{"id":66012,"date":"2022-01-25T18:46:33","date_gmt":"2022-01-25T15:46:33","guid":{"rendered":"https:\/\/www.altoros.com\/blog\/?p=66012"},"modified":"2022-04-29T16:12:23","modified_gmt":"2022-04-29T13:12:23","slug":"osaka-university-cuts-power-consumption-by-13-with-kubernetes-and-ai","status":"publish","type":"post","link":"https:\/\/www.altoros.com\/blog\/osaka-university-cuts-power-consumption-by-13-with-kubernetes-and-ai\/","title":{"rendered":"Osaka University Cuts Power Consumption by 13% with Kubernetes and AI"},"content":{"rendered":"<div id=\"ez-toc-container\" class=\"ez-toc-v2_0_82_2 counter-hierarchy ez-toc-counter ez-toc-transparent ez-toc-container-direction\">\n<div class=\"ez-toc-title-container\">\n<p class=\"ez-toc-title\" style=\"cursor:inherit\">Table of Contents<\/p>\n<span class=\"ez-toc-title-toggle\"><a href=\"#\" class=\"ez-toc-pull-right ez-toc-btn ez-toc-btn-xs ez-toc-btn-default ez-toc-toggle\" aria-label=\"Toggle Table of Content\"><span class=\"ez-toc-js-icon-con\"><span class=\"\"><span class=\"eztoc-hide\" style=\"display:none;\">Toggle<\/span><span class=\"ez-toc-icon-toggle-span\"><svg style=\"fill: #999;color:#999\" xmlns=\"http:\/\/www.w3.org\/2000\/svg\" class=\"list-377408\" width=\"20px\" height=\"20px\" viewBox=\"0 0 24 24\" fill=\"none\"><path d=\"M6 6H4v2h2V6zm14 0H8v2h12V6zM4 11h2v2H4v-2zm16 0H8v2h12v-2zM4 16h2v2H4v-2zm16 0H8v2h12v-2z\" fill=\"currentColor\"><\/path><\/svg><svg style=\"fill: #999;color:#999\" class=\"arrow-unsorted-368013\" xmlns=\"http:\/\/www.w3.org\/2000\/svg\" width=\"10px\" height=\"10px\" viewBox=\"0 0 24 24\" version=\"1.2\" baseProfile=\"tiny\"><path d=\"M18.2 9.3l-6.2-6.3-6.2 6.3c-.2.2-.3.4-.3.7s.1.5.3.7c.2.2.4.3.7.3h11c.3 0 .5-.1.7-.3.2-.2.3-.5.3-.7s-.1-.5-.3-.7zM5.8 14.7l6.2 6.3 6.2-6.3c.2-.2.3-.5.3-.7s-.1-.5-.3-.7c-.2-.2-.4-.3-.7-.3h-11c-.3 0-.5.1-.7.3-.2.2-.3.5-.3.7s.1.5.3.7z\"\/><\/svg><\/span><\/span><\/span><\/a><\/span><\/div>\n<nav><ul class='ez-toc-list ez-toc-list-level-1 ' ><li class='ez-toc-page-1 ez-toc-heading-level-3'><a class=\"ez-toc-link ez-toc-heading-1\" href=\"https:\/\/www.altoros.com\/blog\/osaka-university-cuts-power-consumption-by-13-with-kubernetes-and-ai\/#Increasing_power_consumption_is_a_problem\" >Increasing power consumption is a problem<\/a><\/li><li class='ez-toc-page-1 ez-toc-heading-level-3'><a class=\"ez-toc-link ez-toc-heading-2\" href=\"https:\/\/www.altoros.com\/blog\/osaka-university-cuts-power-consumption-by-13-with-kubernetes-and-ai\/#What_is_the_workload_allocation_optimizer\" >What is the workload allocation optimizer?<\/a><\/li><li class='ez-toc-page-1 ez-toc-heading-level-3'><a class=\"ez-toc-link ez-toc-heading-3\" href=\"https:\/\/www.altoros.com\/blog\/osaka-university-cuts-power-consumption-by-13-with-kubernetes-and-ai\/#How_is_power_consumption_factored_in\" >How is power consumption factored in?<\/a><\/li><li class='ez-toc-page-1 ez-toc-heading-level-3'><a class=\"ez-toc-link ez-toc-heading-4\" href=\"https:\/\/www.altoros.com\/blog\/osaka-university-cuts-power-consumption-by-13-with-kubernetes-and-ai\/#Creating_a_test_environment\" >Creating a test environment<\/a><\/li><li class='ez-toc-page-1 ez-toc-heading-level-3'><a class=\"ez-toc-link ez-toc-heading-5\" href=\"https:\/\/www.altoros.com\/blog\/osaka-university-cuts-power-consumption-by-13-with-kubernetes-and-ai\/#Test_results\" >Test results<\/a><\/li><li class='ez-toc-page-1 ez-toc-heading-level-3'><a class=\"ez-toc-link ez-toc-heading-6\" href=\"https:\/\/www.altoros.com\/blog\/osaka-university-cuts-power-consumption-by-13-with-kubernetes-and-ai\/#Want_details_Watch_the_video\" >Want details? Watch the video!<\/a><\/li><li class='ez-toc-page-1 ez-toc-heading-level-3'><a class=\"ez-toc-link ez-toc-heading-7\" href=\"https:\/\/www.altoros.com\/blog\/osaka-university-cuts-power-consumption-by-13-with-kubernetes-and-ai\/#Further_reading\" >Further reading<\/a><\/li><li class='ez-toc-page-1 ez-toc-heading-level-3'><a class=\"ez-toc-link ez-toc-heading-8\" href=\"https:\/\/www.altoros.com\/blog\/osaka-university-cuts-power-consumption-by-13-with-kubernetes-and-ai\/#About_the_expert\" >About the expert<\/a><\/li><\/ul><\/nav><\/div>\n<h3><span class=\"ez-toc-section\" id=\"Increasing_power_consumption_is_a_problem\"><\/span>Increasing power consumption is a problem<span class=\"ez-toc-section-end\"><\/span><\/h3>\n<p>Over the past few years, <a href=\"https:\/\/en.wikipedia.org\/wiki\/Edge_computing\" rel=\"noopener noreferrer\" target=\"_blank\">edge computing<\/a>, which places computation and data storage closer to the devices where it is being gathered, has become more prevalent due to the widespread adoption of the Internet of Things (IoT). Edge computing using 5G networks may reduce communication time between devices, but management tasks are becoming more and more complex. Additionally, the steady increase of IoT devices and the growing demand for 5G networking has led to a rise in the amount of computing resources necessary to operate such systems.<\/p>\n<p><center><a href=\"https:\/\/www.altoros.com\/blog\/wp-content\/uploads\/2022\/01\/Osaka-University-Kubernetes-cloud-native-workload-allocation-optimizer-WAO-edge-computing-device-management.png\"><img decoding=\"async\" src=\"https:\/\/www.altoros.com\/blog\/wp-content\/uploads\/2022\/01\/Osaka-University-Kubernetes-cloud-native-workload-allocation-optimizer-WAO-edge-computing-device-management-1024x576.png\" alt=\"\" width=\"640\" class=\"aligncenter size-large wp-image-66019\" \/><\/a><small>Managing multiple devices (<a href=\"https:\/\/static.sched.com\/hosted_files\/kccncna2021\/1a\/A_K8s-Based_Workload_Allocation_Optimizer_for_Minimizing_Power_Consumption_YingfengHsu_101421_v3.1.pdf\" rel=\"noopener noreferrer\" target=\"_blank\">Image credit<\/a>)<\/small><\/center><\/p>\n<p>The problem with managing complex systems can be resolved with <a href=\"https:\/\/www.altoros.com\/blog\/tag\/kubernetes\/\">Kubernetes<\/a>. The open-source container orchestration system enables organizations to simplify the management of large amounts of computing resources. However, even with Kubernetes, the total power consumption due to an increased amount of computing resources remains a key issue.