2020-05-04 18:10:20 +00:00
---
type: reference, concepts
2022-05-30 00:08:35 +00:00
stage: Systems
2020-07-31 15:10:14 +00:00
group: Distribution
2020-11-26 06:09:20 +00:00
info: To determine the technical writer assigned to the Stage/Group associated with this page, see https://about.gitlab.com/handbook/engineering/ux/technical-writing/#assignments
2020-05-04 18:10:20 +00:00
---
2020-08-03 15:09:44 +00:00
2021-03-02 15:10:57 +00:00
# Reference architectures **(FREE SELF)**
2020-05-04 18:10:20 +00:00
You can set up GitLab on a single server or scale it up to serve many users.
2020-07-31 15:10:14 +00:00
This page details the recommended Reference Architectures that were built and
2020-12-16 00:09:58 +00:00
verified by the GitLab Quality and Support teams.
2020-05-04 18:10:20 +00:00
2020-07-31 15:10:14 +00:00
Below is a chart representing each architecture tier and the number of users
2021-03-17 15:09:03 +00:00
they can handle. As your number of users grow with time, it's recommended that
2020-07-31 15:10:14 +00:00
you scale GitLab accordingly.
2020-05-04 18:10:20 +00:00
<!-- Internal link: https://docs.google.com/spreadsheets/d/1obYP4fLKkVVDOljaI3 - ozhmCiPtEeMblbBKkf2OADKs/edit#gid=1403207183 -->
2020-07-31 15:10:14 +00:00
For GitLab instances with less than 2,000 users, it's recommended that you use
2020-09-07 15:09:04 +00:00
the [default setup ](#automated-backups ) by
2021-02-11 09:09:18 +00:00
[installing GitLab ](../../install/index.md ) on a single machine to minimize
2020-07-31 15:10:14 +00:00
maintenance and resource costs.
2020-05-04 18:10:20 +00:00
2020-12-16 00:09:58 +00:00
If your organization has more than 2,000 users, the recommendation is to scale the
GitLab components to multiple machine nodes. The machine nodes are grouped by
2020-07-31 15:10:14 +00:00
components. The addition of these nodes increases the performance and
scalability of to your GitLab instance.
2020-05-04 18:10:20 +00:00
When scaling GitLab, there are several factors to consider:
- Multiple application nodes to handle frontend traffic.
- A load balancer is added in front to distribute traffic across the application nodes.
- The application nodes connects to a shared file server and PostgreSQL and Redis services on the backend.
2022-03-30 09:08:12 +00:00
## Available reference architectures
2020-07-31 15:10:14 +00:00
Depending on your workflow, the following recommended reference architectures
may need to be adapted accordingly. Your workload is influenced by factors
including how active your users are, how much automation you use, mirroring,
and repository/change size. Additionally the displayed memory values are
provided by [GCP machine types ](https://cloud.google.com/compute/docs/machine-types ).
For different cloud vendors, attempt to select options that best match the
provided architecture.
2020-05-04 18:10:20 +00:00
2022-01-28 15:14:45 +00:00
### GitLab package (Omnibus)
2020-05-08 18:09:55 +00:00
2022-03-30 09:08:12 +00:00
The following reference architectures, where the GitLab package is used, are available:
2020-05-08 18:09:55 +00:00
- [Up to 1,000 users ](1k_users.md )
- [Up to 2,000 users ](2k_users.md )
- [Up to 3,000 users ](3k_users.md )
- [Up to 5,000 users ](5k_users.md )
- [Up to 10,000 users ](10k_users.md )
- [Up to 25,000 users ](25k_users.md )
- [Up to 50,000 users ](50k_users.md )
2020-05-04 18:10:20 +00:00
2022-01-28 15:14:45 +00:00
### Cloud native hybrid
2021-06-23 12:07:58 +00:00
The following Cloud Native Hybrid reference architectures, where select recommended components can be run in Kubernetes, are available:
2021-08-18 18:10:08 +00:00
- [Up to 2,000 users ](2k_users.md#cloud-native-hybrid-reference-architecture-with-helm-charts-alternative )
2021-07-22 09:08:22 +00:00
- [Up to 3,000 users ](3k_users.md#cloud-native-hybrid-reference-architecture-with-helm-charts-alternative )
2021-07-15 09:09:35 +00:00
- [Up to 5,000 users ](5k_users.md#cloud-native-hybrid-reference-architecture-with-helm-charts-alternative )
2021-06-23 12:07:58 +00:00
- [Up to 10,000 users ](10k_users.md#cloud-native-hybrid-reference-architecture-with-helm-charts-alternative )
2021-07-08 09:09:33 +00:00
- [Up to 25,000 users ](25k_users.md#cloud-native-hybrid-reference-architecture-with-helm-charts-alternative )
2021-06-29 18:07:04 +00:00
- [Up to 50,000 users ](50k_users.md#cloud-native-hybrid-reference-architecture-with-helm-charts-alternative )
2021-06-23 12:07:58 +00:00
2020-12-15 18:10:06 +00:00
A GitLab [Premium or Ultimate ](https://about.gitlab.com/pricing/#self-managed ) license is required
to get assistance from Support with troubleshooting the [2,000 users ](2k_users.md )
and higher reference architectures.
