in [source code](https://github.com/VictoriaMetrics/VictoriaMetrics). Just download VictoriaMetrics and see [how to start it](#how-to-start-victoriametrics).
* Supports [Prometheus querying API](https://prometheus.io/docs/prometheus/latest/querying/api/), so it can be used as Prometheus drop-in replacement in Grafana.
VictoriaMetrics implements [MetricsQL](https://github.com/VictoriaMetrics/VictoriaMetrics/wiki/ExtendedPromQL) query language, which is inspired by PromQL.
* Supports global query view. Multiple Prometheus instances may write data into VictoriaMetrics. Later this data may be used in a single query.
* High performance and good scalability for both [inserts](https://medium.com/@valyala/high-cardinality-tsdb-benchmarks-victoriametrics-vs-timescaledb-vs-influxdb-13e6ee64dd6b)
and [selects](https://medium.com/@valyala/when-size-matters-benchmarking-victoriametrics-vs-timescale-and-influxdb-6035811952d4).
[Outperforms InfluxDB and TimescaleDB by up to 20x](https://medium.com/@valyala/measuring-vertical-scalability-for-time-series-databases-in-google-cloud-92550d78d8ae).
* [Uses 10x less RAM than InfluxDB](https://medium.com/@valyala/insert-benchmarks-with-inch-influxdb-vs-victoriametrics-e31a41ae2893) when working with millions of unique time series (aka high cardinality).
* Optimized for time series with high churn rate. Think about [prometheus-operator](https://github.com/coreos/prometheus-operator) metrics from frequent deployments in Kubernetes.
* High data compression, so [up to 70x more data points](https://medium.com/@valyala/when-size-matters-benchmarking-victoriametrics-vs-timescale-and-influxdb-6035811952d4)
may be crammed into limited storage comparing to TimescaleDB.
* Optimized for storage with high-latency IO and low IOPS (HDD and network storage in AWS, Google Cloud, Microsoft Azure, etc). See [graphs from these benchmarks](https://medium.com/@valyala/high-cardinality-tsdb-benchmarks-victoriametrics-vs-timescaledb-vs-influxdb-13e6ee64dd6b).
* A single-node VictoriaMetrics may substitute moderately sized clusters built with competing solutions such as Thanos, Uber M3, Cortex, InfluxDB or TimescaleDB.
See [vertical scalability benchmarks](https://medium.com/@valyala/measuring-vertical-scalability-for-time-series-databases-in-google-cloud-92550d78d8ae)
and [comparing Thanos to VictoriaMetrics cluster](https://medium.com/@valyala/comparing-thanos-to-victoriametrics-cluster-b193bea1683).
* Easy operation:
* VictoriaMetrics consists of a single [small executable](https://medium.com/@valyala/stripping-dependency-bloat-in-victoriametrics-docker-image-983fb5912b0d) without external dependencies.
* All the configuration is done via explicit command-line flags with reasonable defaults.
* All the data is stored in a single directory pointed by `-storageDataPath` flag.
* Easy and fast backups from [instant snapshots](https://medium.com/@valyala/how-victoriametrics-makes-instant-snapshots-for-multi-terabyte-time-series-data-e1f3fb0e0282)
to S3 or GCS with [vmbackup](https://github.com/VictoriaMetrics/VictoriaMetrics/blob/master/app/vmbackup/README.md) / [vmrestore](https://github.com/VictoriaMetrics/VictoriaMetrics/blob/master/app/vmrestore/README.md).
See [this article](https://medium.com/@valyala/speeding-up-backups-for-big-time-series-databases-533c1a927883) for more details.
* Storage is protected from corruption on unclean shutdown (i.e. OOM, hardware reset or `kill -9`) thanks to [the storage architecture](https://medium.com/@valyala/how-victoriametrics-makes-instant-snapshots-for-multi-terabyte-time-series-data-e1f3fb0e0282).
* Supports metrics' ingestion and [backfilling](#backfilling) via the following protocols:
* Ideally works with big amounts of time series data from Kubernetes, IoT sensors, connected cars, industrial telemetry, financial data and various Enterprise workloads.