<\/p>\n<p><center><a href=\"https:\/\/www.altoros.com\/blog\/wp-content\/uploads\/2022\/01\/Osaka-University-Kubernetes-cloud-native-workload-allocation-optimizer-WAO-increased-power-consumption.png\"><img decoding=\"async\" src=\"https:\/\/www.altoros.com\/blog\/wp-content\/uploads\/2022\/01\/Osaka-University-Kubernetes-cloud-native-workload-allocation-optimizer-WAO-increased-power-consumption-1024x576.png\" alt=\"\" width=\"640\" class=\"aligncenter size-large wp-image-66020\" \/><\/a><small>Power consumption becoming an issue (<a href=\"https:\/\/static.sched.com\/hosted_files\/kccncna2021\/1a\/A_K8s-Based_Workload_Allocation_Optimizer_for_Minimizing_Power_Consumption_YingfengHsu_101421_v3.1.pdf\" rel=\"noopener noreferrer\" target=\"_blank\">Image credit<\/a>)<\/small><\/center><\/p>\n<p>During the KubeCon and Cloud Native Conference North America 2021, <a href=\"https:\/\/www.crunchbase.com\/person\/ying-feng-hsu\" rel=\"noopener noreferrer\" target=\"_blank\">Ying-Feng Hsu<\/a>, Assistant Professor at Matsuoka Laboratory in Osaka University, shared how the institution is working on addressing power consumption concerns. Matsuoka Laboratory regularly conduct experiments to reduce power consumption. The lab has access to two data centers in Osaka with around 350 servers each. Ying-Feng demonstrated a proof-of-concept design for low power consumption policies for an open-source Kubernetes implementation.<\/p>\n<p>Developed by Ying-Feng along with fellow researchers <a href=\"https:\/\/dl.acm.org\/profile\/99658998442\" rel=\"noopener noreferrer\" target=\"_blank\">Kazuhiro Matsuda<\/a> and <a href=\"https:\/\/dl.acm.org\/profile\/99658745774\" rel=\"noopener noreferrer\" target=\"_blank\">Morito Matsuoka<\/a>, the <a href=\"https:\/\/american-cse.org\/sites\/csci2020proc\/pdfs\/CSCI2020-6SccvdzjqC7bKupZxFmCoA\/762400b269\/762400b269.pdf\" rel=\"noopener noreferrer\" target=\"_blank\">proof-of-concept<\/a> uses a neural network to create a <b>workload allocation optimizer<\/b> (WAO) and extending it through Kubernetes to achieve power consumption reduction in an edge computing system. The proof-of-concept solution is an extension of the <a href=\"https:\/\/ieeexplore.ieee.org\/document\/8814506\" rel=\"noopener noreferrer\" target=\"_blank\">previous experiments<\/a> at Osaka University, where the researchers utilized deep learning to reduce data center power consumption by up to 70%.<\/p>\n<div id=\"attachment_66021\" style=\"width: 160px\" class=\"wp-caption alignright\"><a href=\"https:\/\/www.altoros.com\/blog\/wp-content\/uploads\/2022\/01\/Ying-Feng-Hsu.jpg\"><img loading=\"lazy\" decoding=\"async\" aria-describedby=\"caption-attachment-66021\" src=\"https:\/\/www.altoros.com\/blog\/wp-content\/uploads\/2022\/01\/Ying-Feng-Hsu-150x150.jpg\" alt=\"\" width=\"150\" height=\"150\" class=\"size-thumbnail wp-image-66021\" \/><\/a><p id=\"caption-attachment-66021\" class=\"wp-caption-text\"><small>Ying-Feng Hsu<\/small><\/p><\/div>\n<blockquote><p>&#8220;Kubernetes focuses on high performance container orchestration in large-scale environments. However, Kubernetes does not provide container orchestration from the perspective of power consumption reduction, and there are relatively few discussions in the community on how to operate containers while considering both energy saving and service performance. For example, it would be great if we can consider various application requests, as well as different capacities and performances of computing resources before deploying microservices.&#8221; \u2014Ying-Feng Hsu, Osaka University<\/p><\/blockquote>\n<p>&nbsp;<\/p>\n<h3><span class=\"ez-toc-section\" id=\"What_is_the_workload_allocation_optimizer\"><\/span>What is the workload allocation optimizer?<span class=\"ez-toc-section-end\"><\/span><\/h3>\n<p>To better understand how WAO can reduce power consumption, the process for task allocation in Kubernetes should be explained first. The default task scheduler\u2014Kube-scheduler\u2014uses a pod as the smallest deployable unit. A pod may consist of a single or multiple containers. When distributing pods to nodes, Kubernetes does not provide advanced network load balancers. According to the researchers at Osaka University, when allocating tasks to candidate pods, even with <a href=\"https:\/\/metallb.universe.tf\/\" rel=\"noopener noreferrer\" target=\"_blank\">MetalLB<\/a>, Kubernetes clusters only provide simple load balancers with equal probability.<\/p>\n<p><center><a href=\"https:\/\/www.altoros.com\/blog\/wp-content\/uploads\/2022\/01\/Osaka-University-Kubernetes-cloud-native-workload-allocation-optimizer-WAO-kube-scheduler-and-metallb.png\"><img decoding=\"async\" src=\"https:\/\/www.altoros.com\/blog\/wp-content\/uploads\/2022\/01\/Osaka-University-Kubernetes-cloud-native-workload-allocation-optimizer-WAO-kube-scheduler-and-metallb-1024x576.png\" alt=\"\" width=\"640\" class=\"aligncenter size-large wp-image-66023\" \/><\/a><small>The absence of default power consumption optimization in Kubernetes (<a href=\"https:\/\/static.sched.com\/hosted_files\/kccncna2021\/1a\/A_K8s-Based_Workload_Allocation_Optimizer_for_Minimizing_Power_Consumption_YingfengHsu_101421_v3.1.pdf\" rel=\"noopener noreferrer\" target=\"_blank\">Image credit<\/a>)<\/small><\/center><\/p>\n<p>To reduce power consumption in Kubernetes, it is necessary to prioritize the relationship between power consumption and CPU usage when allocating tasks. To achieve this, the team at Osaka University implemented a WAO-based scheduler (WAO-scheduler) to allocate pods and a WAO-based load balancer (WAO-LB) to allocate tasks.<\/p>\n<p>WAO-scheduler is a custom kube-scheduler that ranks nodes using a neural network\u2013based prediction model. A higher rating for a node indicates that it is expected to have a lower increase in power consumption when employing computing resources. The researchers used WAO-LB to define an evaluation formula based on the concept of <a href=\"https:\/\/en.wikipedia.org\/wiki\/Osmotic_pressure\" rel=\"noopener noreferrer\" target=\"_blank\">osmotic pressure<\/a> and add power consumption (PC) and response time (RT) models using neural networks.<\/p>\n<p><center><a href=\"https:\/\/www.altoros.com\/blog\/wp-content\/uploads\/2022\/01\/Osaka-University-Kubernetes-cloud-native-workload-allocation-optimizer-WAO-scheduler-WAO-LB.png\"><img decoding=\"async\" src=\"https:\/\/www.altoros.com\/blog\/wp-content\/uploads\/2022\/01\/Osaka-University-Kubernetes-cloud-native-workload-allocation-optimizer-WAO-scheduler-WAO-LB-1024x576.png\" alt=\"\" width=\"640\" class=\"aligncenter size-large wp-image-66029\" \/><\/a><small>WAO-scheduler and WAO-LB (<a href=\"https:\/\/static.sched.com\/hosted_files\/kccncna2021\/1a\/A_K8s-Based_Workload_Allocation_Optimizer_for_Minimizing_Power_Consumption_YingfengHsu_101421_v3.1.pdf\" rel=\"noopener noreferrer\" target=\"_blank\">Image credit<\/a>)<\/small><\/center><\/p>\n<blockquote><p>&#8220;The main concept of WAO is to use <a href=\"https:\/\/www.altoros.com\/blog\/tag\/machine-learning\/\">machine learning<\/a> to predict server power consumption and perform optimal task allocations. In other words, we created a power consumption model using machine learning. Based on the model, WAO allocates tasks to the server with the least amount of power consumption.