https:\/\/yoast.com\/product\/yoast-seo-premium-wordpress\/ -->\n<title>Understanding how server energy consumption varies with CPU-bound workload - Service Engineering (ICCLab &amp; SPLab)<\/title>\n<meta name=\"robots\" content=\"index, follow, max-snippet:-1, max-image-preview:large, max-video-preview:-1\" \/>\n<link rel=\"canonical\" href=\"https:\/\/blog.zhaw.ch\/icclab\/understanding-how-server-energy-consumption-varies-with-cpu-bound-workload\/\" \/>\n<meta property=\"og:locale\" content=\"en_US\" \/>\n<meta property=\"og:type\" content=\"article\" \/>\n<meta property=\"og:title\" content=\"Understanding how server energy consumption varies with CPU-bound workload\" \/>\n<meta property=\"og:description\" content=\"In some of our project work, we need to understand the energy consumption of our servers. \u00a0One basic test we performed was to determine how server power consumption increases with load. In this blog post we document the results we obtained. Schlagw\u00f6rter: cpu bound workload, server energy consumption\" \/>\n<meta property=\"og:url\" content=\"https:\/\/blog.zhaw.ch\/icclab\/understanding-how-server-energy-consumption-varies-with-cpu-bound-workload\/\" \/>\n<meta property=\"og:site_name\" content=\"Service Engineering (ICCLab &amp; SPLab)\" \/>\n<meta property=\"article:published_time\" content=\"2014-07-31T09:34:28+00:00\" \/>\n<meta property=\"article:modified_time\" content=\"2014-08-14T08:41:51+00:00\" \/>\n<meta property=\"og:image\" content=\"http:\/\/blog.zhaw.ch\/icclab\/files\/2014\/07\/pical1-300x206.jpg\" \/>\n<meta name=\"author\" content=\"Sean Murphy\" \/>\n<meta name=\"twitter:card\" content=\"summary_large_image\" \/>\n<meta name=\"twitter:label1\" content=\"Written by\" \/>\n\t<meta name=\"twitter:data1\" content=\"Sean Murphy\" \/>\n\t<meta name=\"twitter:label2\" content=\"Est. reading time\" \/>\n\t<meta name=\"twitter:data2\" content=\"4 minutes\" \/>\n<script type=\"application\/ld+json\" class=\"yoast-schema-graph\">{\"@context\":\"https:\/\/schema.org\",\"@graph\":[{\"@type\":\"Article\",\"@id\":\"https:\/\/blog.zhaw.ch\/icclab\/understanding-how-server-energy-consumption-varies-with-cpu-bound-workload\/#article\",\"isPartOf\":{\"@id\":\"https:\/\/blog.zhaw.ch\/icclab\/understanding-how-server-energy-consumption-varies-with-cpu-bound-workload\/\"},\"author\":{\"name\":\"Sean Murphy\",\"@id\":\"https:\/\/blog.zhaw.ch\/icclab\/#\/schema\/person\/c87a6eef7e1f4a152aeec5f8b9527b8d\"},\"headline\":\"Understanding how server energy consumption varies with CPU-bound workload\",\"datePublished\":\"2014-07-31T09:34:28+00:00\",\"dateModified\":\"2014-08-14T08:41:51+00:00\",\"mainEntityOfPage\":{\"@id\":\"https:\/\/blog.zhaw.ch\/icclab\/understanding-how-server-energy-consumption-varies-with-cpu-bound-workload\/\"},\"wordCount\":824,\"commentCount\":0,\"image\":{\"@id\":\"https:\/\/blog.zhaw.ch\/icclab\/understanding-how-server-energy-consumption-varies-with-cpu-bound-workload\/#primaryimage\"},\"thumbnailUrl\":\"http:\/\/blog.zhaw.ch\/icclab\/files\/2014\/07\/pical1-300x206.jpg\",\"keywords\":[\"cpu bound workload\",\"server energy consumption\"],\"articleSection\":[\"Articles\"],\"inLanguage\":\"en-US\",\"potentialAction\":[{\"@type\":\"CommentAction\",\"name\":\"Comment\",\"target\":[\"https:\/\/blog.zhaw.ch\/icclab\/understanding-how-server-energy-consumption-varies-with-cpu-bound-workload\/#respond\"]}]},{\"@type\":\"WebPage\",\"@id\":\"https:\/\/blog.zhaw.ch\/icclab\/understanding-how-server-energy-consumption-varies-with-cpu-bound-workload\/\",\"url\":\"https:\/\/blog.zhaw.ch\/icclab\/understanding-how-server-energy-consumption-varies-with-cpu-bound-workload\/\",\"name\":\"Understanding