Supported OS Linux Windows Mac OS

概要

CoreWeave インテグレーションにより、Datadog は CoreWeave の Prometheus メトリクスをスクレイピングし、Prometheus が提供するタグ (ポッド、コンテナ、ネームスペースなど) を含むタグを完全に補完してインポートすることができます。

Datadog を通じて利用パターンを追跡し、組織による CoreWeave クラウドプラットフォームの利用状況の理解を深め、最適化します。

また、このインテグレーションは、組織がどのように請求されているかを視覚化し、CoreWeave Cloud 内のどこから費用が発生しているかを特定します。請求の異常を検出し、異常が発生した場合にアラートを受け取ることで、チームは変化に迅速に対処し、どのポッドやネームスペースに最も費用がかかっているかを判断できます。

セットアップ

ステップ 1: CoreWeave のアクセストークンを取得する

まずは、CoreWeave アクセストークンを取得します。トークンの生成、閲覧、削除ができるのは組織管理者のみです。

以下の手順に従って、CoreWeave アカウントにインテグレーションを追加し、ベアラートークンを作成します。

  1. CoreWeave のアクセストークンページにアクセスし、Create a New Token をクリックします。
  2. トークン名を入力します。できれば Datadog に固有の名前を入力します。
  3. トークンを Available Namespaces から Selected Namespaces に移動して、トークンにネームスペースを割り当てます。Datadog は、CoreWeave インテグレーションを使用する場合、1 つのネームスペースに 1 つのアクセストークンを推奨します。
  4. Generate をクリックします。

ステップ 2 でこのアクセストークンが必要になります。

ステップ 2: CoreWeave アカウントを Datadog に接続する

まず、ステップ 1 のアクセストークンキーをコピーします。

  1. CoreWeave インテグレーションタイルに移動します。
  2. アカウントの Name を入力します。
  3. CoreWeave アカウントのアクセストークンキーをアクセストークンフィールドに貼り付けます。
  4. オプションで、これらのログにタグを定義することができます。
  5. その後、Save をクリックします。