&#8221; \u2014Ying-Feng Hsu, Osaka University<\/p><\/blockquote>\n<p>&nbsp;<\/p>\n<h3><span class=\"ez-toc-section\" id=\"How_is_power_consumption_factored_in\"><\/span>How is power consumption factored in?<span class=\"ez-toc-section-end\"><\/span><\/h3>\n<p>By default, when a pod deployment request is received, kube-scheduler determines to which node each pod in the scheduling queue should be placed based on available resources and constraints. This process includes three primary orderly phases: filtering, scoring, and binding. In the filtering phase, kube-scheduler lists all the available nodes that meet the pod&#8217;s resource requirement. Next, in the scoring phase, the available nodes are ranked based on their scores, which are calculated with default priority functions. The node with the highest score is then selected. Finally, in the binding phase, kube-scheduler, via the API server, notifies the selection of the best node in which a pod is formed.<\/p>\n<p><center><a href=\"https:\/\/www.altoros.com\/blog\/wp-content\/uploads\/2022\/01\/Osaka-University-Kubernetes-cloud-native-workload-allocation-optimizer-WAO-kube-scheduler-process.png\"><img decoding=\"async\" src=\"https:\/\/www.altoros.com\/blog\/wp-content\/uploads\/2022\/01\/Osaka-University-Kubernetes-cloud-native-workload-allocation-optimizer-WAO-kube-scheduler-process-1024x576.png\" alt=\"\" width=\"640\" class=\"aligncenter size-large wp-image-66055\" \/><\/a><small>Scheduling process in kube-scheduler (<a href=\"https:\/\/static.sched.com\/hosted_files\/kccncna2021\/1a\/A_K8s-Based_Workload_Allocation_Optimizer_for_Minimizing_Power_Consumption_YingfengHsu_101421_v3.1.pdf\" rel=\"noopener noreferrer\" target=\"_blank\">Image credit<\/a>)<\/small><\/center><\/p>\n<p>To account for energy savings, the researchers designed a power consumption efficient control component called the <i>Power Consumption\u2013based Scorer<\/i> (PCS) for the WAO-scheduler. The team created a neutral network\u2013based PC model and deployed it to the TensorFlow Serving server in Kubernetes. The following table summarizes the structure and hyperparameters of the PC model.<\/p>\n<p><center><\/p>\n<table>\n<tbody>\n<tr>\n<td valign=\"middle\" colspan=\"2\" width=\"312\" style=\"vertical-align: middle;\">\n<small><center>Input<\/center><\/small>\n<\/td>\n<td width=\"312\">\n<small>CPU usage, temperature around node<\/small>\n<\/td>\n<\/tr>\n<tr>\n<td valign=\"middle\" rowspan=\"4\" width=\"156\" style=\"vertical-align: middle;\">\n<small><center>Neural network model<\/center><\/small>\n<\/td>\n<td width=\"156\">\n<small><center>Layers<\/center><\/small>\n<\/td>\n<td width=\"312\">\n<small>1 hidden layer with 5,000 nodes<\/small>\n<\/td>\n<\/tr>\n<tr>\n<td valign=\"middle\" width=\"156\" style=\"vertical-align: middle;\">\n<small><center>Optimizer<\/center><\/small>\n<\/td>\n<td width=\"312\">\n<small>Adam (with hyperparameters of lr=0.0005, beta1=0.9, beta2=0.999, epsilon=1e-8, decay=0.0)<\/small>\n<\/td>\n<\/tr>\n<tr>\n<td width=\"156\">\n<small><center>Loss function<\/center><\/small>\n<\/td>\n<td width=\"312\">\n<small>mean squared error (MSE)<\/small>\n<\/td>\n<\/tr>\n<tr>\n<td width=\"156\">\n<small><center>Batch size<\/center><\/small>\n<\/td>\n<td width=\"312\">\n<small>256<\/small>\n<\/td>\n<\/tr>\n<tr>\n<td colspan=\"2\" width=\"312\">\n<small><center>Output<\/center><\/small>\n<\/td>\n<td width=\"312\">\n<small>Power consumption of each node<\/small>\n<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n<p><\/center><\/p>\n<p>After taking all the available nodes during the filter phase, PCS first collects information on each node, including resource usage and temperature. Next, PCS predicts the increase of power consumption in  each node via <a href=\"https:\/\/www.altoros.com\/blog\/tag\/TensorFlow\/\">TensorFlow<\/a> and ranks them accordingly. In this manner, WAO-scheduler not only factors in the increased computing resources due to pods deployment, but also the increase in power consumption in each node.<\/p>\n<p><center><a href=\"https:\/\/www.altoros.com\/blog\/wp-content\/uploads\/2022\/01\/Osaka-University-Kubernetes-cloud-native-workload-allocation-optimizer-WAO-scheduler-architecture.png\"><img decoding=\"async\" src=\"https:\/\/www.altoros.com\/blog\/wp-content\/uploads\/2022\/01\/Osaka-University-Kubernetes-cloud-native-workload-allocation-optimizer-WAO-scheduler-architecture-1024x576.png\" alt=\"\" width=\"640\" class=\"aligncenter size-large wp-image-66057\" \/><\/a><small>WAO-scheduler architecture (<a href=\"https:\/\/static.sched.com\/hosted_files\/kccncna2021\/1a\/A_K8s-Based_Workload_Allocation_Optimizer_for_Minimizing_Power_Consumption_YingfengHsu_101421_v3.1.pdf\" rel=\"noopener noreferrer\" target=\"_blank\">Image credit<\/a>)<\/small><\/center><\/p>\n<p>WAO-LB periodically retrieves pod information using kube-apiserver. When receiving a request from a client, WAO-LB first collects information about each pod, such as CPU, memory, and network status from cAdvisor. WAO-LB then predicts the increase in power consumption applying  PC and RT models. The following table summarizes the team&#8217;s implementation of the RT model.<\/p>\n<p><center><\/p>\n<table>\n<tbody>\n<tr>\n<td valign=\"middle\" colspan=\"2\" width=\"312\" style=\"vertical-align: middle;\">\n<small><center>Input<\/center><\/small>\n<\/td>\n<td width=\"312\">\n<small>CPU usage, memory information, network information, temperature around node, date-time information<\/small>\n<\/td>\n<\/tr>\n<tr>\n<td valign=\"middle\" rowspan=\"4\" width=\"156\" style=\"vertical-align: middle;\">\n<small><center>Neural network model<\/center><\/small>\n<\/td>\n<td width=\"156\">\n<small><center>Layers<\/center><\/small>\n<\/td>\n<td width=\"312\">\n<small>3 hidden layers with 2,000 nodes each<\/small>\n<\/td>\n<\/tr>\n<tr>\n<td valign=\"middle\" width=\"156\" style=\"vertical-align: middle;\">\n<small><center>Optimizer<\/center><\/small>\n<\/td>\n<td width=\"312\">\n<small>Adam (with hyperparameters of lr=0.0005, beta1=0.9, beta2=0.999, epsilon=1e-10, decay=0.00001)<\/small>\n<\/td>\n<\/tr>\n<tr>\n<td width=\"156\">\n<small><center>Loss function<\/center><\/small>\n<\/td>\n<td width=\"312\">\n<small>MSE<\/small>\n<\/td>\n<\/tr>\n<tr>\n<td width=\"156\">\n<small><center>Batch size<\/center><\/small>\n<\/td>\n<td width=\"312\">\n<small>256<\/small>\n<\/td>\n<\/tr>\n<tr>\n<td colspan=\"2\" width=\"312\">\n<small><center>Output<\/center><\/small>\n<\/td>\n<td width=\"312\">\n<small>Response time against user&rsquo;s request<\/small>\n<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n<p><\/center><\/p>\n<p>WAO-LB uses both the PC model and the RT model to evaluate the priority of clients&#8217; task allocation to nodes based on power consumption and response times.