* Has open source [cluster version](https://github.com/VictoriaMetrics/VictoriaMetrics/tree/cluster).
## Operation
### Table of contents
- [How to start VictoriaMetrics](#how-to-start-victoriametrics)
- [Prometheus setup](#prometheus-setup)
- [Grafana setup](#grafana-setup)
- [How to upgrade VictoriaMetrics?](#how-to-upgrade-victoriametrics)
- [How to apply new config to VictoriaMetrics?](#how-to-apply-new-config-to-victoriametrics)
- [How to send data from InfluxDB-compatible agents such as Telegraf?](#how-to-send-data-from-influxdb-compatible-agents-such-as-telegraf)
- [How to send data from Graphite-compatible agents such as StatsD?](#how-to-send-data-from-graphite-compatible-agents-such-as-statsd)
- [Scalability and cluster version](#scalability-and-cluster-version)
- [Alerting](#alerting)
- [Security](#security)
- [Tuning](#tuning)
- [Monitoring](#monitoring)
- [Troubleshooting](#troubleshooting)
- [Backfilling](#backfilling)
- [Profiling](#profiling)
- [Integrations](#integrations)
- [Roadmap](#roadmap)
- [Contacts](#contacts)
- [Community and contributions](#community-and-contributions)
- [Reporting bugs](#reporting-bugs)
- [Victoria Metrics Logo](#victoria-metrics-logo)
- [Logo Usage Guidelines](#logo-usage-guidelines)
- [Font used:](#font-used)
- [Color Palette:](#color-palette)
- [We kindly ask:](#we-kindly-ask)
### How to start VictoriaMetrics
Just start VictoriaMetrics [executable](https://github.com/VictoriaMetrics/VictoriaMetrics/releases)
or [docker image](https://hub.docker.com/r/victoriametrics/victoria-metrics/) with the desired command-line flags.
The following command-line flags are used the most:
*`-storageDataPath` - path to data directory. VictoriaMetrics stores all the data in this directory. Default path is `victoria-metrics-data` in current working directory.
*`-retentionPeriod` - retention period in months for the data. Older data is automatically deleted. Default period is 1 month.
*`-httpListenAddr` - TCP address to listen to for http requests. By default, it listens port `8428` on all the network interfaces.
*`-graphiteListenAddr` - TCP and UDP address to listen to for Graphite data. By default, it is disabled.
*`-opentsdbListenAddr` - TCP and UDP address to listen to for OpenTSDB data over telnet protocol. By default, it is disabled.
*`-opentsdbHTTPListenAddr` - TCP address to listen to for HTTP OpenTSDB data over `/api/put`. By default, it is disabled.
Pass `-help` to see all the available flags with description and default values.
It is recommended setting up [monitoring](#monitoring) for VictoriaMetrics.
Note that Influx line protocol expects [timestamps in *nanoseconds* by default](https://docs.influxdata.com/influxdb/v1.7/write_protocols/line_protocol_tutorial/#timestamp),
while VictoriaMetrics stores them with *milliseconds* precision.
### How to send data from Graphite-compatible agents such as [StatsD](https://github.com/etsy/statsd)?
1) Enable Graphite receiver in VictoriaMetrics by setting `-graphiteListenAddr` command line flag. For instance,
the following command will enable Graphite receiver in VictoriaMetrics on TCP and UDP port `2003`:
helps to spin up VictoriaMetrics, Prometheus and Grafana with one command.
More details may be found [here](https://github.com/VictoriaMetrics/VictoriaMetrics/tree/master/deployment/docker#folder-contains-basic-images-and-tools-for-building-and-running-victoria-metrics-in-docker).
### Setting up service
Read [these instructions](https://github.com/VictoriaMetrics/VictoriaMetrics/issues/43) on how to set up VictoriaMetrics as a service in your OS.
VictoriaMetrics can create [instant snapshots](https://medium.com/@valyala/how-victoriametrics-makes-instant-snapshots-for-multi-terabyte-time-series-data-e1f3fb0e0282)
for all the data stored under `-storageDataPath` directory.
Navigate to `http://<victoriametrics-addr>:8428/snapshot/create` in order to create an instant snapshot.
