유휴 VM 권장사항을 구성하는 방법에 대한 자세한 내용은 유휴 VM 권장사항 구성을 참조하세요.
가격 책정
유휴 VM 권장사항은 무료로 제공됩니다. 리소스 사용량 감소를 위한 권장사항을 사용하면 비용을 절약할 수 있습니다.
제한사항
다음과 같은 경우 독립형 VM에 유휴 VM 권장사항을 사용할 수 없습니다.
로컬 SSD를 사용하는 인스턴스
GPU/TPU를 사용하는 인스턴스
App Engine Flex 리소스
Dataflow 리소스
Google Kubernetes Engine 리소스
유휴 VM 인스턴스 감지 작동 방식
Compute Engine은 이전 사용량 측정항목을 기반으로 유휴 VM 인스턴스에 대해 권장사항을 생성합니다. 기본적으로 이전 관찰 기간은 14일이고, 신규 VM의 경우에는 VM 생성 다음 날부터 시작됩니다.
기본 관찰 기간을 변경하여 수신되는 권장사항을 맞춤설정할 수 있습니다.
권장사항을 생성하기 위해 알고리즘은 최근 관찰 기간에서 CPU 및 네트워크 사용량을 고려합니다.
CPU 및 네트워크 사용량이 미리 정의된 임곗값보다 낮으면 추천자가 VM을 유휴 상태로 분류합니다.
권장사항 실행 빈도
VM이 생성되고 관찰 기간 동안 최소 하루 이상 실행된 다음 Compute Engine이 이에 대한 유휴 VM 권장사항 생성을 시작합니다.
새 권장사항은 하루에 한 번 이상 생성됩니다.
머신 유형 크기 조정 권장사항과의 관계
머신 유형 크기 조정 권장사항은 VM에 가장 적합한 크기를 추천합니다. 유휴 VM의 크기 축소를 위한 머신 유형 권장사항을 수신할 수 있습니다.
사용률이 낮은 유휴 VM을 계속 실행해야 하는 경우 머신 유형 권장사항을 사용하여 해당 VM에 가장 적합한 크기를 선택할 수 있습니다.
권장사항 맞춤설정
Compute Engine을 사용하면 권장사항 알고리즘에 사용되는 구성을 변경하여 프로젝트에 대해 수신되는 권장사항을 맞춤설정할 수 있습니다.
특히 기본 관찰 기간을 변경하여 워크로드, 애플리케이션, 인프라 요구에 적합한 권장사항을 수신할 수 있습니다.
[[["이해하기 쉬움","easyToUnderstand","thumb-up"],["문제가 해결됨","solvedMyProblem","thumb-up"],["기타","otherUp","thumb-up"]],[["이해하기 어려움","hardToUnderstand","thumb-down"],["잘못된 정보 또는 샘플 코드","incorrectInformationOrSampleCode","thumb-down"],["필요한 정보/샘플이 없음","missingTheInformationSamplesINeed","thumb-down"],["번역 문제","translationIssue","thumb-down"],["기타","otherDown","thumb-down"]],["최종 업데이트: 2025-07-26(UTC)"],[[["\u003cp\u003eCompute Engine identifies idle virtual machines (VMs) that have been unused for 1 to 14 days to help reduce resource waste and lower compute costs.\u003c/p\u003e\n"],["\u003cp\u003eIdle VM recommendations are generated automatically based on Cloud Monitoring service metrics, using CPU and network usage data.\u003c/p\u003e\n"],["\u003cp\u003eThe observation period for detecting idle VMs can be adjusted between 1 and 14 days to tailor recommendations to different workload patterns.\u003c/p\u003e\n"],["\u003cp\u003eIdle VM recommendations are provided free of charge, but are not available for VMs with local SSDs, GPUs/TPUs, App Engine flex resources, Dataflow resources, or Google Kubernetes Engine resources.\u003c/p\u003e\n"],["\u003cp\u003eIf you must keep an idle VM running, you can use machine type recommendations to help choose an optimal, potentially smaller, machine type.\u003c/p\u003e\n"]]],[],null,["*** ** * ** ***\n\nCompute Engine provides idle VM recommendations to help you identify\nvirtual machine (VM) instances that have not been used over the previous 1 to\n14 days. You can use idle VM recommendations to find and stop idle VM instances\nto reduce wasting resources and reduce your compute bill.\n\nCompute Engine generates recommendations automatically based on system\nmetrics gathered by the Cloud Monitoring service. You can configure idle VM\nrecommendations to receive more or fewer recommendations.\n\nThis page provides conceptual information on how Compute Engine generates\nidle VM recommendations and on the parameters you can use to configure them.\n\n- For more information on how to identify idle VMs and take action, see [View and apply idle VM recommendations](/compute/docs/instances/viewing-and-applying-idle-vm-recommendations).\n- For more information on how to configure idle VM recommendations, see [Configure idle VM recommendations](/compute/docs/instances/configuring-idle-vm-recommendations).\n\nPricing\n\nIdle VM recommendations are available free of charge. Using recommendations\nto reduce your resource usage can result in cost savings.