<\/p>\n<p><center><a href=\"https:\/\/www.altoros.com\/blog\/wp-content\/uploads\/2022\/01\/Osaka-University-Kubernetes-cloud-native-workload-allocation-optimizer-WAO-LB-architecture.png\"><img decoding=\"async\" src=\"https:\/\/www.altoros.com\/blog\/wp-content\/uploads\/2022\/01\/Osaka-University-Kubernetes-cloud-native-workload-allocation-optimizer-WAO-LB-architecture-1024x576.png\" alt=\"\" width=\"640\" class=\"aligncenter size-large wp-image-66056\" \/><\/a><small>The WAO-LB architecture (<a href=\"https:\/\/static.sched.com\/hosted_files\/kccncna2021\/1a\/A_K8s-Based_Workload_Allocation_Optimizer_for_Minimizing_Power_Consumption_YingfengHsu_101421_v3.1.pdf\" rel=\"noopener noreferrer\" target=\"_blank\">Image credit<\/a>)<\/small><\/center><\/p>\n<p>Both the PC and RT machine learning models are based on neural networks. According to the team, neural networks performed the best when compared to other <a href=\"https:\/\/www.altoros.com\/blog\/ml-toolkit-for-tensorflow-out-of-the-box-algorithms-to-boost-training-data-by-50x\/\">machine learning approaches<\/a>, such as random forest and support vector machines.<\/p>\n<p><center><a href=\"https:\/\/www.altoros.com\/blog\/wp-content\/uploads\/2022\/01\/Osaka-University-Kubernetes-cloud-native-workload-allocation-optimizer-WAO-maching-learning-models.png\"><img decoding=\"async\" src=\"https:\/\/www.altoros.com\/blog\/wp-content\/uploads\/2022\/01\/Osaka-University-Kubernetes-cloud-native-workload-allocation-optimizer-WAO-maching-learning-models-1024x576.png\" alt=\"\" width=\"640\" class=\"aligncenter size-large wp-image-66060\" \/><\/a><small>PC and RT models use neural networks (<a href=\"https:\/\/static.sched.com\/hosted_files\/kccncna2021\/1a\/A_K8s-Based_Workload_Allocation_Optimizer_for_Minimizing_Power_Consumption_YingfengHsu_101421_v3.1.pdf\" rel=\"noopener noreferrer\" target=\"_blank\">Image credit<\/a>)<\/small><\/center><\/p>\n<p>&nbsp;<\/p>\n<h3><span class=\"ez-toc-section\" id=\"Creating_a_test_environment\"><\/span>Creating a test environment<span class=\"ez-toc-section-end\"><\/span><\/h3>\n<p>To evaluate WAO, the researchers at Osaka University tested the proof-of-concept solution on a private data center consisting of 200 Fujitsu Primergy RX2530 M4 servers. Each server was equipped with two Intel Xeon Silver 4108 CPUs (8 cores x2), 16 GB of memory, and 1 TB HDD. The edge data center is located in the Konohana Building, Osaka, Japan, of NTT West. To simulate a client in the experiment, the team used a desktop computer that was located roughly 10 km away from the data center.<\/p>\n<p>Next, the team prepared a service that performs object detection based on TensorFlow. Object detection has many use cases in IoT, such as security cameras, <a href=\"https:\/\/www.altoros.com\/blog\/what-it-takes-to-build-and-train-neural-networks-for-autonomous-vehicles\/\">self-driving vehicles<\/a>, and mobile apps. When the service receives a compressed image from a client device, it annotates objects and organisms in the image, compress it again, and sends it back to the client device.<\/p>\n<p>The power consumption value depends on both CPU usage and the temperature around the server. A server&#8217;s power consumption increased significantly when CPU usage was between 10% and 30%, Ying-Feng noted. Temperature also influenced a server&#8217;s power consumption. When temperature is low, fans will rotate at a low speed. On the other hand, when the temperature is high, fans will rotate at a higher speed.<\/p>\n<p><center><a href=\"https:\/\/www.altoros.com\/blog\/wp-content\/uploads\/2022\/01\/Osaka-University-Kubernetes-cloud-native-workload-allocation-optimizer-WAO-power-consumption-based-on-CPU-usage.png\"><img decoding=\"async\" src=\"https:\/\/www.altoros.com\/blog\/wp-content\/uploads\/2022\/01\/Osaka-University-Kubernetes-cloud-native-workload-allocation-optimizer-WAO-power-consumption-based-on-CPU-usage-1024x576.png\" alt=\"\" width=\"640\" class=\"aligncenter size-large wp-image-66067\" \/><\/a><small>The highest power occurring at between 10% and 30% CPU usage (<a href=\"https:\/\/static.sched.com\/hosted_files\/kccncna2021\/1a\/A_K8s-Based_Workload_Allocation_Optimizer_for_Minimizing_Power_Consumption_YingfengHsu_101421_v3.1.pdf\" rel=\"noopener noreferrer\" target=\"_blank\">Image credit<\/a>)<\/small><\/center><\/p>\n<p>With a preset temperature of 24\u00b0C, WAO-scheduler reduced more power consumption compared to other temperatures, as the server fans started to rotate if the temperature fluctuated above or below 24\u00b0C. Thus, the team went with a fixed temperature parameter of 24\u00b0C for both WAO-scheduler and WAO-LB.<\/p>\n<p><center><a href=\"https:\/\/www.altoros.com\/blog\/wp-content\/uploads\/2022\/01\/Osaka-University-Kubernetes-cloud-native-workload-allocation-optimizer-WAO-power-consumption-based-on-temperature.png\"><img decoding=\"async\" src=\"https:\/\/www.altoros.com\/blog\/wp-content\/uploads\/2022\/01\/Osaka-University-Kubernetes-cloud-native-workload-allocation-optimizer-WAO-power-consumption-based-on-temperature-1024x576.png\" alt=\"\" width=\"640\" class=\"aligncenter size-large wp-image-66068\" \/><\/a><small>24\u00b0C as the optimal temperature for evaluation (<a href=\"https:\/\/static.sched.com\/hosted_files\/kccncna2021\/1a\/A_K8s-Based_Workload_Allocation_Optimizer_for_Minimizing_Power_Consumption_YingfengHsu_101421_v3.1.pdf\" rel=\"noopener noreferrer\" target=\"_blank\">Image credit<\/a>)<\/small><\/center><\/p>\n<p>WAO-LB uses an evaluation formula that factors in both power consumption and response times. The researchers define the formula as <b>Evaluation Value = &alpha;*PC + &beta;*RT<\/b>, where PC and RT are power consumption and response time indices. &alpha; and &beta; are weights of each index (&alpha; + &beta; = 1).<\/p>\n<blockquote><p>&#8220;The parameter of &alpha; and &beta; in the evaluation are weights on indices. These values can change depending on the application requirements. For example, applications related to self-driving, low response time is critical. On the other hand, for non-real-time related applications, lowering power consumption can be prioritized. In our experiment, we chose an object detection application, and we observed that a correlation of -0.569 between increased power consumption and response time [was optimal].