検証

  1. coreweave. というプレフィックスを持つメトリクスを確認します。このメトリクスが存在すれば、インテグレーションは機能しています。

収集データ

メトリクス

coreweave.ALERTS
(count)
coreweave.ALERTS_FOR_STATE
(count)
coreweave.billing_cpu
(count)
coreweave.billing_gpu
(count)
coreweave.billing_ip_address
(count)
coreweave.billing_mem
(count)
coreweave.billing_object_storage_usage_bytes
(count)
coreweave.billing_object_storage_usage_total_bytes
(count)
coreweave.billing_pod_cost_hourly
(count)
coreweave.billing_pod_cost_total
(count)
coreweave.billing_pod_cpu
(count)
coreweave.billing_pod_cpu_cost_hourly
(count)
coreweave.billing_pod_gpu
(count)
coreweave.billing_pod_gpu_cost_hourly
(count)
coreweave.billing_pod_mem
(count)
coreweave.billing_pod_mem_cost_hourly
(count)
coreweave.billing_resource_usage_gpu
(count)
coreweave.billing_resource_usage_storage
(count)
coreweave.cluster_namespace_pod_cpu_activekube_pod_container_resource_requests
(count)
coreweave.cluster_namespace_pod_memory_activekube_pod_container_resource_requests
(count)
coreweave.cluster_namespace_deployment_actual_replicas_count
(count)
coreweave.cluster_namespace_deployment_container_cpu_usage_seconds_total_sum_rate
(count)
coreweave.cluster_namespace_deployment_container_memory_usage_bytes_sum
(count)
coreweave.cluster_namespace_deploymentkube_pod_container_resource_requests_cpu_cores_sum
(count)
coreweave.cluster_namespace_deploymentkube_pod_container_resource_requests_memory_bytes_sum
(count)
coreweave.container_cpu_usage_seconds_total
(count)
Cumulative cpu time consumed in seconds.
coreweave.container_fs_reads_bytes_total
(count)
Cumulative count of bytes read
coreweave.container_fs_writes_bytes_total
(count)
Cumulative count of bytes written
coreweave.container_memory_cache
(gauge)
Number of bytes of page cache memory.
coreweave.container_memory_rss
(gauge)
Size of RSS in bytes.
coreweave.container_memory_usage_bytes
(gauge)
Current memory usage in bytes, including all memory regardless of when it was accessed
coreweave.container_memory_working_set_bytes
(gauge)
Current working set in bytes.
coreweave.container_network_receive_bytes_total
(count)
Cumulative count of bytes received
coreweave.container_network_receive_packets_total
(count)
Cumulative count of packets received
coreweave.container_network_transmit_bytes_total
(count)
Cumulative count of bytes transmitted
coreweave.container_network_transmit_packets_total
(count)
Cumulative count of packets transmitted
coreweave.coreweave_billed_amount
(gauge)
CoreWeave billed amount based on the billing label
coreweave.coreweave_billed_usage
(gauge)
CoreWeave billed usage based on the billing label
coreweave.coreweave_customer_reference
(gauge)
CoreWeave customer reference showing the customer ID
coreweave.coreweave_pricing
(gauge)
CoreWeave product pricing for specific tenants
coreweave.coreweave_subscription
(gauge)
CoreWeave subscription ID
coreweave.coreweave_subscription_active
(gauge)
CoreWeave subscription active status, will return 1 if active
coreweave.coreweave_subscription_amount_billed
(gauge)
Total Billed usage in US cents since the beginning of the customer subscription
coreweave.coreweave_subscription_current_balance
(gauge)
Unpaid balance in USd (US cents) for already-invoiced charges
coreweave.coreweave_subscription_current_billing_amount
(gauge)
Accumulated charges in USd (US cents) for the current period (not yet invoiced)
coreweave.count_up0
(count)
coreweave.count_up1
(count)
coreweave.instance_node_cpu_ratio
(count)
coreweave.kube_configmap_annotations
(gauge)
Kubernetes annotations converted to Prometheus labels.
coreweave.kube_configmap_created
(gauge)
Unix creation timestamp
coreweave.kube_configmap_info
(gauge)
[STABLE] Information about configmap.
coreweave.kube_configmap_labels
(gauge)
Kubernetes labels converted to Prometheus labels.
coreweave.kube_configmap_metadata_resource_version
(gauge)
Resource version representing a specific version of the configmap.