[Read more about our definition of scaled architectures ](https://about.gitlab.com/support/#definition-of-scaled-architecture ).
2022-03-30 09:08:12 +00:00
## Validation and test results
The [Quality Engineering team ](https://about.gitlab.com/handbook/engineering/quality/quality-engineering/ )
does regular smoke and performance tests for the reference architectures to ensure they
remain compliant.
### Why we perform the tests
The Quality Department has a focus on measuring and improving the performance
of GitLab, as well as creating and validating reference architectures that
self-managed customers can rely on as performant configurations.
For more information, see our [handbook page ](https://about.gitlab.com/handbook/engineering/quality/performance-and-scalability/ ).
2022-01-26 15:12:36 +00:00
2022-03-30 09:08:12 +00:00
### How we perform the tests
2022-01-26 15:12:36 +00:00
2022-03-30 09:08:12 +00:00
Testing occurs against all reference architectures and cloud providers in an automated and ad-hoc fashion. This is done by two tools:
- The [GitLab Environment Toolkit ](https://gitlab.com/gitlab-org/gitlab-environment-toolkit ) for building the environments.
- The [GitLab Performance Tool ](https://gitlab.com/gitlab-org/quality/performance ) for performance testing.
Network latency on the test environments between components on all Cloud Providers were measured at < 5ms. Note that this is shared as an observation and not as an implicit recommendation .
We aim to have a "test smart" approach where architectures tested have a good range that can also apply to others. Testing focuses on 10k Omnibus on GCP as the testing has shown this is a good bellwether for the other architectures and cloud providers as well as Cloud Native Hybrids.
The Standard Reference Architectures are designed to be platform agnostic, with everything being run on VMs via [Omnibus GitLab ](https://docs.gitlab.com/omnibus/ ). While testing occurs primarily on GCP, ad-hoc testing has shown that they perform similarly on equivalently specced hardware on other Cloud Providers or if run on premises (bare-metal).
Testing on these reference architectures is performed with the
[GitLab Performance Tool ](https://gitlab.com/gitlab-org/quality/performance )
at specific coded workloads, and the throughputs used for testing are
calculated based on sample customer data. Select the
[reference architecture ](#available-reference-architectures ) that matches your scale.
Each endpoint type is tested with the following number of requests per second (RPS)
per 1,000 users:
- API: 20 RPS
- Web: 2 RPS
- Git (Pull): 2 RPS
- Git (Push): 0.4 RPS (rounded to nearest integer)
### How to interpret the results
NOTE:
2022-05-24 12:09:04 +00:00
Read our blog post on [how our QA team leverages GitLab performance testing tool ](https://about.gitlab.com/blog/2020/02/18/how-were-building-up-performance-testing-of-gitlab/ ).
2022-03-30 09:08:12 +00:00
Testing is done publicly and all results are shared.
2022-01-26 15:12:36 +00:00
The following table details the testing done against the reference architectures along with the frequency and results. Additional testing is continuously evaluated, and the table is updated accordingly.