The page will return the following JSON response:
```
{"status":"ok","snapshot":"<snapshot-name>"}
```
Snapshots are created under `<-storageDataPath>/snapshots` directory, where `<-storageDataPath>`
is the command-line flag value. Snapshots can be archived to backup storage at any time
with [vmbackup](https://github.com/VictoriaMetrics/VictoriaMetrics/blob/master/app/vmbackup/README.md).
The `http://<victoriametrics-addr>:8428/snapshot/list` page contains the list of available snapshots.
Navigate to `http://<victoriametrics-addr>:8428/snapshot/delete?snapshot=<snapshot-name>` in order
to delete `<snapshot-name>` snapshot.
Navigate to `http://<victoriametrics-addr>:8428/snapshot/delete_all` in order to delete all the snapshots.
Steps for restoring from a snapshot:
1. Stop VictoriaMetrics with `kill -INT`.
2. Restore snapshot contents from backup with [vmrestore](https://github.com/VictoriaMetrics/VictoriaMetrics/blob/master/app/vmrestore/README.md)
to the directory pointed by `-storageDataPath`.
3. Start VictoriaMetrics.
### How to delete time series?
Send a request to `http://<victoriametrics-addr>:8428/api/v1/admin/tsdb/delete_series?match[]=<timeseries_selector_for_delete>`,
where `<timeseries_selector_for_delete>` may contain any [time series selector](https://prometheus.io/docs/prometheus/latest/querying/basics/#time-series-selectors)
for metrics to delete. After that all the time series matching the given selector are deleted. Storage space for
the deleted time series isn't freed instantly - it is freed during subsequent merges of data files.
It is recommended verifying which metrics will be deleted with the call to `http://<victoria-metrics-addr>:8428/api/v1/series?match[]=<timeseries_selector_for_delete>`
before actually deleting the metrics.
### How to export time series?
Send a request to `http://<victoriametrics-addr>:8428/api/v1/export?match[]=<timeseries_selector_for_export>`,
where `<timeseries_selector_for_export>` may contain any [time series selector](https://prometheus.io/docs/prometheus/latest/querying/basics/#time-series-selectors)
for metrics to export. Use `{__name__!=""}` selector for fetching all the time series.
The response would contain all the data for the selected time series in [JSON streaming format](https://en.wikipedia.org/wiki/JSON_streaming#Line-delimited_JSON).
Each request to `/api/v1/import` can load up to a single vCPU core on VictoriaMetrics. Import speed can be improved by splitting the original file into smaller parts
and importing them concurrently. Note that the original file must be split on newlines.
at `http://<victoriametrics-addr>:8428/federate?match[]=<timeseries_selector_for_federation>`.
Optional `start` and `end` args may be added to the request in order to scrape the last point for each selected time series on the `[start ... end]` interval.
`start` and `end` may contain either unix timestamp in seconds or [RFC3339](https://www.ietf.org/rfc/rfc3339.txt) values. By default, the last point
on the interval `[now - max_lookback ... now]` is scraped for each time series. The default value for `max_lookback` is `5m` (5 minutes), but it can be overridden.
For instance, `/federate?match[]=up&max_lookback=1h` would return last points on the `[now - 1h ... now]` interval. This may be useful for time series federation
with scrape intervals exceeding `5m`.
### Capacity planning
A rough estimation of the required resources for ingestion path:
* RAM size: less than 1KB per active time series. So, ~1GB of RAM is required for 1M active time series.
Time series is considered active if new data points have been added to it recently or if it has been recently queried.
The number of active time series may be obtained from `vm_cache_entries{type="storage/hour_metric_ids"}` metric
exproted on the `/metrics` page.
VictoriaMetrics stores various caches in RAM. Memory size for these caches may be limited by `-memory.allowedPercent` flag.
* CPU cores: a CPU core per 300K inserted data points per second. So, ~4 CPU cores are required for processing
the insert stream of 1M data points per second. The ingestion rate may be lower for high cardinality data or for time series with high number of labels.
