Cloud Composer는 환경의 클러스터에서 이러한 자동 확장 처리를 구성합니다. 이렇게 하면 클러스터의 노드 수, 머신 유형, 작업자 수가 자동으로 확장됩니다.
확장 및 성능 매개변수
자동 확장 외에도 스케줄러, 웹 서버, 작업자의 CPU, 메모리, 디스크 한도를 조정하여 환경의 확장 및 성능 매개변수를 제어할 수 있습니다. 이렇게 하면 자동 확장 기능에서 제공하는 수평 확장 외에도 환경을 수직으로 확장할 수 있습니다. Airflow 스케줄러, 웹 서버, 작업자의 확장 및 성능 매개변수는 언제든지 조정 가능합니다.
환경의 환경 크기 성능 매개변수는 Airflow 데이터베이스가 포함된 관리형 Cloud Composer 인프라의 성능 매개변수를 제어합니다. 인프라 성능이 우수한 DAG 및 태스크를 대량으로 실행하려는 경우에는 더 큰 환경 크기를 선택하는 것이 좋습니다. 예를 들어 환경 크기가 클수록 환경에서 최소 지연으로 처리할 수 있는 Airflow 태스크 로그 항목 양이 증가합니다.
여러 스케줄러
Airflow 2는 동시에 1개 이상의 Airflow 스케줄러를 사용할 수 있습니다. 이 Airflow 기능은 HA 스케줄러라고도 합니다. Cloud Composer 2에서는 환경의 스케줄러 수를 설정하고 언제든지 조정할 수 있습니다. Cloud Composer는 환경의 스케줄러 수를 자동으로 확장하지 않습니다.
[[["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: 2025-09-03(UTC)"],[[["\u003cp\u003eCloud Composer 2 environments automatically scale the number of workers based on the demands of your DAGs and tasks, increasing workers during heavy loads and removing them during low usage.\u003c/p\u003e\n"],["\u003cp\u003eCloud Composer 2 uses GKE's Horizontal Pod Autoscaler (HPA), Cluster Autoscaler (CA), and Node auto-provisioning (NAP) to manage autoscaling, adjusting the number of nodes, machine types, and workers.\u003c/p\u003e\n"],["\u003cp\u003eYou can manually adjust CPU, memory, and disk limits for schedulers, web servers, and workers, providing vertical scaling in addition to the horizontal autoscaling.\u003c/p\u003e\n"],["\u003cp\u003eThe environment size parameter affects the managed Cloud Composer infrastructure's performance, impacting how many Airflow task log entries your environment can process efficiently.\u003c/p\u003e\n"],["\u003cp\u003eCloud Composer 2 allows for multiple Airflow schedulers, which you can configure, but the number of schedulers does not automatically scale, and the Airflow database's disk space scales automatically to adapt to the demand.\u003c/p\u003e\n"]]],[],null,["**Cloud Composer 3** \\| [Cloud Composer 2](/composer/docs/composer-2/environment-scaling \"View this page for Cloud Composer 2\") \\| [Cloud Composer 1](/composer/docs/composer-1/environment-scaling \"View this page for Cloud Composer 1\")\n\n\u003cbr /\u003e\n\n| **Note:** This page is **not yet revised for Cloud Composer 3** and displays content for Cloud Composer 2.\n\n\u003cbr /\u003e\n\n\u003cbr /\u003e\n\n\u003cbr /\u003e\n\n\u003cbr /\u003e\n\n\u003cbr /\u003e\n\nThis page describes how environment scaling works in Cloud Composer 2.\n\nOther pages about scaling:\n\n- For a guide about selecting optimal scale and performance parameters for your environment, see [Optimize environment performance and costs](/composer/docs/composer-2/optimize-environments).\n- For information about scaling your environments, see [Scale environments](/composer/docs/composer-2/scale-environments).\n\nAutoscaling environments\n\nCloud Composer 2 environments automatically scale in response to the demands\nof your executed DAGs and tasks:\n\n- If your environment experiences a heavy load, Cloud Composer automatically increases the number of workers in your environment.\n- If your environment does not use some of its workers, these workers are removed to save environment resources and costs.\n- You can set the minimum and maximum number of workers for your environment. Cloud Composer automatically scales your environment within the set limits. You can adjust these limits at any time.\n\nThe number of workers is adjusted based on\nthe [Scaling Factor Target](/composer/docs/composer-2/monitor-environments#worker-metrics) metric. This metric is\ncalculated based on:\n\n- Current number of workers\n- Number of Celery tasks in the Celery queue, that are not assigned to a worker\n- Number of idle workers\n- `celery.worker_concurrency` Airflow configuration option\n\nCloud Composer autoscaling uses three different autoscalers\nprovided by GKE:\n\n- [Horizontal Pod Autoscaler (HPA)](/kubernetes-engine/docs/concepts/horizontalpodautoscaler)\n- [Cluster Autoscaler (CA)](/kubernetes-engine/docs/concepts/cluster-autoscaler)\n- [Node auto-provisioning (NAP)](/kubernetes-engine/docs/how-to/node-auto-provisioning)\n\nCloud Composer configures these autoscalers in the environment's\ncluster. This automatically scales the number of nodes in the cluster, the\nmachine type and the number of workers.\n\nScale and performance parameters\n\nIn addition to autoscaling, you can control the scale and performance\nparameters of your environment by adjusting the CPU, memory, and disk limits\nfor schedulers, web server, and workers. By doing so you can scale your\nenvironment vertically, in addition to the horizontal scaling provided by the\nautoscaling feature. You can adjust the scale and performance parameters of\nAirflow schedulers, web server, and workers at any time.\n\nThe *environment size* performance parameter of your environment controls the\nperformance parameters of the managed Cloud Composer infrastructure\nthat includes the Airflow database. Consider selecting a larger environment\nsize if you want to run a large number of DAGs and tasks with higher\ninfrastructure performance. For example, larger environment's size increases\nthe amount of Airflow task log entries that your environment can process with\nminimal delay.\n| **Note:** Environment size is different from the environment presets. Environment presets, which you can select when you create an environment, determine all limits, scale, and performance parameters of your environment, including the environment size. Environment size determines only the performance parameters of the managed Cloud Composer infrastructure of your environment.\n\nMultiple schedulers\n\nAirflow 2 can use more than one Airflow scheduler at the same time. This\nAirflow feature is also known as the **HA scheduler**. In Cloud Composer 2,\nyou can set the number of schedulers for your environment and adjust it at any\ntime. Cloud Composer does not automatically scale the number of\nschedulers in your environment.\n\nFor more information about configuring the number of schedulers for your\nenvironment, see [Scale environments](/composer/docs/composer-2/scale-environments#scheduler-count).\n\nDatabase disk space\n\nDisk space for the Airflow database automatically increases to accommodate the\ndemand.\n\n\nWhat's next\n\n- [Scale environments](/composer/docs/composer-2/scale-environments)\n- [Cloud Composer pricing](/composer/pricing)\n- [Create environments](/composer/docs/composer-2/create-environments)\n- [Environment architecture](/composer/docs/composer-2/environment-architecture)"]]