coreweave.kube_deployment_annotations
(gauge)
Kubernetes annotations converted to Prometheus labels.
coreweave.kube_deployment_created
(gauge)
[STABLE] Unix creation timestamp
coreweave.kube_deployment_labels
(gauge)
Kubernetes labels converted to Prometheus labels.
coreweave.kube_deployment_metadata_generation
(gauge)
Sequence number representing a specific generation of the desired state.
coreweave.kube_deployment_spec_paused
(gauge)
Whether the deployment is paused and will not be processed by the deployment controller.
coreweave.kube_deployment_spec_replicas
(gauge)
Number of desired pods for a deployment.
coreweave.kube_deployment_spec_strategy_rollingupdate_max_surge
(gauge)
Maximum number of replicas that can be scheduled above the desired number of replicas during a rolling update of a deployment.
coreweave.kube_deployment_spec_strategy_rollingupdate_max_unavailable
(gauge)
Maximum number of unavailable replicas during a rolling update of a deployment.
coreweave.kube_deployment_status_condition
(gauge)
The current status conditions of a deployment.
coreweave.kube_deployment_status_observed_generation
(gauge)
The generation observed by the deployment controller.
coreweave.kube_deployment_status_replicas
(gauge)
The number of replicas per deployment.
coreweave.kube_deployment_status_replicas_available
(gauge)
The number of available replicas per deployment.
coreweave.kube_deployment_status_replicas_ready
(gauge)
The number of ready replicas per deployment.
coreweave.kube_deployment_status_replicas_unavailable
(gauge)
The number of unavailable replicas per deployment.
coreweave.kube_deployment_status_replicas_updated
(gauge)
The number of updated replicas per deployment.
coreweave.kube_endpoint_address
(gauge)
[STABLE] Information about Endpoint available and non available addresses.
coreweave.kube_endpoint_address_available
(gauge)
Number of addresses available in endpoint.
coreweave.kube_endpoint_address_not_ready
(gauge)
Number of addresses not ready in endpoint
coreweave.kube_endpoint_annotations
(gauge)
Kubernetes annotations converted to Prometheus labels.
coreweave.kube_endpoint_created
(gauge)
Unix creation timestamp
coreweave.kube_endpoint_info
(gauge)
Information about endpoint.
coreweave.kube_endpoint_labels
(gauge)
[STABLE] Kubernetes labels converted to Prometheus labels.
coreweave.kube_endpoint_ports
(gauge)
Information about the Endpoint ports.
coreweave.kube_ingress_annotations
(gauge)
Kubernetes annotations converted to Prometheus labels.
coreweave.kube_ingress_created
(gauge)
Unix creation timestamp
coreweave.kube_ingress_info
(gauge)
[STABLE] Information about ingress.
coreweave.kube_ingress_labels
(gauge)
Kubernetes labels converted to Prometheus labels.
coreweave.kube_ingress_metadata_resource_version
(gauge)
Resource version representing a specific version of ingress.
coreweave.kube_ingress_path
(gauge)
Ingress host, paths and backend service information.
coreweave.kube_ingress_tls
(gauge)
Ingress TLS host and secret information.
coreweave.kube_namespace_annotations
(gauge)
Kubernetes annotations converted to Prometheus labels.
coreweave.kube_namespace_created
(gauge)
Unix creation timestamp
coreweave.kube_namespace_labels
(gauge)
Kubernetes labels converted to Prometheus labels.
coreweave.kube_namespace_status_phase
(gauge)
[STABLE] kubernetes namespace status phase.
coreweave.kube_networkpolicy_annotations
(gauge)
Kubernetes annotations converted to Prometheus labels.
coreweave.kube_networkpolicy_created
(gauge)
Unix creation timestamp of network policy
coreweave.kube_networkpolicy_labels
(gauge)
Kubernetes labels converted to Prometheus labels.
coreweave.kube_networkpolicy_spec_egress_rules
(gauge)
Number of egress rules on the networkpolicy
coreweave.kube_networkpolicy_spec_ingress_rules
(gauge)
Number of ingress rules on the networkpolicy
coreweave.kube_persistentvolumeclaim_access_mode
(gauge)