how server energy consumption varies with CPU-bound workload - Service Engineering (ICCLab &amp; SPLab)\",\"isPartOf\":{\"@id\":\"https:\/\/blog.zhaw.ch\/icclab\/#website\"},\"primaryImageOfPage\":{\"@id\":\"https:\/\/blog.zhaw.ch\/icclab\/understanding-how-server-energy-consumption-varies-with-cpu-bound-workload\/#primaryimage\"},\"image\":{\"@id\":\"https:\/\/blog.zhaw.ch\/icclab\/understanding-how-server-energy-consumption-varies-with-cpu-bound-workload\/#primaryimage\"},\"thumbnailUrl\":\"http:\/\/blog.zhaw.ch\/icclab\/files\/2014\/07\/pical1-300x206.jpg\",\"datePublished\":\"2014-07-31T09:34:28+00:00\",\"dateModified\":\"2014-08-14T08:41:51+00:00\",\"author\":{\"@id\":\"https:\/\/blog.zhaw.ch\/icclab\/#\/schema\/person\/c87a6eef7e1f4a152aeec5f8b9527b8d\"},\"breadcrumb\":{\"@id\":\"https:\/\/blog.zhaw.ch\/icclab\/understanding-how-server-energy-consumption-varies-with-cpu-bound-workload\/#breadcrumb\"},\"inLanguage\":\"en-US\",\"potentialAction\":[{\"@type\":\"ReadAction\",\"target\":[\"https:\/\/blog.zhaw.ch\/icclab\/understanding-how-server-energy-consumption-varies-with-cpu-bound-workload\/\"]}]},{\"@type\":\"ImageObject\",\"inLanguage\":\"en-US\",\"@id\":\"https:\/\/blog.zhaw.ch\/icclab\/understanding-how-server-energy-consumption-varies-with-cpu-bound-workload\/#primaryimage\",\"url\":\"https:\/\/blog.zhaw.ch\/icclab\/files\/2014\/07\/pical1.jpg\",\"contentUrl\":\"https:\/\/blog.zhaw.ch\/icclab\/files\/2014\/07\/pical1.jpg\",\"width\":481,\"height\":331},{\"@type\":\"BreadcrumbList\",\"@id\":\"https:\/\/blog.zhaw.ch\/icclab\/understanding-how-server-energy-consumption-varies-with-cpu-bound-workload\/#breadcrumb\",\"itemListElement\":[{\"@type\":\"ListItem\",\"position\":1,\"name\":\"Startseite\",\"item\":\"https:\/\/blog.zhaw.ch\/icclab\/\"},{\"@type\":\"ListItem\",\"position\":2,\"name\":\"Understanding how server energy consumption varies with CPU-bound workload\"}]},{\"@type\":\"WebSite\",\"@id\":\"https:\/\/blog.zhaw.ch\/icclab\/#website\",\"url\":\"https:\/\/blog.zhaw.ch\/icclab\/\",\"name\":\"Service Engineering (ICCLab &amp; SPLab)\",\"description\":\"A Blog of the ZHAW Zurich University of Applied Sciences\",\"potentialAction\":[{\"@type\":\"SearchAction\",\"target\":{\"@type\":\"EntryPoint\",\"urlTemplate\":\"https:\/\/blog.zhaw.ch\/icclab\/?s={search_term_string}\"},\"query-input\":{\"@type\":\"PropertyValueSpecification\",\"valueRequired\":true,\"valueName\":\"search_term_string\"}}],\"inLanguage\":\"en-US\"},{\"@type\":\"Person\",\"@id\":\"https:\/\/blog.zhaw.ch\/icclab\/#\/schema\/person\/c87a6eef7e1f4a152aeec5f8b9527b8d\",\"name\":\"Sean Murphy\",\"image\":{\"@type\":\"ImageObject\",\"inLanguage\":\"en-US\",\"@id\":\"https:\/\/secure.gravatar.com\/avatar\/4514cb0ddfe236fd05d5ddb715bc19e1e1e35dafa16bb1b911e6094d278211d6?s=96&d=mm&r=g\",\"url\":\"https:\/\/secure.gravatar.com\/avatar\/4514cb0ddfe236fd05d5ddb715bc19e1e1e35dafa16bb1b911e6094d278211d6?s=96&d=mm&r=g\",\"contentUrl\":\"https:\/\/secure.gravatar.com\/avatar\/4514cb0ddfe236fd05d5ddb715bc19e1e1e35dafa16bb1b911e6094d278211d6?s=96&d=mm&r=g\",\"caption\":\"Sean Murphy\"},\"url\":\"https:\/\/blog.zhaw.ch\/icclab\/author\/murp\/\"}]}<\/script>\n<!