\n\nLimitations\n\nFor standalone VMs, idle VM recommendations are not available in following cases:\n\n- Instances with local SSDs\n- Instances with GPU/TPUs\n- App Engine flex resources\n- Dataflow resources\n- Google Kubernetes Engine resources\n\nHow detection of idle VM instances works\n\nCompute Engine generates recommendations about idle VM instances based\non historical usage metrics. By default, the historical observation period is\n14 days, or, for new VMs, starting one day after VM creation.\nBy changing the default observation period, you can customize the\nrecommendations that you receive.\n\nTo generate recommendations, the algorithm considers the CPU and network usage\nin the last observation period.\nIf CPU and network usage are below predefined thresholds,\nthe Recommender classifies the VM as idle.\n| **Note:** Google designs the default thresholds so that if monitoring agents generate the majority of CPU and network usage, then your VM is classified as idle. The default threshold values that are currently used to classify a VM as idle might change in the future.\n\nFrequency of recommendations\n\nAfter a VM is created and running for at least one day during the\nobservation period, Compute Engine begins generating idle VM\nrecommendations for it.\nNew recommendations are generated once per day.\n\nRelation with machine type sizing recommendations\n\n[Machine type sizing recommendations](/compute/docs/instances/apply-sizing-recommendations-for-instances)\nrecommend the optimal size for a VM. You might receive a machine type\nrecommendation to downsize an idle VM.\n\nIf you must keep an idle VM running despite its low utilization, you can use\nmachine type recommendations to help you choose the optimal size for\nunderutilized VMs.\n\nCustomize recommendations\n| **Preview**\n|\n|\n| This product or feature is subject to the \"Pre-GA Offerings Terms\" in the General Service Terms section\n| of the [Service Specific Terms](/terms/service-terms#1).\n|\n| Pre-GA products and features are available \"as is\" and might have limited support.\n|\n| For more information, see the\n| [launch stage descriptions](/products#product-launch-stages).\n\nCompute Engine lets you customize the recommendations you receive for\nyour project by changing the configuration used by the recommendation algorithm.\nIn particular, by changing the default observation period, you can receive\nrecommendations that better fit your workloads, applications,\nand infrastructure needs.\n\nTo learn how to modify the configuration for your project,\nsee [Configure idle VM recommendations](/compute/docs/instances/configuring-idle-vm-recommendations).\n\nChoose the right configuration\n\nThis section describes the values that you can set for the configuration.\nChanging these values affects the recommendations that you receive.\n\nThe observation period\n\nSet the observation period to modify the duration of the window used to\ncalculate the recommendations.\n\nYou can set the observation period to a value between 1 day and 14 days, using\na string with the total number of seconds, followed by the letter `s`.\n\n- For an observation period of 1 day, use `\"86400s\"`.\n- For an observation period of 14 days, use `\"1209600s\"`.\n\nBy default, the observation period is 14 days.\n\n- Use a shorter observation period if you want to receive recommendations based on short-term changes in your workload.\n- Use a longer observation period if you want to receive recommendations that are not affected by short-term fluctuations in your workload.\n\nWhat's next\n\n- Learn how to [view and apply idle VM recommendations](/compute/docs/instances/viewing-and-applying-idle-vm-recommendations).\n- Learn how to [configure idle VM recommendations](/compute/docs/instances/configuring-idle-vm-recommendations)."]]