The access mode(s) specified by the persistent volume claim.
coreweave.kube_persistentvolumeclaim_annotations
(gauge)
Kubernetes annotations converted to Prometheus labels.
coreweave.kube_persistentvolumeclaim_created
(gauge)
Unix creation timestamp
coreweave.kube_persistentvolumeclaim_info
(gauge)
Information about persistent volume claim.
coreweave.kube_persistentvolumeclaim_labels
(gauge)
Kubernetes labels converted to Prometheus labels.
coreweave.kube_persistentvolumeclaim_resource_requests_storage_bytes
(gauge)
[STABLE] The capacity of storage requested by the persistent volume claim.
coreweave.kube_persistentvolumeclaim_status_phase
(gauge)
The phase the persistent volume claim is currently in.
coreweave.kube_pod_annotations
(gauge)
Kubernetes annotations converted to Prometheus labels.
coreweave.kube_pod_container_info
(gauge)
Information about a container in a pod.
coreweave.kube_pod_container_resource_limits
(gauge)
The number of requested limit resource by a container.
coreweave.kube_pod_container_resource_requests
(gauge)
The number of requested request resource by a container.
coreweave.kube_pod_container_state_started
(gauge)
Start time in unix timestamp for a pod container.
coreweave.kube_pod_container_status_ready
(gauge)
Describes whether the containers readiness check succeeded.
coreweave.kube_pod_container_status_restarts_total
(count)
The number of container restarts per container.
coreweave.kube_pod_container_status_running
(gauge)
[STABLE] Describes whether the container is currently in running state.
coreweave.kube_pod_container_status_terminated
(gauge)
Describes whether the container is currently in terminated state.
coreweave.kube_pod_container_status_waiting
(gauge)
Describes whether the container is currently in waiting state.
coreweave.kube_pod_created
(gauge)
Unix creation timestamp
coreweave.kube_pod_deletion_timestamp
(gauge)
Unix deletion timestamp
coreweave.kube_pod_info
(gauge)
Information about pod.
coreweave.kube_pod_init_container_info
(gauge)
Information about an init container in a pod.
coreweave.kube_pod_init_container_resource_requests
(gauge)
The number of requested request resource by an init container.
coreweave.kube_pod_init_container_status_ready
(gauge)
Describes whether the init containers readiness check succeeded.
coreweave.kube_pod_init_container_status_restarts_total
(count)
The number of restarts for the init container.
coreweave.kube_pod_init_container_status_running
(gauge)
Describes whether the init container is currently in running state.
coreweave.kube_pod_init_container_status_terminated
(gauge)
Describes whether the init container is currently in terminated state.
coreweave.kube_pod_init_container_status_terminated_reason
(gauge)
Describes the reason the init container is currently in terminated state.
coreweave.kube_pod_init_container_status_waiting
(gauge)
Describes whether the init container is currently in waiting state.
coreweave.kube_pod_ips
(gauge)
Pod IP addresses
coreweave.kube_pod_labels
(gauge)
Kubernetes labels converted to Prometheus labels.
coreweave.kube_pod_owner
(gauge)
Information about the Pod's owner.
coreweave.kube_pod_restart_policy
(gauge)
Describes the restart policy in use by this pod.
coreweave.kube_pod_running_info
(count)
coreweave.kube_pod_spec_volumes_persistentvolumeclaims_info
(gauge)
Information about persistentvolumeclaim volumes in a pod.
coreweave.kube_pod_spec_volumes_persistentvolumeclaims_readonly
(gauge)
Describes whether a persistentvolumeclaim is mounted read only.
coreweave.kube_pod_start_time
(gauge)
[STABLE] Start time in unix timestamp for a pod.
coreweave.kube_pod_status_container_ready_time
(gauge)
Readiness achieved time in unix timestamp for a pod containers.
coreweave.kube_pod_status_phase
(gauge)
The pods current phase.
coreweave.kube_pod_status_qos_class
(gauge)
The pods current qosClass.
coreweave.kube_pod_status_ready
(gauge)
Describes whether the pod is ready to serve requests.