< style >
table.test-coverage td {
2022-02-22 18:17:57 +00:00
border-top: 1px solid #dbdbdb ;
2022-01-26 15:12:36 +00:00
border-left: 1px solid #dbdbdb ;
border-right: 1px solid #dbdbdb ;
border-bottom: 1px solid #dbdbdb ;
}
table.test-coverage th {
2022-02-22 18:17:57 +00:00
border-top: 1px solid #dbdbdb ;
2022-01-26 15:12:36 +00:00
border-left: 1px solid #dbdbdb ;
border-right: 1px solid #dbdbdb ;
border-bottom: 1px solid #dbdbdb ;
}
< / style >
< table class = "test-coverage" >
< col >
< colgroup span = "2" > < / colgroup >
< colgroup span = "2" > < / colgroup >
< tr >
< th rowspan = "2" > Reference< br / > Architecture< / th >
< th style = "text-align: center" colspan = "2" scope = "colgroup" > GCP (* also proxy for Bare-Metal)< / th >
< th style = "text-align: center" colspan = "2" scope = "colgroup" > AWS< / th >
< th style = "text-align: center" colspan = "2" scope = "colgroup" > Azure< / th >
< / tr >
< tr >
< th scope = "col" > Omnibus< / th >
< th scope = "col" > Cloud Native Hybrid< / th >
< th scope = "col" > Omnibus< / th >
< th scope = "col" > Cloud Native Hybrid< / th >
< th scope = "col" > Omnibus< / th >
< / tr >
< tr >
< th scope = "row" > 1k< / th >
2022-04-29 18:08:18 +00:00
< td > < a href = "https://gitlab.com/gitlab-org/quality/performance/-/wikis/Benchmarks/Latest/1k" > Weekly< / a > < / td >
2022-01-26 15:12:36 +00:00
< td > < / td >
< td > < / td >
< td > < / td >
< td > < / td >
< / tr >
< tr >
< th scope = "row" > 2k< / th >
2022-04-29 18:08:18 +00:00
< td > < a href = "https://gitlab.com/gitlab-org/quality/performance/-/wikis/Benchmarks/Latest/2k" > Weekly< / a > < / td >
2022-01-26 15:12:36 +00:00
< td > < / td >
< td > < / td >
< td > < / td >
< td > < / td >
< / tr >
< tr >
< th scope = "row" > 3k< / th >
< td > < a href = "https://gitlab.com/gitlab-org/quality/performance/-/wikis/Benchmarks/Latest/3k" > Weekly< / a > < / td >
< td > < / td >
< td > < / td >
2022-04-29 18:08:18 +00:00
< td > < a href = "https://gitlab.com/gitlab-org/quality/performance/-/wikis/Benchmarks/Latest/3k_hybrid_aws_services" > Weekly< / a > < / td >
2022-01-26 15:12:36 +00:00
< td > < / td >
< / tr >
< tr >
< th scope = "row" > 5k< / th >
< td > < a href = "https://gitlab.com/gitlab-org/quality/performance/-/wikis/Benchmarks/Latest/5k" > Weekly< / a > < / td >
< td > < / td >
< td > < / td >
< td > < / td >
< td > < / td >
< / tr >
< tr >
< th scope = "row" > 10k< / th >
< td > < a href = "https://gitlab.com/gitlab-org/quality/performance/-/wikis/Benchmarks/Latest/10k" > Daily< / a > < / td >
2022-04-29 18:08:18 +00:00
< td > < a href = "https://gitlab.com/gitlab-org/quality/performance/-/wikis/Benchmarks/Latest/10k_hybrid" > Weekly< / a > < / td >
< td > < a href = "https://gitlab.com/gitlab-org/quality/performance/-/wikis/Benchmarks/Latest/10k_aws" > Weekly< / a > < / td >
< td > < a href = "https://gitlab.com/gitlab-org/quality/performance/-/wikis/Benchmarks/Latest/10k_hybrid_aws_services" > Weekly< / a > < / td >
2022-01-26 15:12:36 +00:00
< td > < a href = "https://gitlab.com/gitlab-org/quality/performance/-/wikis/Past-Results/10k" > Ad-Hoc< / a > < / td >
< / tr >
< tr >
< th scope = "row" > 25k< / th >
< td > < a href = "https://gitlab.com/gitlab-org/quality/performance/-/wikis/Benchmarks/Latest/25k" > Weekly< / a > < / td >
< td > < / td >
< td > < / td >
< td > < / td >
< td > < a href = "https://gitlab.com/gitlab-org/quality/performance/-/wikis/Past-Results/25k" > Ad-Hoc< / a > < / td >
< / tr >
< tr >
< th scope = "row" > 50k< / th >
< td > < a href = "https://gitlab.com/gitlab-org/quality/performance/-/wikis/Benchmarks/Latest/50k" > Weekly< / a > < / td >
< td > < / td >
< td > < a href = "https://gitlab.com/gitlab-org/quality/performance/-/wikis/Past-Results/50k" > Ad-Hoc (inc Cloud Services)< / a > < / td >
< td > < / td >
< td > < / td >
< / tr >
< / table >
2022-03-30 09:08:12 +00:00
## Cost to run
2022-01-26 15:12:36 +00:00
2022-05-20 12:08:50 +00:00
The following table details the cost to run the different reference architectures across GCP, AWS, and Azure. Bare-metal costs are not included here as it varies widely depending on each customer configuration.