-- \/ Yoast SEO Premium plugin. -->","yoast_head_json":{"title":"Understanding how server energy consumption varies with CPU-bound workload - Service Engineering (ICCLab &amp; SPLab)","robots":{"index":"index","follow":"follow","max-snippet":"max-snippet:-1","max-image-preview":"max-image-preview:large","max-video-preview":"max-video-preview:-1"},"canonical":"https:\/\/blog.zhaw.ch\/icclab\/understanding-how-server-energy-consumption-varies-with-cpu-bound-workload\/","og_locale":"en_US","og_type":"article","og_title":"Understanding how server energy consumption varies with CPU-bound workload","og_description":"In some of our project work, we need to understand the energy consumption of our servers. \u00a0One basic test we performed was to determine how server power consumption increases with load. In this blog post we document the results we obtained. Schlagw\u00f6rter: cpu bound workload, server energy consumption","og_url":"https:\/\/blog.zhaw.ch\/icclab\/understanding-how-server-energy-consumption-varies-with-cpu-bound-workload\/","og_site_name":"Service Engineering (ICCLab &amp; SPLab)","article_published_time":"2014-07-31T09:34:28+00:00","article_modified_time":"2014-08-14T08:41:51+00:00","og_image":[{"url":"http:\/\/blog.zhaw.ch\/icclab\/files\/2014\/07\/pical1-300x206.jpg","type":"","width":"","height":""}],"author":"Sean Murphy","twitter_card":"summary_large_image","twitter_misc":{"Written by":"Sean Murphy","Est. reading time":"4 minutes"},"schema":{"@context":"https:\/\/schema.org","@graph":[{"@type":"Article","@id":"https:\/\/blog.zhaw.ch\/icclab\/understanding-how-server-energy-consumption-varies-with-cpu-bound-workload\/#article","isPartOf":{"@id":"https:\/\/blog.zhaw.ch\/icclab\/understanding-how-server-energy-consumption-varies-with-cpu-bound-workload\/"},"author":{"name":"Sean Murphy","@id":"https:\/\/blog.zhaw.ch\/icclab\/#\/schema\/person\/c87a6eef7e1f4a152aeec5f8b9527b8d"},"headline":"Understanding how server energy consumption varies with CPU-bound workload","datePublished":"2014-07-31T09:34:28+00:00","dateModified":"2014-08-14T08:41:51+00:00","mainEntityOfPage":{"@id":"https:\/\/blog.zhaw.ch\/icclab\/understanding-how-server-energy-consumption-varies-with-cpu-bound-workload\/"},"wordCount":824,"commentCount":0,"image":{"@id":"https:\/\/blog.zhaw.ch\/icclab\/understanding-how-server-energy-consumption-varies-with-cpu-bound-workload\/#primaryimage"},"thumbnailUrl":"http:\/\/blog.zhaw.ch\/icclab\/files\/2014\/07\/pical1-300x206.jpg","keywords":["cpu bound workload","server energy consumption"],"articleSection":["Articles"],"inLanguage":"en-US","potentialAction":[{"@type":"CommentAction","name":"Comment","target":["https:\/\/blog.zhaw.ch\/icclab\/understanding-how-server-energy-consumption-varies-with-cpu-bound-workload\/#respond"]}]},{"@type":"WebPage","@id":"https:\/\/blog.zhaw.ch\/icclab\/understanding-how-server-energy-consumption-varies-with-cpu-bound-workload\/","url":"https:\/\/blog.zhaw.ch\/icclab\/understanding-how-server-energy-consumption-varies-with-cpu-bound-workload\/","name":"Understanding how server energy consumption varies with CPU-bound workload - Service Engineering (ICCLab &amp; SPLab)","isPartOf":{"@id":"https:\/\/blog.zhaw.ch\/icclab\/#website"},"primaryImageOfPage":{"@id":"https:\/\/blog.zhaw.ch\/icclab\/understanding-how-server-energy-consumption-varies-with-cpu-bound-workload\/#primaryimage"},"image":{"@id":"https:\/\/blog.zhaw.ch\/icclab\/understanding-how-server-energy-consumption-varies-with-cpu-bound-workload\/#primaryimage"},"thumbnailUrl":"http:\/\/blog.zhaw.ch\/icclab\/files\/2014\/07\/pical1-300x206.jpg","datePublished":"2014-07-31T09:34:28+00:00","dateModified":"2014-08-14T08:41:51+00:00","author":{"@id":"https:\/\/blog.zhaw.ch\/icclab\/#\/schema\/person\/c87a6eef7e1f4a152aeec5f8b9527b8d"},"breadcrumb":{"@id":"https:\/\/blog.zhaw.ch\/icclab\/understanding-how-server-energy-consumption-varies-with-cpu-bound-workload\/#breadcrumb"},"inLanguage":"en-US","potentialAction":[{"@type":"ReadAction","target":["https:\/\/blog.zhaw.ch\/icclab\/understanding-how-server-energy-consumption-varies-with-cpu-bound-workload\/"]}]},{"@type":"ImageObject","inLanguage":"en-US","@id":"https:\/\/blog.zhaw.ch\/icclab\/understanding-how-server-energy-consumption-varies-with-cpu-bound-workload\/#primaryimage","url":"https:\/\/blog.zhaw.ch\/icclab\/files\/2014\/07\/pical1.jpg","contentUrl":"https:\/\/blog.zhaw.ch\/icclab\/files\/2014\/07\/pical1.jpg","width":481,"height":331},{"@type":"BreadcrumbList","@id":"https:\/\/blog.zhaw.ch\/icclab\/understanding-how-server-energy-consumption-varies-with-cpu-bound-workload\/#breadcrumb","itemListElement":[{"@type":"ListItem","position":1,"name":"Startseite","item":"https:\/\/blog.zhaw.ch\/icclab\/"},{"@type":"ListItem","position":2,"name":"Understanding how server energy consumption varies with CPU-bound workload"}]},{"@type":"WebSite","@id":"https:\/\/blog.zhaw.ch\/icclab\/#website","url":"https:\/\/blog.zhaw.ch\/icclab\/","name":"Service Engineering (ICCLab &amp; SPLab)","description":"A Blog of the ZHAW Zurich University of Applied Sciences","potentialAction":[{"@type":"SearchAction","target":{"@type":"EntryPoint","urlTemplate":"https:\/\/blog.zhaw.ch\/icclab\/?s={search_term_string}"},"query-input":{"@type":"PropertyValueSpecification","valueRequired":true,"valueName":"search_term_string"}}],"inLanguage":"en-US"},{"@type":"Person","@id":"https:\/\/blog.zhaw.ch\/icclab\/#\/schema\/person\/c87a6eef7e1f4a152aeec5f8b9527b8d","name":"Sean Murphy","image":{"@type":"ImageObject","inLanguage":"en-US","@id":"https:\/\/secure.gravatar.com\/avatar\/4514cb0ddfe236fd05d5ddb715bc19e1e1e35dafa16bb1b911e6094d278211d6?s=96&d=mm&r=g","url":"https:\/\/secure.gravatar.com\/avatar\/4514cb0ddfe236fd05d5ddb715bc19e1e1e35dafa16bb1b911e6094d278211d6?s=96&d=mm&r=g","contentUrl":"https:\/\/secure.gravatar.com\/avatar\/4514cb0ddfe236fd05d5ddb715bc19e1e1e35dafa16bb1b911e6094d278211d6?s=96&d=mm&r=g","caption":"Sean Murphy"},"url":"https:\/\/blog.zhaw.ch\/icclab\/author\/murp\/"}]}},"_links":{"self":[{"href":"https:\/\/blog.zhaw.ch\/icclab\/wp-json\/wp\/v2\/posts\/5563","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/blog.zhaw.ch\/icclab\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/blog.zhaw.ch\/icclab\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/blog.zhaw.ch\/icclab\/wp-json\/wp\/v2\/users\/92"}],"replies":[{"embeddable":true,"href":"https:\/\/blog.zhaw.ch\/icclab\/wp-json\/wp\/v2\/comments?post=5563"}],"version-history":[{"count":6,"href":"https:\/\/blog.zhaw.ch\/icclab\/wp-json\/wp\/v2\/posts\/5563\/revisions"}],"predecessor-version":[{"id":5709,"href":"https:\/\/blog.zhaw.ch\/icclab\/wp-json\/wp\/v2\/posts\/5563\/revisions\/5709"}],"wp:attachment":[{"href":"https:\/\/blog.zhaw.ch\/icclab\/wp-json\/wp\/v2\/media?parent=5563"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/blog.zhaw.ch\/icclab\/wp-json\/wp\/v2\/categories?post=5563"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/blog.zhaw.ch\/icclab\/wp-json\/wp\/v2\/tags?post=5563"},{"taxonomy":"features","embeddable":true,"href":"https:\/\/blog.zhaw.ch\/icclab\/wp-json\/wp\/v2\/features?post=5563"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}