filmov
tv
PromCon 2016: The Prometheus Time Series Database - Björn Rabenstein
Показать описание
* Abstract:
Various time series databases (TSDBs) have been implemented on top of key-value stores with BigTable semantics. The TSDB that sits at the core of the Prometheus monitoring system started with a similar approach and was built on top of LevelDB. The Prometheus server as we know it today, however, uses a highly optimized custom storage layer for bulk sample data, enabling a single server to sustain an ingestion rate of 500,000 samples per second belonging to millions of time series. Very recent improvements of the in-memory representation of sample data resulted in an outstanding compression level of 1.3 bytes per sample in a typical production setup. A journey from fundamental challenges of TSDB design to details of the Prometheus storage layer.
* Speaker biography:
Björn is a production engineer at SoundCloud and one of the main Prometheus developers. Previously, he was a Site Reliability Engineer at Google and a number cruncher for science.
* Slides:
* PromCon website:
Various time series databases (TSDBs) have been implemented on top of key-value stores with BigTable semantics. The TSDB that sits at the core of the Prometheus monitoring system started with a similar approach and was built on top of LevelDB. The Prometheus server as we know it today, however, uses a highly optimized custom storage layer for bulk sample data, enabling a single server to sustain an ingestion rate of 500,000 samples per second belonging to millions of time series. Very recent improvements of the in-memory representation of sample data resulted in an outstanding compression level of 1.3 bytes per sample in a typical production setup. A journey from fundamental challenges of TSDB design to details of the Prometheus storage layer.
* Speaker biography:
Björn is a production engineer at SoundCloud and one of the main Prometheus developers. Previously, he was a Site Reliability Engineer at Google and a number cruncher for science.
* Slides:
* PromCon website:
0:42:57
PromCon 2016: The Prometheus Time Series Database - Björn Rabenstein
0:52:18
PromCon 2016: The History of Prometheus at SoundCloud - Tobias Schmidt
0:24:47
PromCon 2016: Dynamic Monitoring with Prometheus and Rancher - Chris Urwin, Edward Marshall
0:31:30
PromCon 2016: Scaling to a Million Machines with Prometheus - Matthew Campbell
0:30:09
PromCon 2016: Multitenant, Scale-Out Prometheus - Tom Wilkie
0:30:21
PromCon 2016: So You Want to Write an Exporter - Brian Brazil
0:38:28
PromCon 2016: Deploying Prometheus at DigitalOcean - Carlos Amedee
0:26:59
PromCon 2016: Prometheus Is Good for Your Small Startup - Ignacio P. Carretero
0:23:20
PromCon 2016: Prometheus Design and Philosophy - Why It Is the Way It Is - Julius Volz
0:33:59
PromCon 2016: Alerting in the Prometheus Universe - Fabian Reinartz
0:05:38
PromCon 2016: Lightning Talks - Deploying Full Prometheus Stacks via Juju Charms - JuanJo Ciarlante
0:29:55
PromCon 2017: Cortex: Prometheus as a Service, One Year On - Tom Wilkie
0:01:30
PromCon 2016: Conference Recap
0:09:33
PromCon 2016: Welcome and Introduction - Julius Volz
0:06:33
PromCon 2016: Lightning Talks - Why We Love Prometheus (and You Should Too) - Gil Fliker
0:38:00
PromCon 2016: Graphing MySQL Performance with Prometheus and Grafana - Roman Vynar
0:03:36
PromCon 2016: Lightning Talks - Prometheus on the Small Scale - Jonas Große Sundrup
0:24:32
PromCon 2016: Life of a Label - Brian Brazil
0:28:16
PromCon 2016: Prometheus as a Customer-Facing Monitoring Tool for SpatialOS - Dmytro Kislov
0:04:06
PromCon 2016: Lightning Talks - PromQL You Probably (Maybe) Don't Use - Brian Brazil
0:05:02
PromCon 2016: Lightning Talks - vulcan: An API-compatible Alternative to Prometheus - Ian Hansen
0:08:11
PromCon 2016: Lightning Talks - An Exploration of the Formal Properties of PromQL - Brian Brazil
0:31:56
PromCon EU 2019: OpenMetrics: What Does It Mean for You
0:05:15
PromCon 2017: Lightning Talk - Prometheus@Home - Richard Hartmann
Комментарии