&#8221; \u2014Ying-Feng Hsu, Osaka University<\/p><\/blockquote>\n<p>&nbsp;<\/p>\n<h3><span class=\"ez-toc-section\" id=\"Test_results\"><\/span>Test results<span class=\"ez-toc-section-end\"><\/span><\/h3>\n<p>In order to evaluate WAO, the researchers initially performed the test using kube-scheduler and MetalLB to get a baseline measurement. Next, the team ran the test under three different scenarios.<\/p>\n<ol>\n<li>WAO-scheduler and MetalLB<\/li>\n<li>kube-scheduler and WAO-LB<\/li>\n<li>WAO-scheduler and WAO-LB<\/li>\n<\/ol>\n<p>In the <i>first scenario<\/i>, the combination of WAO-scheduler and MetalLB saved power regardless of the number of allocated pods. The team observed an 8% reduction to power consumption with 10 pods that are utilizing 20% of the CPU.<\/p>\n<p><center><a href=\"https:\/\/www.altoros.com\/blog\/wp-content\/uploads\/2022\/01\/Osaka-University-Kubernetes-cloud-native-workload-allocation-optimizer-WAO-scheduler-metallb-evaluation.png\"><img decoding=\"async\" src=\"https:\/\/www.altoros.com\/blog\/wp-content\/uploads\/2022\/01\/Osaka-University-Kubernetes-cloud-native-workload-allocation-optimizer-WAO-scheduler-metallb-evaluation-1024x576.png\" alt=\"\" width=\"640\" class=\"aligncenter size-large wp-image-66138\" \/><\/a><small>WAO-scheduler and MetalLB (<a href=\"https:\/\/static.sched.com\/hosted_files\/kccncna2021\/1a\/A_K8s-Based_Workload_Allocation_Optimizer_for_Minimizing_Power_Consumption_YingfengHsu_101421_v3.1.pdf\" rel=\"noopener noreferrer\" target=\"_blank\">Image credit<\/a>)<\/small><\/center><\/p>\n<p>In the <i>second scenario<\/i>, the combination of kube-scheduler and WAO-LB saved power, except when response time is heavily weighted. When power consumption is prioritized, the team saw a 9.9% reduction to power consumption at a total CPU usage of 20%.<\/p>\n<p><center><a href=\"https:\/\/www.altoros.com\/blog\/wp-content\/uploads\/2022\/01\/Osaka-University-Kubernetes-cloud-native-workload-allocation-optimizer-kube-scheduler-wao-lb-evaluation.png\"><img decoding=\"async\" src=\"https:\/\/www.altoros.com\/blog\/wp-content\/uploads\/2022\/01\/Osaka-University-Kubernetes-cloud-native-workload-allocation-optimizer-kube-scheduler-wao-lb-evaluation-1024x576.png\" alt=\"\" width=\"640\" class=\"aligncenter size-large wp-image-66137\" \/><\/a><small>kube-scheduler and WAO-LB (<a href=\"https:\/\/static.sched.com\/hosted_files\/kccncna2021\/1a\/A_K8s-Based_Workload_Allocation_Optimizer_for_Minimizing_Power_Consumption_YingfengHsu_101421_v3.1.pdf\" rel=\"noopener noreferrer\" target=\"_blank\">Image credit<\/a>)<\/small><\/center><\/p>\n<p>In the <i>third scenario<\/i>, the complete WAO solution was able to achieve a 13% reduction to power consumption at a total CPU usage of 27%.<\/p>\n<p><center><a href=\"https:\/\/www.altoros.com\/blog\/wp-content\/uploads\/2022\/01\/Osaka-University-Kubernetes-cloud-native-workload-allocation-optimizer-WAO-scheduler-wao-lb-evaluation.png\"><img decoding=\"async\" src=\"https:\/\/www.altoros.com\/blog\/wp-content\/uploads\/2022\/01\/Osaka-University-Kubernetes-cloud-native-workload-allocation-optimizer-WAO-scheduler-wao-lb-evaluation-1024x576.png\" alt=\"\" width=\"640\" class=\"aligncenter size-large wp-image-66136\" \/><\/a><small>WAO-scheduler and WAO-LB (<a href=\"https:\/\/static.sched.com\/hosted_files\/kccncna2021\/1a\/A_K8s-Based_Workload_Allocation_Optimizer_for_Minimizing_Power_Consumption_YingfengHsu_101421_v3.1.pdf\" rel=\"noopener noreferrer\" target=\"_blank\">Image credit<\/a>)<\/small><\/center><\/p>\n<p>After the tests, the researches concluded that WAO-scheduler and WAO-LB can achieve the highest power consumption reduction rates at a total CPU usage between 15% and 39%.<\/p>\n<p><center><a href=\"https:\/\/www.altoros.com\/blog\/wp-content\/uploads\/2022\/01\/Osaka-University-Kubernetes-cloud-native-workload-allocation-optimizer-WAO-scheduler-wao-lb-power-reduction.png\"><img decoding=\"async\" src=\"https:\/\/www.altoros.com\/blog\/wp-content\/uploads\/2022\/01\/Osaka-University-Kubernetes-cloud-native-workload-allocation-optimizer-WAO-scheduler-wao-lb-power-reduction-1024x576.png\" alt=\"\" width=\"640\" class=\"aligncenter size-large wp-image-66135\" \/><\/a><small>Power consumption rate (<a href=\"https:\/\/static.sched.com\/hosted_files\/kccncna2021\/1a\/A_K8s-Based_Workload_Allocation_Optimizer_for_Minimizing_Power_Consumption_YingfengHsu_101421_v3.1.pdf\" rel=\"noopener noreferrer\" target=\"_blank\">Image credit<\/a>)<\/small><\/center><\/p>\n<blockquote><p>&#8220;In general, [CPU] utilization in a typical data center is often between 20% and 40%. This result tells us that the proposed Kubernetes-based WAO can achieve the highest power consumption reduction under a common data center CPU usage scenario.&#8221; \u2014Ying-Feng Hsu, Osaka University<\/p><\/blockquote>\n<p>Besides the reduction in power consumption, WAO was able to achieve faster response times, except when power consumption is heavily weighted.<\/p>\n<p><center><a href=\"https:\/\/www.altoros.com\/blog\/wp-content\/uploads\/2022\/01\/Osaka-University-Kubernetes-cloud-native-workload-allocation-optimizer-WAO-scheduler-wao-lb-response-time.png\"><img decoding=\"async\" src=\"https:\/\/www.altoros.com\/blog\/wp-content\/uploads\/2022\/01\/Osaka-University-Kubernetes-cloud-native-workload-allocation-optimizer-WAO-scheduler-wao-lb-response-time-1024x576.png\" alt=\"\" width=\"640\" class=\"aligncenter size-large wp-image-66139\" \/><\/a><small>Response time test results (<a href=\"https:\/\/static.sched.com\/hosted_files\/kccncna2021\/1a\/A_K8s-Based_Workload_Allocation_Optimizer_for_Minimizing_Power_Consumption_YingfengHsu_101421_v3.1.pdf\" rel=\"noopener noreferrer\" target=\"_blank\">Image credit<\/a>)<\/small><\/center><\/p>\n<p>As edge computing becomes more widespread, power consumption in data centers will continue to increase. Solutions like WAO can provide organizations with a tool for managing resources and costs related to power consumption. Read more about WAO in the <a href=\"https:\/\/american-cse.org\/sites\/csci2020proc\/pdfs\/CSCI2020-6SccvdzjqC7bKupZxFmCoA\/762400b269\/762400b269.pdf\" rel=\"noopener noreferrer\" target=\"_blank\">Osaka University paper<\/a>.<\/p>\n<p>&nbsp;<\/p>\n<h3><span class=\"ez-toc-section\" id=\"Want_details_Watch_the_video\"><\/span>Want details? Watch the video!<span class=\"ez-toc-section-end\"><\/span><\/h3>\n<p><small><a href=\"https:\/\/www.crunchbase.com\/person\/ying-feng-hsu\" rel=\"noopener noreferrer\" target=\"_blank\">Ying-Feng Hsu<\/a> provides an overview of the workload allocation optimizer and how it can reduce power consumptions in data centers.