coreweave.kube_pod_status_ready_time
(gauge)
Readiness achieved time in unix timestamp for a pod.
coreweave.kube_pod_status_reason
(gauge)
The pod status reasons
coreweave.kube_pod_status_scheduled
(gauge)
[STABLE] Describes the status of the scheduling process for the pod.
coreweave.kube_pod_status_scheduled_time
(gauge)
Unix timestamp when pod moved into scheduled status
coreweave.kube_pod_tolerations
(gauge)
Information about the pod tolerations
coreweave.kube_replicaset_annotations
(gauge)
Kubernetes annotations converted to Prometheus labels.
coreweave.kube_replicaset_created
(gauge)
[STABLE] Unix creation timestamp
coreweave.kube_replicaset_labels
(gauge)
Kubernetes labels converted to Prometheus labels.
coreweave.kube_replicaset_metadata_generation
(gauge)
Sequence number representing a specific generation of the desired state.
coreweave.kube_replicaset_owner
(gauge)
Information about the ReplicaSet's owner.
coreweave.kube_replicaset_spec_replicas
(gauge)
Number of desired pods for a ReplicaSet.
coreweave.kube_replicaset_status_fully_labeled_replicas
(gauge)
The number of fully labeled replicas per ReplicaSet.
coreweave.kube_replicaset_status_observed_generation
(gauge)
The generation observed by the ReplicaSet controller.
coreweave.kube_replicaset_status_ready_replicas
(gauge)
[STABLE] The number of ready replicas per ReplicaSet.
coreweave.kube_replicaset_status_replicas
(gauge)
The number of replicas per ReplicaSet.
coreweave.kube_resourcequota
(gauge)
Information about resource quota.
coreweave.kube_resourcequota_created
(gauge)
Unix creation timestamp
coreweave.kube_secret_annotations
(gauge)
Kubernetes annotations converted to Prometheus labels.
coreweave.kube_secret_created
(gauge)
Unix creation timestamp
coreweave.kube_secret_info
(gauge)
Information about secret.
coreweave.kube_secret_labels
(gauge)
Kubernetes labels converted to Prometheus labels.
coreweave.kube_secret_metadata_resource_version
(gauge)
Resource version representing a specific version of secret.
coreweave.kube_secret_type
(gauge)
Type about secret.
coreweave.kube_service_annotations
(gauge)
Kubernetes annotations converted to Prometheus labels.
coreweave.kube_service_created
(gauge)
Unix creation timestamp
coreweave.kube_service_info
(gauge)
Information about service.
coreweave.kube_service_labels
(gauge)
Kubernetes labels converted to Prometheus labels.
coreweave.kube_service_spec_type
(gauge)
Type about service.
coreweave.kube_service_status_load_balancer_ingress
(gauge)
Service load balancer ingress status
coreweave.kubelet_container_log_filesystem_used_bytes
(gauge)
[ALPHA] Bytes used by the container's logs on the filesystem.
coreweave.kubevirt_vnc_active_connections
(gauge)
Amount of active VNC connections, broken down by namespace and vmi name
coreweave.namespace_cpukube_pod_container_resource_requests_sum
(count)
coreweave.namespace_memorykube_pod_container_resource_requests_sum
(count)
coreweave.node_boot_time_seconds
(gauge)
Node boot time, in unixtime.
coreweave.node_context_switches_total
(count)
Total number of context switches.
coreweave.node_cooling_device_cur_state
(gauge)
Current throttle state of the cooling device
coreweave.node_cpu_seconds_total
(count)
Seconds the CPUs spent in each mode.
coreweave.node_disk_io_time_seconds_total
(count)
Total seconds spent doing I/Os.
coreweave.node_disk_read_bytes_total
(count)
The total number of bytes read successfully.
coreweave.node_disk_write_time_seconds_total
(count)
This is the total number of seconds spent by all writes.
coreweave.node_disk_written_bytes_total
(count)
The total number of bytes written successfully.
coreweave.node_filesystem_avail_bytes
(gauge)
Filesystem space available to non-root users in bytes.
coreweave.node_filesystem_size_bytes
(gauge)
Filesystem size in bytes.
coreweave.node_load1
(gauge)
1m load average.
coreweave.node_load15
(gauge)
15m load average.