See [this article](https://medium.com/@valyala/insert-benchmarks-with-inch-influxdb-vs-victoriametrics-e31a41ae2893) for details.
If you see lower numbers per CPU core, then it is likely active time series info doesn't fit caches,
so you need more RAM for lowering CPU usage.
* Storage space: less than a byte per data point on average. So, ~260GB is required for storing a month-long insert stream
of 100K data points per second.
The actual storage size heavily depends on data randomness (entropy). Higher randomness means higher storage size requirements.
4) Now Prometheus should write data into all the configured `remote_write` urls in parallel.
5) Set up [Promxy](https://github.com/jacksontj/promxy) in front of all the VictoriaMetrics replicas.
6) Set up Prometheus datasource in Grafana that points to Promxy.
If you have Prometheus HA pairs with replicas `r1` and `r2` in each pair, then configure each `r1`
to write data to `victoriametrics-addr-1`, while each `r2` should write data to `victoriametrics-addr-2`.
### Multiple retentions
Just start multiple VictoriaMetrics instances with distinct values for the following flags:
*`-retentionPeriod`
*`-storageDataPath`, so the data for each retention period is saved in a separate directory
*`-httpListenAddr`, so clients may reach VictoriaMetrics instance with proper retention
### Downsampling
There is no downsampling support at the moment, but:
- VictoriaMetrics is optimized for querying big amounts of raw data. See benchmark results for heavy queries
in [this article](https://medium.com/@valyala/measuring-vertical-scalability-for-time-series-databases-in-google-cloud-92550d78d8ae).
- VictoriaMetrics has good compression for on-disk data. See [this article](https://medium.com/@valyala/victoriametrics-achieving-better-compression-for-time-series-data-than-gorilla-317bc1f95932)
for details.
These properties reduce the need in downsampling. We plan to implement downsampling in the future.
See [this issue](https://github.com/VictoriaMetrics/VictoriaMetrics/issues/36) for details.
### Multi-tenancy
Single-node VictoriaMetrics doesn't support multi-tenancy. Use [cluster version](https://github.com/VictoriaMetrics/VictoriaMetrics/tree/cluster) instead.
### Scalability and cluster version
Though single-node VictoriaMetrics cannot scale to multiple nodes, it is optimized for resource usage - storage size / bandwidth / IOPS, RAM, CPU.
This means that a single-node VictoriaMetrics may scale vertically and substitute a moderately sized cluster built with competing solutions
such as Thanos, Uber M3, InfluxDB or TimescaleDB. See [vertical scalability benchmarks](https://medium.com/@valyala/measuring-vertical-scalability-for-time-series-databases-in-google-cloud-92550d78d8ae).
So try single-node VictoriaMetrics at first and then [switch to cluster version](https://github.com/VictoriaMetrics/VictoriaMetrics/tree/cluster) if you still need
horizontally scalable long-term remote storage for really large Prometheus deployments.
[Contact us](mailto:info@victoriametrics.com) for paid support.
### Alerting
VictoriaMetrics doesn't support rule evaluation and alerting yet, so these actions must be performed either
on [Prometheus side](https://prometheus.io/docs/alerting/overview/) or on [Grafana side](https://grafana.com/docs/alerting/rules/).
### Security
Do not forget protecting sensitive endpoints in VictoriaMetrics when exposing it to untrusted networks such as the internet.
Consider setting the following command-line flags:
*`-tls`, `-tlsCertFile` and `-tlsKeyFile` for switching from HTTP to HTTPS.
*`-httpAuth.username` and `-httpAuth.password` for protecting all the HTTP endpoints
with [HTTP Basic Authentication](https://en.wikipedia.org/wiki/Basic_access_authentication).
*`-deleteAuthKey` for protecting `/api/v1/admin/tsdb/delete_series` endpoint. See [how to delete time series](#how-to-delete-time-series).
*`-snapshotAuthKey` for protecting `/snapshot*` endpoints. See [how to work with snapshots](#how-to-work-with-snapshots).
Explicitly set internal network interface for TCP and UDP ports for data ingestion with Graphite and OpenTSDB formats.
For example, substitute `-graphiteListenAddr=:2003` with `-graphiteListenAddr=<internal_iface_ip>:2003`.