<\/small><\/p>\n<p><center><iframe loading=\"lazy\" width=\"560\" height=\"315\" src=\"https:\/\/www.youtube.com\/embed\/TXa1lj7FIZA\" title=\"YouTube video player\" frameborder=\"0\" allow=\"accelerometer; autoplay; clipboard-write; encrypted-media; gyroscope; picture-in-picture\" allowfullscreen><\/iframe><\/center><\/p>\n<p>&nbsp;<\/p>\n<h3><span class=\"ez-toc-section\" id=\"Further_reading\"><\/span>Further reading<span class=\"ez-toc-section-end\"><\/span><\/h3>\n<ul>\n<li><a href=\"https:\/\/www.altoros.com\/blog\/shell-builds-10000-ai-models-on-kubernetes-in-less-than-a-day\/\">Shell Builds 10,000 AI Models on Kubernetes in Less than a Day<\/a><\/li>\n<li><a href=\"https:\/\/www.altoros.com\/blog\/denso-delivers-an-iot-prototype-per-week-with-kubernetes\/\">Denso Delivers an IoT Prototype per Week with Kubernetes<\/a><\/li>\n<li><a href=\"https:\/\/www.altoros.com\/blog\/the-pompeii-museum-develops-a-mobile-app-on-kubernetes-in-six-weeks\/\">The Pompeii Museum Develops a Mobile App on Kubernetes in Six Weeks<\/a><\/li>\n<\/ul>\n<p>&nbsp;<\/p>\n<h3><span class=\"ez-toc-section\" id=\"About_the_expert\"><\/span>About the expert<span class=\"ez-toc-section-end\"><\/span><\/h3>\n<div style=\"padding-left: 0px; padding-bottom: 30px; padding-right: 0px; padding-top: 0px;\">\n<p><a href=\"https:\/\/www.altoros.com\/blog\/wp-content\/uploads\/2022\/01\/Ying-Feng-Hsu-bio.png\"><img decoding=\"async\" src=\"https:\/\/www.altoros.com\/blog\/wp-content\/uploads\/2022\/01\/Ying-Feng-Hsu-bio-150x150.png\" alt=\"\" width=\"120\" class=\"alignright size-thumbnail wp-image-66156\" \/><\/a><\/p>\n<p><small><a href=\"https:\/\/www.crunchbase.com\/person\/ying-feng-hsu\" rel=\"noopener noreferrer\" target=\"_blank\">Ying-Feng Hsu<\/a> is Assistant Professor at Matsuoka Laboratory, Osaka University. His research revolves around machine learning and cloud computing, with a special focus on the use of data center operation, power consumption optimization, and network intrusion detection systems. In the past, Ying-Feng has worked as a data science expert on national projects, such as Pediatric Health Information System (PHIS) and Project Tycho.<\/small>\n<\/div>\n<hr \/>\n<p><center><small>This blog post was written by <a href=\"https:\/\/www.altoros.com\/blog\/author\/carlo\/\">Carlo Gutierrez<\/a>, edited by <a href=\"https:\/\/www.altoros.com\/blog\/author\/sophie.turol\/\">Sophia Turol<\/a> and <a href=\"https:\/\/www.altoros.com\/blog\/author\/alex\/\">Alex Khizhniak<\/a>.<\/small><\/center><\/p>\n","protected":false},"excerpt":{"rendered":"<p>Increasing power consumption is a problem<\/p>\n<p>Over the past few years, edge computing, which places computation and data storage closer to the devices where it is being gathered, has become more prevalent due to the widespread adoption of the Internet of Things (IoT). Edge computing using 5G networks may reduce communication [&#8230;]<\/p>\n","protected":false},"author":32,"featured_media":66153,"comment_status":"open","ping_status":"closed","sticky":false,"template":"","format":"standard","meta":{"_acf_changed":false,"content-type":"","footnotes":"","_links_to":"","_links_to_target":""},"categories":[26],"tags":[873,815,117,912,748],"class_list":["post-66012","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-use-cases","tag-cloud-native","tag-digital-transformation","tag-iot","tag-kubernetes","tag-machine-learning"],"acf":[],"yoast_head":"<!-- This site is optimized with the Yoast SEO plugin v27.4 - https:\/\/yoast.com\/product\/yoast-seo-wordpress\/ -->\n<title>Osaka University Cuts Power Consumption by 13% with Kubernetes and AI | Altoros<\/title>\n<meta name=\"description\" content=\"Relying on deep learning, the institution builds a workload allocation optimizer using neural network\u2014to enable efficient edge computing.\" \/>\n<meta name=\"robots\" content=\"index, follow, max-snippet:-1, max-image-preview:large, max-video-preview:-1\" \/>\n<link rel=\"canonical\" href=\"https:\/\/www.altoros.com\/blog\/osaka-university-cuts-power-consumption-by-13-with-kubernetes-and-ai\/\" \/>\n<meta property=\"og:locale\" content=\"en_US\" \/>\n<meta property=\"og:type\" content=\"article\" \/>\n<meta property=\"og:title\" content=\"Osaka University Cuts Power Consumption by 13% with Kubernetes and AI | Altoros\" \/>\n<meta property=\"og:description\" content=\"Increasing power consumption is a problem Over the past few years, edge computing, which places computation and data storage closer to the devices where it is being gathered, has become more prevalent due to the widespread adoption of the Internet of Things (IoT). Edge computing using 5G networks may reduce communication [...]\" \/>\n<meta property=\"og:url\" content=\"https:\/\/www.altoros.com\/blog\/osaka-university-cuts-power-consumption-by-13-with-kubernetes-and-ai\/\" \/>\n<meta property=\"og:site_name\" content=\"Altoros\" \/>\n<meta property=\"article:published_time\" content=\"2022-01-25T15:46:33+00:00\" \/>\n<meta property=\"article:modified_time\" content=\"2022-04-29T13:12:23+00:00\" \/>\n<meta property=\"og:image\" content=\"https:\/\/www.altoros.com\/blog\/wp-content\/uploads\/2022\/01\/Osaka-University-Kubernetes-cloud-native-workload-allocation-optimizer-WAO-scheduler-WAO-LB-power-consumption-reduction-digital-transformation.gif\" \/>\n\t<meta property=\"og:image:width\" content=\"1024\" \/>\n\t<meta property=\"og:image:height\" content=\"576\" \/>\n\t<meta property=\"og:image:type\" content=\"image\/gif\" \/>\n<meta name=\"author\" content=\"Carlo Gutierrez\" \/>\n<meta name=\"twitter:label1\" content=\"Written by\" \/>\n\t<meta name=\"twitter:data1\" content=\"Carlo Gutierrez\" \/>\n\t<meta name=\"twitter:label2\" content=\"Est. reading time\" \/>\n\t<meta name=\"twitter:data2\" content=\"10 minutes\" \/>\n<script type=\"application\/ld+json\" class=\"yoast-schema-graph\">{\"@context\":\"https:\\\/\\\/schema.org\",\"@graph\":[{\"@type\":\"Article\",\"@id\":\"https:\\\/\\\/www.altoros.com\\\/blog\\\/osaka-university-cuts-power-consumption-by-13-with-kubernetes-and-ai\\\/#article\",\"isPartOf\":{\"@id\":\"https:\\\/\\\/www.altoros.com\\\/blog\\\/osaka-university-cuts-power-consumption-by-13-with-kubernetes-and-ai\\\/\"},\"author\":{\"name\":\"Carlo Gutierrez\",\"@id\":\"https:\\\/\\\/www.altoros.com\\\/blog\\\/#\\\/schema\\\/person\\\/833e109f77de753b2b472dca0236b442\"},\"headline\":\"Osaka University Cuts Power Consumption by 13% with Kubernetes and AI\",\"datePublished\":\"2022-01-25T15:46:33+00:00\",\"dateModified\":\"2022-04-29T13:12:23+00:00\",\"mainEntityOfPage\":{\"@id\":\"https:\\\/\\\/www.altoros.com\\\/blog\\\/osaka-university-cuts-power-consumption-by-13-with-kubernetes-and-ai\\\/\"},\"wordCount\":1955,\"commentCount\":0,\"image\":{\"@id\":\"https:\\\/\\\/www.altoros.com\\\/blog\\\/osaka-university-cuts-power-consumption-by-13-with-kubernetes-and-ai\\\/#primaryimage\"},\"thumbnailUrl\":\"https:\\\/\\\/www.altoros.com\\\/blog\\\/wp-content\\\/uploads\\\/2022\\\/01\\\/Osaka-University-Kubernetes-cloud-native-workload-allocation-optimizer-WAO-scheduler-WAO-LB-power-consumption-reduction-digital-transformation.gif\",\"keywords\":[\"Cloud-Native\",\"Digital Transformation\",\"IoT\",\"Kubernetes\",\"Machine Learning\"],\"articleSection\":[\"Use Cases\"],\"inLanguage\":\"en-US\",\"potentialAction\":[{\"@type\":\"CommentAction\",\"name\":\"Comment\",\"target\":[\"https:\\\/\\\/www.altoros.com\\\/blog\\\/osaka-university-cuts-power-consumption-by-13-with-kubernetes-and-ai\\\/#respond\"]}]},{\"@type\":\"WebPage\",\"@id\":\"https:\\\/\\\/www.altoros.com\\\/blog\\\/osaka-university-cuts-power-consumption-by-13-with-kubernetes-and-ai\\\/\",\"url\":\"https:\\\/\\\/www.altoros.com\\\/blog\\\/osaka-university-cuts-power-consumption-by-13-with-kubernetes-and-ai\\\/\",\"name\":\"Osaka University Cuts Power Consumption by 13% with Kubernetes and AI | Altoros\",\"isPartOf\":{\"@id\":\"https:\\\/\\\/www.altoros.com\\\/blog\\\/#website\"},\"primaryImageOfPage\":{\"@id\":\"https:\\\/\\\/www.altoros.com\\\/blog\\\/osaka-university-cuts-power-consumption-by-13-with-kubernetes-and-ai\\\/#primaryimage\"},\"image\":{\"@id\":\"https:\\\/\\\/www.altoros.com\\\/blog\\\/osaka-university-cuts-power-consumption-by-13-with-kubernetes-and-ai\\\/#primaryimage\"},\"thumbnailUrl\":\"https:\\\/\\\/www.altoros.com\\\/blog\\\/wp-content\\\/uploads\\\/2022\\\/01\\\/Osaka-University-Kubernetes-cloud-native-workload-allocation-optimizer-WAO-scheduler-WAO-LB-power-consumption-reduction-digital-transformation.gif\",\"datePublished\":\"2022-01-25T15:46:33+00:00\",\"dateModified\":\"2022-04-29T13:12:23+00:00\",\"author\":{\"@id\":\"https:\\\/\\\/www.altoros.com\\\/blog\\\/#\\\/schema\\\/person\\\/833e109f77de753b2b472dca0236b442\"},\"breadcrumb\":{\"@id\":\"https:\\\/\\\/www.altoros.com\\\/blog\\\/osaka-university-cuts-power-consumption-by-13-with-kubernetes-and-ai\\\/#breadcrumb\"},\"inLanguage\":\"en-US\",\"potentialAction\":[{\"@type\":\"ReadAction\",\"target\":[\"https:\\\/\\\/www.altoros.com\\\/blog\\\/osaka-university-cuts-power-consumption-by-13-with-kubernetes-and-ai\\\/\"]}]},{\"@type\":\"ImageObject\",\"inLanguage\":\"en-US\",\"@id\":\"https:\\\/\\\/www.altoros.com\\\/blog\\\/osaka-university-cuts-power-consumption-by-13-with-kubernetes-and-ai\\\/#primaryimage\",\"url\":\"https:\\\/\\\/www.altoros.com\\\/blog\\\/wp-content\\\/uploads\\\/2022\\\/01\\\/Osaka-University-Kubernetes-cloud-native-workload-allocation-optimizer-WAO-scheduler-WAO-LB-power-consumption-reduction-digital-transformation.gif\",\"contentUrl\":\"https:\\\/\\\/www.altoros.com\\\/blog\\\/wp-content\\\/uploads\\\/2022\\\/01\\\/Osaka-University-Kubernetes-cloud-native-workload-allocation-optimizer-WAO-scheduler-WAO-LB-power-consumption-reduction-digital-transformation.gif\",\"width\":1024,\"height\":576},{\"@type\":\"BreadcrumbList\",\"@id\":\"https:\\\/\\\/www.altoros.com\\\/blog\\\/osaka-university-cuts-power-consumption-by-13-with-kubernetes-and-ai\\\/#breadcrumb\",\"itemListElement\":[{\"@type\":\"ListItem\",\"position\":1,\"name\":\"Home\",\"item\":\"https:\\\/\\\/www.altoros.com\\\/blog\\\/\"},{\"@type\":\"ListItem\",\"position\":2,\"name\":\"Osaka University Cuts Power Consumption by 13% with Kubernetes and AI\"}]},{\"@type\":\"WebSite\",\"@id\":\"https:\\\/\\\/www.altoros.com\\\/blog\\\/#website\",\"url\":\"https:\\\/\\\/www.altoros.com\\\/blog\\\/\",\"name\":\"Altoros\",\"description\":\"Insight\",\"potentialAction\":[{\"@type\":\"SearchAction\",\"target\":{\"@type\":\"EntryPoint\",\"urlTemplate\":\"https:\\\/\\\/www.altoros.com\\\/blog\\\/?s={search_term_string}\"},\"query-input\":{\"@type\":\"PropertyValueSpecification\",\"valueRequired\":true,\"valueName\":\"search_term_string\"}}],\"inLanguage\":\"en-US\"},{\"@type\":\"Person\",\"@id\":\"https:\\\/\\\/www.altoros.com\\\/blog\\\/#\\\/schema\\\/person\\\/833e109f77de753b2b472dca0236b442\",\"name\":\"Carlo Gutierrez\",\"image\":{\"@type\":\"ImageObject\",\"inLanguage\":\"en-US\",\"@id\":\"https:\\\/\\\/www.altoros.com\\\/blog\\\/wp-content\\\/uploads\\\/2021\\\/02\\\/CG_portrait-2-96x96.jpg\",\"url\":\"https:\\\/\\\/www.altoros.com\\\/blog\\\/wp-content\\\/uploads\\\/2021\\\/02\\\/CG_portrait-2-96x96.jpg\",\"contentUrl\":\"https:\\\/\\\/www.altoros.com\\\/blog\\\/wp-content\\\/uploads\\\/2021\\\/02\\\/CG_portrait-2-96x96.jpg\",\"caption\":\"Carlo Gutierrez\"},\"description\":\"Carlo Gutierrez is a Technical Writer at Altoros. As part of the editorial team, his focus has been on emerging technologies such as Cloud Foundry, Kubernetes, blockchain, and the Internet of Things. Prior to Altoros, he primarily wrote about enterprise and consumer technology. Carlo has over 12 years of experience in the publishing industry. Previously, he served as an Editor for PC World Philippines and Questex Asia, as well as a Designer for Tropa Entertainment.\",\"url\":\"https:\\\/\\\/www.altoros.com\\\/blog\\\/author\\\/carlo\\\/\"}]}<\/script>\n<!-- \/ Yoast SEO plugin. -->","yoast_head_json":{"title":"Osaka University Cuts Power Consumption by 13% with Kubernetes and AI | Altoros","description":"Relying on deep learning, the institution builds a workload allocation optimizer using neural network\u2014to enable efficient edge computing.","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:\/\/www.altoros.com\/blog\/osaka-university-cuts-power-consumption-by-13-with-kubernetes-and-ai\/","og_locale":"en_US","og_type":"article","og_title":"Osaka University Cuts Power Consumption by 13% with Kubernetes and AI | Altoros","og_description":"Increasing power consumption is a problem Over the past few years, edge computing, which places computation and data storage closer to the devices where it is being gathered, has become more prevalent due to the widespread adoption of the Internet of Things (IoT). Edge computing using 5G networks may reduce communication [...]","og_url":"https:\/\/www.altoros.com\/blog\/osaka-university-cuts-power-consumption-by-13-with-kubernetes-and-ai\/","og_site_name":"Altoros","article_published_time":"2022-01-25T15:46:33+00:00","article_modified_time":"2022-04-29T13:12:23+00:00","og_image":[{"width":1024,"height":576,"url":"https:\/\/www.altoros.com\/blog\/wp-content\/uploads\/2022\/01\/Osaka-University-Kubernetes-cloud-native-workload-allocation-optimizer-WAO-scheduler-WAO-LB-power-consumption-reduction-digital-transformation.gif","type":"image\/gif"}],"author":"Carlo