[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-119321-en":3,"doc-seo-119321-105":30,"detail-sidebar-cat-0-en-105":92},{"code":4,"msg":5,"data":6},0,"success",{"doc_id":7,"user_id":8,"nickname":9,"user_avatar":10,"doc_module":4,"category_id":11,"category_name":12,"doc_title":13,"doc_description":14,"doc_content":15,"file_id":16,"file_url":17,"file_type":18,"file_size":19,"view_count":20,"is_deleted":4,"is_public":20,"is_downloadable":20,"audit_status":20,"page_count":21,"language":22,"language_code":23,"site_id":24,"html_lang":23,"table_of_contents":25,"faqs":26,"seo_title":27,"seo_description":14,"update_tm":28,"read_time":29},119321,1374391975076,"Riley","https://ap-avatar.wpscdn.com/avatar/14000253ca4ec9f6853?x-image-process=image/resize,m_fixed,w_180,h_180&k=1783305029341752051",6,"Technology","Machine Learning Operations(MLOps) - from global landscape to practice in Al4EOSC","Machine Learning Operations (MLOps) practices connect the reliable deployment and maintenance of machine learning models to a structured engineering culture. The material reviews MLOps definitions from multiple references, then explains automation levels for model training, validation/testing, continuous training and delivery, and continuous monitoring with triggering. It further surveys open-source platforms using capability criteria across orchestration, distributed training, code management, model development, testing/validation, deployment/inference, data and experiment versioning, metadata storage, and performance monitoring, illustrating how real-world adoption depends on project needs and existing services.","Machine Learning Operations(MLOps):from global landscape to practice in Al4EOSC  \nEGI Conference 2024.October 3d,2024.Lecce,Italy  \nValentin Kozlov¹)*,Lisana Berberi¹,Borja Esteban Sanchis¹,Giang Nguyen²),Judith Sainz-Pardo Diaz³,Amanda Calatrava⁴,Germán Molto⁴,Viet Tran²,Alvaro Lopez Garcia³)  \n*valentin.kozlov@kit.edu  \n1)KIT 2)IISAS 3)IFCA-CSIC 4)UPV  \n# ● Machine Learning Operations(MLOps)definition(s)\n\n● MLOps landscape of platforms &tools  \n● Al4EOSC MLOps practices  \nWikipedia:MLOps is a paradigm that aims to deploy and maintain machine learning models inproduction reliably and efficiently.  \nGoogle:MLOps is an MLengineering culture and practice that aims at unifying ML systemdevelopment(Dev)and ML system operation(Ops).  \nDatabricks*):MLOps is the set of processes and automation for managing data,code and modelsto improve performance stability and long-term efficiency in ML systems  \n*)The Big Book of MLOps:Second Edition(Databricks)  \nMLOps:  \nculture,practices,processes of automation for managing data,development,models,operations  \n# MLOps automation levels (Google)\n\nCeoSC  \nModel Training,Validation/TestingLevel 1Level 2Dev dataExperimentSource codeDev dataExperimentSource codeML PipelineManually  \nCI/CD  \nPipelineOrchestratorContinuousNew dataML PipelineModel registryContinuousNew dataML PipelineModel registryMonitoring &TriggeringMonitoring &TriggeringFunded by  \nmanualthe European Unionautomated  \nMLOps automation levels (Google)  \nManuallyLevel 0Offline dataExperimentModel registryData validation,preparation,Model Training,Validation/Testing  Level 1Level 2Devd● Automated ML pipeline withdeDevLevel 1odeContinuous Training(CT)  \n● Model Continuous Delivery(CD)ator(not the pipeline!)  \nNew(● Continuous Monitoring(CM)listryNew(CI/CD)istry&Triggering!Monitoring&TriggeringPrecIcTIonFunded bymanualthe European Union  \nautomated6  \n# MLOps platforms &tools (subset)\n\nOnly open source is considered  \n·Orchestration (O)  \n● Distributed Training(DT)  \n● Code Management(CM)  \n● Model Development(MDV)  \n● Model Testing/Validation (MTV)  \n*)from L.Berberi“Machine Learning Operations Landscape:Platforms and Tools”,submitted  \n# MLOps Platforms\n\nCriteria to assess MLOps Platforms)  \n● Model Inference(MI)  \n● Model Deployment(MDP)  \n·Experiment Tracking and Metadata Store(ETMS)  \n● Data Versioning and Management (DVM)  \n● Model Performance Monitoring(MPM)  \nReal-world implementation depends on your needs and other already implemented services  \n| Product   | DT  \u003Cbr>Distri-  \u003Cbr>buted  \u003Cbr>Train-  \u003Cbr>ing   | CM  \u003Cbr>Code  \u003Cbr>Man-  \u003Cbr>age-  \u003Cbr>ment   | MDV  \u003Cbr>Model  \u003Cbr>Devel-  \u003Cbr>opment   | MTV  \u003Cbr>Model  \u003Cbr>Test-  \u003Cbr>ing/-  \u003Cbr>Valida-  \u003Cbr>tion   | MDP  \u003Cbr>MI  \u003Cbr>Model Model  \u003Cbr>Deploy-  \u003Cbr>Infer-   l  \u003Cbr>ment  \u003Cbr>ence   | DVM  \u003Cbr>ETMS  \u003Cbr>Experiment Data Ver-  \u003Cbr>Tracking  \u003Cbr>sioning  \u003Cbr>and Meta-  and Man-  \u003Cbr>data Store agement   | MPM  \u003Cbr>Model Per-  \u003Cbr>Full Partial  \u003Cbr>formance  \u003Cbr>Score  \u003Cbr>Score  \u003Cbr>Monitoring   |  | O  \u003Cbr>GitHub Orche-  \u003Cbr>Stars stration   |\n| --- | --- | --- | --- | --- | --- | --- | --- | --- | --- |\n| √√  \u003Cbr>MLflow  \u003Cbr>17.1 K  \u003Cbr>√√    √√  \u003Cbr>√√  \u003Cbr>14.4 K  \u003Cbr>Prefect  \u003Cbr>√√  \u003Cbr>√√  \u003Cbr>13.6 K  \u003Cbr>Kubeflow  \u003Cbr>Dagster  \u003Cbr>10 K  \u003Cbr>√√  \u003Cbr>√√  \u003Cbr>√  \u003Cbr>√√  \u003Cbr>√  \u003Cbr>8.1 K  \u003Cbr>W&B(WB)  \u003Cbr>MetaFlow  \u003Cbr>√√  \u003Cbr>7.5 K  \u003Cbr>√  \u003Cbr>Mage  \u003Cbr>6.9 K  \u003Cbr>√√  \u003Cbr>Pachyderm  \u003Cbr>√  \u003Cbr>√  \u003Cbr>√√  \u003Cbr>√√  \u003Cbr>6.1 K  \u003Cbr>√√  \u003Cbr>√√  \u003Cbr>√√  \u003Cbr>√√  \u003Cbr>√√  \u003Cbr>√√  \u003Cbr>5.2 K  \u003Cbr>ClearML  \u003Cbr>√√  \u003Cbr>√√  \u003Cbr>√√  \u003Cbr>√√  \u003Cbr>√√  \u003Cbr>√√  \u003Cbr>4.7 K  \u003Cbr>√√  \u003Cbr>Flyte  \u003Cbr>√√  \u003Cbr>√√  \u003Cbr>√√  \u003Cbr>√√  \u003Cbr>4.2 K  \u003Cbr>Seldon core  \u003Cbr>√√  \u003Cbr>√√  \u003Cbr>√√  \u003Cbr>√√  \u003Cbr>√√  \u003Cbr>3.6 K  \u003Cbr>√√  \u003Cbr>ZenML  \u003Cbr>√√  \u003Cbr>√√  \u003Cbr>√√  \u003Cbr>3.5 K  \u003Cbr>√√  \u003Cbr>√√  \u003Cbr>Polyaxon  \u003Cbr>√√  \u003Cbr>√  \u003Cbr>√√  \u003Cbr>√√  \u003Cbr>√√  \u003Cbr>√√  \u003Cbr>TFX  \u003Cbr>√√  \u003Cbr>2.1 K  \u003Cbr>√√  \u003Cbr>√√  \u003Cbr>1.5 K  \u003Cbr>MLeap  \u003Cbr>√√  \u003Cbr>√√  \u003Cbr>1.2 K  \u003Cbr>MLRun  \u003Cbr>√√  \u003Cbr>√√  \u003Cbr>√√  \u003Cbr>√√   |  |  |  |  |  |  |  |  |  |","cbCaiiNssCcB04nq","https://ap.wps.com/l/cbCaiiNssCcB04nq","pdf",3846866,1,22,"English","en",105,"# Machine Learning Operations (MLOps) definition(s)\n## MLOps landscape of platforms & tools\n# MLOps automation levels\n## CeoSC: Level 0/1/2 and CI/CD\n# MLOps platforms & tools (subset)\n## Criteria to assess MLOps platforms\n## Comparison table of open-source tools","[{\"question\":\"What is MLOps and what problem does it address?\",\"answer\":\"MLOps aims to deploy and maintain machine learning models reliably and efficiently. It unifies machine learning system development with system operations and uses processes and automation to manage data, code, and models for stability and long-term efficiency.\"},{\"question\":\"How do the MLOps automation levels differ?\",\"answer\":\"The slides present a progression from manual workflows (offline experimentation and basic validation/training) to automated ML pipelines. Higher levels introduce continuous training/delivery and continuous monitoring with monitoring and triggering.\"},{\"question\":\"Which capabilities are used to compare MLOps platforms?\",\"answer\":\"Platforms are assessed across distributed training, code management, model development, testing/validation, deployment/inference, experiment tracking and metadata storage, data versioning and management, and model performance monitoring.