### Tuning
* There is no need in VictoriaMetrics tuning since it uses reasonable defaults for command-line flags,
which are automatically adjusted for the available CPU and RAM resources.
* There is no need in Operating System tuning since VictoriaMetrics is optimized for default OS settings.
The only option is increasing the limit on [the number of open files in the OS](https://medium.com/@muhammadtriwibowo/set-permanently-ulimit-n-open-files-in-ubuntu-4d61064429a),
so Prometheus instances could establish more connections to VictoriaMetrics.
* The recommended filesystem is `ext4`, the recommended persistent storage is [persistent HDD-based disk on GCP](https://cloud.google.com/compute/docs/disks/#pdspecs),
since it is protected from hardware failures via internal replication and it can be [resized on the fly](https://cloud.google.com/compute/docs/disks/add-persistent-disk#resize_pd).
If you plan storing more than 1TB of data on `ext4` partition or plan extending it to more than 16TB,
then the following options are recommended to pass to `mkfs.ext4`:
```
mkfs.ext4 ... -O 64bit,huge_file,extent -T huge
```
### Monitoring
VictoriaMetrics exports internal metrics in Prometheus format on the `/metrics` page.
Add this page to Prometheus' scrape config in order to collect VictoriaMetrics metrics.
There are officials Grafana dashboards for [single-node VictoriaMetrics](https://grafana.com/dashboards/10229) and [clustered VictoriaMetrics](https://grafana.com/grafana/dashboards/11176).
*`vm_cache_entries{type="storage/hour_metric_ids"}` - the number of time series with new data points during the last hour
aka active time series.
*`rate(vm_new_timeseries_created_total[5m])` - time series churn rate.
*`vm_rows{type="indexdb"}` - the number of rows in inverted index. High value for this number usually mean high churn rate for time series.
* Sum of `vm_rows{type="storage/big"}` and `vm_rows{type="storage/small"}` - total number of `(timestamp, value)` data points
in the database.
* Sum of all the `vm_cache_size_bytes` metrics - the total size of all the caches in the database.
*`vm_allowed_memory_bytes` - the maximum allowed size for caches in the database. It is calculated as `system_memory * <-memory.allowedPercent> / 100`,
where `system_memory` is the amount of system memory and `-memory.allowedPercent` is the corresponding flag value.
*`vm_rows_inserted_total` - the total number of inserted rows since VictoriaMetrics start.
### Troubleshooting
* It is recommended to use default command-line flag values (i.e. don't set them explicitly) until the need
in tweaking these flag values arises.
* If VictoriaMetrics works slowly and eats more than a CPU core per 100K ingested data points per second,
then it is likely you have too many active time series for the current amount of RAM.
It is recommended increasing the amount of RAM on the node with VictoriaMetrics in order to improve
ingestion performance.
Another option is to increase `-memory.allowedPercent` command-line flag value. Be careful with this
option, since too big value for `-memory.allowedPercent` may result in high I/O usage.
* VictoriaMetrics requires free disk space for [merging data files to bigger ones](https://medium.com/@valyala/how-victoriametrics-makes-instant-snapshots-for-multi-terabyte-time-series-data-e1f3fb0e0282).
It may slow down when there is no enough free space left. So make sure `-storageDataPath` directory
has at least 20% of free space comparing to disk size.
* If VictoriaMetrics doesn't work because of certain parts are corrupted due to disk errors,
then just remove directoreis with broken parts. This will recover VictoriaMetrics at the cost
of data loss stored in the broken parts. In the future, `vmrecover` tool will be created
for automatic recovering from such errors.
### Backfilling
Make sure that configured `-retentionPeriod` covers timestamps for the backfilled data.
It is recommended disabling query cache with `-search.disableCache` command-line flag when writing
historical data with timestamps from the past, since the cache assumes that the data is written with
the current timestamps. Query cache can be enabled after the backfilling is complete.
### Profiling
VictoriaMetrics provides handlers for collecting the following [Go profiles](https://blog.golang.org/profiling-go-programs):
- Memory profile. It can be collected with the following command:
