Kepler

Supported OS Linux Windows Mac OS

통합 버전1.0.0
이 페이지는 아직 한국어로 제공되지 않으며 번역 작업 중입니다. 번역에 관한 질문이나 의견이 있으시면 언제든지 저희에게 연락해 주십시오.

Overview

This check brings Kuberenetes workloads energy usage estimates from Kepler into Datadog.

Setup

Installation

For Agent v7.21+ / v6.21+, follow the instructions below to install the Kepler check on your host. See Use Community Integrations to install with the Docker Agent or earlier versions of the Agent.

  1. Run the following command to install the integration:

    datadog-agent integration install -t datadog-kepler==<INTEGRATION_VERSION>

  2. Configure your integration similar to core integrations.

Configuration

  1. Edit the kepler.d/conf.yaml file, in the conf.d/ folder at the root of your Agent’s configuration directory to start collecting your kepler performance data. See the sample kepler.d/conf.yaml for all available configuration options.

  2. Restart the Agent.

Validation

Run the Agent’s status subcommand and look for kepler under the Checks section.

Data Collected

Metrics

kepler.container.usage.bpf_cpu_time.count
(count)		Aggregated value in bpfcputime_ms value from bpf
kepler.container.usage.bpf_net_rx_irq.count
(count)		Aggregated value in bpfnetrx_irq value from bpf
kepler.container.usage.bpf_net_tx_irq.count
(count)		Aggregated value in bpfnettx_irq value from bpf
kepler.container.usage.bpf_page_cache_hit.count
(count)		Aggregated value in bpfpagecache_hit value from bpf
kepler.container.usage.cache_miss.count
(count)		Aggregated value in cache_miss value from bpf
kepler.container.usage.core_joules.count
(count)		Aggregated value in core value from trainedpowermodel
kepler.container.usage.cpu_cycles.count
(count)		Aggregated value in cpu_cycles value from bpf
kepler.container.usage.cpu_instructions.count
(count)		Aggregated value in cpu_instructions value from bpf
kepler.container.usage.dram_joules.count
(count)		Aggregated value in dram value from trainedpowermodel
kepler.container.usage.gpu_joules.count
(count)		Aggregated value in gpu value from nvidia
kepler.container.usage.joules.count
(count)		Aggregated value in joules value from
kepler.container.usage.package_joules.count
(count)		Aggregated value in package value from trainedpowermodel
kepler.container.usage.platform_joules.count
(count)		Aggregated value in platform value from trainedpowermodel
kepler.container.usage.task_clock.count
(count)		Aggregated value in taskclockms value from bpf
kepler.container.usage.uncore_joules.count
(count)		Aggregated value in uncore value from trainedpowermodel
kepler.exporter.build_info
(gauge)		A metric with a constant '1' value labeled by version, revision, branch, goversion from which kepler_exporter was built, and the goos and goarch for the build.
kepler.node.usage.dram_joules.count
(count)		Aggregated value in dram value from trainedpowermodel
kepler.node_info.count
(count)		Aggregated value in value from os
kepler.node.usage.platform_joules.count
(count)		Aggregated value in platform value from trainedpowermodel
kepler.node.usage.uncore_joules.count
(count)		Aggregated value in uncore value from trainedpowermodel
kepler.container.usage.bpf_block_irq.count
(count)		Aggregated value in bpfblockirq value from bpf
kepler.container.usage.other_joules.count
(count)		Aggregated value in other value from trainedpowermodel
kepler.node.usage.core_joules.count
(count)		Aggregated value in core value from trainedpowermodel
kepler.node.usage.package_joules.count
(count)		Aggregated value in package value from trainedpowermodel
kepler.promhttp.metric.handler.errors.count
(count)		Total number of internal errors encountered by the promhttp metric handler.

Events

Kepler does not include any events.

Troubleshooting

Need help? Contact Datadog support.