\"}]","Machine Learning Operations(MLOps) - from global landscape to practice in Al4EOSC | PDF",1785723705,55,{"code":4,"msg":31,"data":32},"ok",{"site_id":24,"language":23,"slug":33,"title":13,"keywords":34,"description":14,"schema_data":35,"social_meta":87,"head_meta":89,"extra_data":91,"updated_unix":28},"machine-learning-operationsmlops-from-global-landscape-to-practice-in-al4eosc","",{"@graph":36,"@context":86},[37,54,69],{"@type":38,"itemListElement":39},"BreadcrumbList",[40,44,48,51],{"item":41,"name":42,"@type":43,"position":20},"https://docshare.wps.com","Home","ListItem",{"item":45,"name":46,"@type":43,"position":47},"https://docshare.wps.com/document/","Document",2,{"item":49,"name":12,"@type":43,"position":50},"https://docshare.wps.com/document/technology/",3,{"item":52,"name":13,"@type":43,"position":53},"https://docshare.wps.com/document/machine-learning-operationsmlops-from-global-landscape-to-practice-in-al4eosc/119321/",4,{"url":52,"name":13,"@type":55,"author":56,"headline":13,"publisher":58,"fileFormat":61,"inLanguage":23,"description":14,"dateModified":62,"datePublished":63,"encodingFormat":61,"isAccessibleForFree":64,"interactionStatistic":65},"DigitalDocument",{"name":9,"@type":57},"Person",{"url":41,"name":59,"@type":60},"DocShare","Organization","application/pdf","2026-08-04","2026-08-03",true,{"@type":66,"interactionType":67,"userInteractionCount":20},"InteractionCounter",{"@type":68},"ViewAction",{"@type":70,"mainEntity":71},"FAQPage",[72,78,82],{"name":73,"@type":74,"acceptedAnswer":75},"What is MLOps and what problem does it address?","Question",{"text":76,"@type":77},"MLOps aims to deploy and maintain machine learning models reliably and efficiently. It unifies machine learning system development with system operations and uses processes and automation to manage data, code, and models for stability and long-term efficiency.","Answer",{"name":79,"@type":74,"acceptedAnswer":80},"How do the MLOps automation levels differ?",{"text":81,"@type":77},"The slides present a progression from manual workflows (offline experimentation and basic validation/training) to automated ML pipelines. Higher levels introduce continuous training/delivery and continuous monitoring with monitoring and triggering.",{"name":83,"@type":74,"acceptedAnswer":84},"Which capabilities are used to compare MLOps platforms?",{"text":85,"@type":77},"Platforms are assessed across distributed training, code management, model development, testing/validation, deployment/inference, experiment tracking and metadata storage, data versioning and management, and model performance monitoring.","https://schema.org",{"og:url":52,"og:type":88,"og:title":13,"og:site_name":59,"og:description":14},"article",{"robots":90,"canonical":52},"index,follow",{"doc_id":7,"site_id":24},{"code":4,"msg":5,"data":93},[94,98,102,106,111,114,119,124,129,132,136],{"id":20,"doc_module":4,"doc_module_name":46,"category_name":95,"show_sort_weight":96,"slug":97},"Story & Novel",90,"story-novel",{"id":47,"doc_module":4,"doc_module_name":46,"category_name":99,"show_sort_weight":100,"slug":101},"Literature",80,"literature",{"id":53,"doc_module":4,"doc_module_name":46,"category_name":103,"show_sort_weight":104,"slug":105},"Exam",70,"exam",{"id":107,"doc_module":4,"doc_module_name":46,"category_name":108,"show_sort_weight":109,"slug":110},5,"Comic",60,"comic",{"id":11,"doc_module":4,"doc_module_name":46,"category_name":12,"show_sort_weight":112,"slug":113},50,"technology",{"id":115,"doc_module":4,"doc_module_name":46,"category_name":116,"show_sort_weight":117,"slug":118},7,"Healthcare",40,"healthcare",{"id":120,"doc_module":4,"doc_module_name":46,"category_name":121,"show_sort_weight":122,"slug":123},8,"Research & Report",30,"research-report",{"id":125,"doc_module":4,"doc_module_name":46,"category_name":126,"show_sort_weight":127,"slug":128},9,"Religion & Spirituality",20,"religion-spirituality",{"id":127,"doc_module":4,"doc_module_name":46,"category_name":130,"show_sort_weight":127,"slug":131},"World Cup","world-cup",{"id":133,"doc_module":4,"doc_module_name":46,"category_name":134,"show_sort_weight":133,"slug":135},10,"Lifestyle","lifestyle",{"id":137,"doc_module":4,"doc_module_name":46,"category_name":138,"show_sort_weight":107,"slug":139},19,"General","general"]