coreweave.node_load5
(gauge)
5m load average.
coreweave.node_memory_Active_anon_bytes
(gauge)
Memory information field Activeanonbytes.
coreweave.node_memory_Active_bytes
(gauge)
Memory information field Active_bytes.
coreweave.node_memory_Active_file_bytes
(gauge)
Memory information field Activefilebytes.
coreweave.node_memory_AnonHugePages_bytes
(gauge)
Memory information field AnonHugePages_bytes.
coreweave.node_memory_AnonPages_bytes
(gauge)
Memory information field AnonPages_bytes.
coreweave.node_memory_Bounce_bytes
(gauge)
Memory information field Bounce_bytes.
coreweave.node_memory_Buffers_bytes
(gauge)
Memory information field Buffers_bytes.
coreweave.node_memory_Cached_bytes
(gauge)
Memory information field Cached_bytes.
coreweave.node_memory_CommitLimit_bytes
(gauge)
Memory information field CommitLimit_bytes.
coreweave.node_memory_Committed_AS_bytes
(gauge)
Memory information field CommittedASbytes.
coreweave.node_memory_DirectMap1G_bytes
(gauge)
Memory information field DirectMap1G_bytes.
coreweave.node_memory_DirectMap2M_bytes
(gauge)
Memory information field DirectMap2M_bytes.
coreweave.node_memory_DirectMap4k_bytes
(gauge)
Memory information field DirectMap4k_bytes.
coreweave.node_memory_Dirty_bytes
(gauge)
Memory information field Dirty_bytes.
coreweave.node_memory_HardwareCorrupted_bytes
(gauge)
Memory information field HardwareCorrupted_bytes.
coreweave.node_memory_HugePages_Free
(gauge)
Memory information field HugePages_Free.
coreweave.node_memory_HugePages_Total
(gauge)
Memory information field HugePages_Total.
coreweave.node_memory_Hugepagesize_bytes
(gauge)
Memory information field Hugepagesize_bytes.
coreweave.node_memory_Inactive_anon_bytes
(gauge)
Memory information field Inactiveanonbytes.
coreweave.node_memory_Inactive_bytes
(gauge)
Memory information field Inactive_bytes.
coreweave.node_memory_Mapped_bytes
(gauge)
Memory information field Mapped_bytes.
coreweave.node_memory_MemAvailable_bytes
(gauge)
Memory information field MemAvailable_bytes.
coreweave.node_memory_MemFree_bytes
(gauge)
Memory information field MemFree_bytes.
coreweave.node_memory_MemTotal_bytes
(gauge)
Memory information field MemTotal_bytes.
coreweave.node_namespace_podkube_pod_info
(count)
coreweave.node_namespace_pod_container_container_cpu_usage_seconds_total_sum_irate
(count)
coreweave.node_namespace_pod_container_container_memory_cache
(count)
coreweave.node_namespace_pod_container_container_memory_rss
(count)
coreweave.node_namespace_pod_container_container_memory_working_set_bytes
(count)
coreweave.node_network_receive_bytes_total
(count)
Network device statistic receive_bytes.
coreweave.node_network_receive_drop_total
(count)
Network device statistic receive_drop.
coreweave.node_network_receive_errs_total
(count)
Network device statistic receive_errs.
coreweave.node_network_receive_packets_total
(count)
Network device statistic receive_packets.
coreweave.node_network_transmit_bytes_total
(count)
Network device statistic transmit_bytes.
coreweave.node_network_transmit_drop_total
(count)
Network device statistic transmit_drop.
coreweave.node_network_transmit_errs_total
(count)
Network device statistic transmit_errs.
coreweave.node_network_transmit_packets_total
(count)
Network device statistic transmit_packets.
coreweave.node_uname_info
(gauge)
Labeled system information as provided by the uname system call.
coreweave.node_uptime
(count)
coreweave.object_storage_quota_bytes
(count)
coreweave.object_storage_total_object_count
(count)
coreweave.object_storage_total_quota_bytes
(count)
coreweave.scrape_duration_seconds
(count)
coreweave.scrape_samples_post_metric_relabeling
(count)
coreweave.scrape_samples_scraped
(count)
coreweave.scrape_series_added
(count)
coreweave.up
(count)

サービスチェック

coreweave には、サービスのチェック機能は含まれません。

イベント

coreweave には、イベントは含まれません。

トラブルシューティング

ご不明な点は、Datadog のサポートチームまでお問合せください。

PREVIEWING: maxime/fix-typo-secret-doc