Gutierrez","twitter_misc":{"Written by":"Carlo Gutierrez","Est. reading time":"10 minutes"},"schema":{"@context":"https:\/\/schema.org","@graph":[{"@type":"Article","@id":"https:\/\/www.altoros.com\/blog\/osaka-university-cuts-power-consumption-by-13-with-kubernetes-and-ai\/#article","isPartOf":{"@id":"https:\/\/www.altoros.com\/blog\/osaka-university-cuts-power-consumption-by-13-with-kubernetes-and-ai\/"},"author":{"name":"Carlo Gutierrez","@id":"https:\/\/www.altoros.com\/blog\/#\/schema\/person\/833e109f77de753b2b472dca0236b442"},"headline":"Osaka University Cuts Power Consumption by 13% with Kubernetes and AI","datePublished":"2022-01-25T15:46:33+00:00","dateModified":"2022-04-29T13:12:23+00:00","mainEntityOfPage":{"@id":"https:\/\/www.altoros.com\/blog\/osaka-university-cuts-power-consumption-by-13-with-kubernetes-and-ai\/"},"wordCount":1955,"commentCount":0,"image":{"@id":"https:\/\/www.altoros.com\/blog\/osaka-university-cuts-power-consumption-by-13-with-kubernetes-and-ai\/#primaryimage"},"thumbnailUrl":"https:\/\/www.altoros.com\/blog\/wp-content\/uploads\/2022\/01\/Osaka-University-Kubernetes-cloud-native-workload-allocation-optimizer-WAO-scheduler-WAO-LB-power-consumption-reduction-digital-transformation.gif","keywords":["Cloud-Native","Digital Transformation","IoT","Kubernetes","Machine Learning"],"articleSection":["Use Cases"],"inLanguage":"en-US","potentialAction":[{"@type":"CommentAction","name":"Comment","target":["https:\/\/www.altoros.com\/blog\/osaka-university-cuts-power-consumption-by-13-with-kubernetes-and-ai\/#respond"]}]},{"@type":"WebPage","@id":"https:\/\/www.altoros.com\/blog\/osaka-university-cuts-power-consumption-by-13-with-kubernetes-and-ai\/","url":"https:\/\/www.altoros.com\/blog\/osaka-university-cuts-power-consumption-by-13-with-kubernetes-and-ai\/","name":"Osaka University Cuts Power Consumption by 13% with Kubernetes and AI | Altoros","isPartOf":{"@id":"https:\/\/www.altoros.com\/blog\/#website"},"primaryImageOfPage":{"@id":"https:\/\/www.altoros.com\/blog\/osaka-university-cuts-power-consumption-by-13-with-kubernetes-and-ai\/#primaryimage"},"image":{"@id":"https:\/\/www.altoros.com\/blog\/osaka-university-cuts-power-consumption-by-13-with-kubernetes-and-ai\/#primaryimage"},"thumbnailUrl":"https:\/\/www.altoros.com\/blog\/wp-content\/uploads\/2022\/01\/Osaka-University-Kubernetes-cloud-native-workload-allocation-optimizer-WAO-scheduler-WAO-LB-power-consumption-reduction-digital-transformation.gif","datePublished":"2022-01-25T15:46:33+00:00","dateModified":"2022-04-29T13:12:23+00:00","author":{"@id":"https:\/\/www.altoros.com\/blog\/#\/schema\/person\/833e109f77de753b2b472dca0236b442"},"breadcrumb":{"@id":"https:\/\/www.altoros.com\/blog\/osaka-university-cuts-power-consumption-by-13-with-kubernetes-and-ai\/#breadcrumb"},"inLanguage":"en-US","potentialAction":[{"@type":"ReadAction","target":["https:\/\/www.altoros.com\/blog\/osaka-university-cuts-power-consumption-by-13-with-kubernetes-and-ai\/"]}]},{"@type":"ImageObject","inLanguage":"en-US","@id":"https:\/\/www.altoros.com\/blog\/osaka-university-cuts-power-consumption-by-13-with-kubernetes-and-ai\/#primaryimage","url":"https:\/\/www.altoros.com\/blog\/wp-content\/uploads\/2022\/01\/Osaka-University-Kubernetes-cloud-native-workload-allocation-optimizer-WAO-scheduler-WAO-LB-power-consumption-reduction-digital-transformation.gif","contentUrl":"https:\/\/www.altoros.com\/blog\/wp-content\/uploads\/2022\/01\/Osaka-University-Kubernetes-cloud-native-workload-allocation-optimizer-WAO-scheduler-WAO-LB-power-consumption-reduction-digital-transformation.gif","width":1024,"height":576},{"@type":"BreadcrumbList","@id":"https:\/\/www.altoros.com\/blog\/osaka-university-cuts-power-consumption-by-13-with-kubernetes-and-ai\/#breadcrumb","itemListElement":[{"@type":"ListItem","position":1,"name":"Home","item":"https:\/\/www.altoros.com\/blog\/"},{"@type":"ListItem","position":2,"name":"Osaka University Cuts Power Consumption by 13% with Kubernetes and AI"}]},{"@type":"WebSite","@id":"https:\/\/www.altoros.com\/blog\/#website","url":"https:\/\/www.altoros.com\/blog\/","name":"Altoros","description":"Insight","potentialAction":[{"@type":"SearchAction","target":{"@type":"EntryPoint","urlTemplate":"https:\/\/www.altoros.com\/blog\/?s={search_term_string}"},"query-input":{"@type":"PropertyValueSpecification","valueRequired":true,"valueName":"search_term_string"}}],"inLanguage":"en-US"},{"@type":"Person","@id":"https:\/\/www.altoros.com\/blog\/#\/schema\/person\/833e109f77de753b2b472dca0236b442","name":"Carlo Gutierrez","image":{"@type":"ImageObject","inLanguage":"en-US","@id":"https:\/\/www.altoros.com\/blog\/wp-content\/uploads\/2021\/02\/CG_portrait-2-96x96.jpg","url":"https:\/\/www.altoros.com\/blog\/wp-content\/uploads\/2021\/02\/CG_portrait-2-96x96.jpg","contentUrl":"https:\/\/www.altoros.com\/blog\/wp-content\/uploads\/2021\/02\/CG_portrait-2-96x96.jpg","caption":"Carlo Gutierrez"},"description":"Carlo Gutierrez is a Technical Writer at Altoros. As part of the editorial team, his focus has been on emerging technologies such as Cloud Foundry, Kubernetes, blockchain, and the Internet of Things. Prior to Altoros, he primarily wrote about enterprise and consumer technology. Carlo has over 12 years of experience in the publishing industry. Previously, he served as an Editor for PC World Philippines and Questex Asia, as well as a Designer for Tropa Entertainment.","url":"https:\/\/www.altoros.com\/blog\/author\/carlo\/"}]}},"_links":{"self":[{"href":"https:\/\/www.altoros.com\/blog\/wp-json\/wp\/v2\/posts\/66012","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.altoros.com\/blog\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/www.altoros.com\/blog\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/www.altoros.com\/blog\/wp-json\/wp\/v2\/users\/32"}],"replies":[{"embeddable":true,"href":"https:\/\/www.altoros.com\/blog\/wp-json\/wp\/v2\/comments?post=66012"}],"version-history":[{"count":115,"href":"https:\/\/www.altoros.com\/blog\/wp-json\/wp\/v2\/posts\/66012\/revisions"}],"predecessor-version":[{"id":66696,"href":"https:\/\/www.altoros.com\/blog\/wp-json\/wp\/v2\/posts\/66012\/revisions\/66696"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/www.altoros.com\/blog\/wp-json\/wp\/v2\/media\/66153"}],"wp:attachment":[{"href":"https:\/\/www.altoros.com\/blog\/wp-json\/wp\/v2\/media?parent=66012"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.altoros.com\/blog\/wp-json\/wp\/v2\/categories?post=66012"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.altoros.com\/blog\/wp-json\/wp\/v2\/tags?post=66012"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}