2022-04-29 18:08:18 +00:00
2022-01-26 15:12:36 +00:00
< table class = "test-coverage" >
< col >
< colgroup span = "2" > < / colgroup >
< colgroup span = "2" > < / colgroup >
< tr >
< th rowspan = "2" > Reference< br / > Architecture< / th >
< th style = "text-align: center" colspan = "2" scope = "colgroup" > GCP< / th >
< th style = "text-align: center" colspan = "2" scope = "colgroup" > AWS< / th >
< th style = "text-align: center" colspan = "2" scope = "colgroup" > Azure< / th >
< / tr >
< tr >
< th scope = "col" > Omnibus< / th >
< th scope = "col" > Cloud Native Hybrid< / th >
< th scope = "col" > Omnibus< / th >
< th scope = "col" > Cloud Native Hybrid< / th >
< th scope = "col" > Omnibus< / th >
< / tr >
< tr >
< th scope = "row" > 1k< / th >
< td > < a href = "https://cloud.google.com/products/calculator#id=a6d6a94a-c7dc-4c22-85c4-7c5747f272ed" > Calculated cost< / a > < / td >
< td > < / td >
2022-04-08 12:08:48 +00:00
< td > < a href = "https://calculator.aws/#/estimate?id=b51f178f4403b69a63f6eb33ea425f82de3bf249" > Calculated cost< / a > < / td >
2022-01-26 15:12:36 +00:00
< td > < / td >
2022-04-08 12:08:48 +00:00
< td > < a href = "https://azure.com/e/1adf30bef7e34ceba9efa97c4470417b" > Calculated cost< / a > < / td >
2022-01-26 15:12:36 +00:00
< / tr >
< tr >
< th scope = "row" > 2k< / th >
< td > < a href = "https://cloud.google.com/products/calculator#id=84d11491-d72a-493c-a16e-650931faa658" > Calculated cost< / a > < / td >
< td > < / td >
2022-04-08 12:08:48 +00:00
< td > < a href = "https://calculator.aws/#/estimate?id=dce36b5cb6ab25211f74e47233d77f58fefb54e2" > Calculated cost< / a > < / td >
2022-01-26 15:12:36 +00:00
< td > < / td >
2022-04-08 12:08:48 +00:00
< td > < a href = "https://azure.com/e/72764902f3854f798407fb03c3de4b6f" > Calculated cost< / a > < / td >
2022-01-26 15:12:36 +00:00
< / tr >
< tr >
< th scope = "row" > 3k< / th >
< td > < a href = "https://cloud.google.com/products/calculator/#id=ac4838e6-9c40-4a36-ac43-6d1bc1843e08" > Calculated cost< / a > < / td >
< td > < / td >
2022-04-08 12:08:48 +00:00
< td > < a href = "https://calculator.aws/#/estimate?id=b1c5b4e32e990eaeb035a148255132bd28988760" > Calculated cost< / a > < / td >
2022-01-26 15:12:36 +00:00
< td > < / td >
2022-04-08 12:08:48 +00:00
< td > < a href = "https://azure.com/e/0dbfc575051943b9970e5d8ace03680d" > Calculated cost< / a > < / td >
2022-01-26 15:12:36 +00:00
< / tr >
< tr >
< th scope = "row" > 5k< / th >
< td > < a href = "https://cloud.google.com/products/calculator/#id=8742e8ea-c08f-4e0a-b058-02f3a1c38a2f" > Calculated cost< / a > < / td >
< td > < / td >
2022-04-08 12:08:48 +00:00
< td > < a href = "https://calculator.aws/#/estimate?id=2bf1af883096e6f4c6efddb4f3c35febead7fec2" > Calculated cost< / a > < / td >
2022-01-26 15:12:36 +00:00
< td > < / td >
2022-04-08 12:08:48 +00:00
< td > < a href = "https://azure.com/e/8f618711ffec4b039f1581871ca6a7c9" > Calculated cost< / a > < / td >
2022-01-26 15:12:36 +00:00
< / tr >
< tr >
< th scope = "row" > 10k< / th >
< td > < a href = "https://cloud.google.com/products/calculator#id=e77713f6-dc0b-4bb3-bcef-cea904ac8efd" > Calculated cost< / a > < / td >
< td > < / td >
2022-04-08 12:08:48 +00:00
< td > < a href = "https://calculator.aws/#/estimate?id=1d374df13c0f2088d332ab0134f5b1d0f717259e" > Calculated cost< / a > < / td >
2022-01-26 15:12:36 +00:00
< td > < / td >
2022-04-08 12:08:48 +00:00
< td > < a href = "https://azure.com/e/de3da8286dda4d4db1362932bc75410b" > Calculated cost< / a > < / td >
2022-01-26 15:12:36 +00:00
< / tr >
< tr >
< th scope = "row" > 25k< / th >
< td > < a href = "https://cloud.google.com/products/calculator#id=925386e1-c01c-4c0a-8d7d-ebde1824b7b0" > Calculated cost< / a > < / td >
< td > < / td >
2022-04-08 12:08:48 +00:00
< td > < a href = "https://calculator.aws/#/estimate?id=46fe6a6e9256d9b7779fae59fbbfa7e836942b7d" > Calculated cost< / a > < / td >
2022-01-26 15:12:36 +00:00
< td > < / td >
2022-04-08 12:08:48 +00:00
< td > < a href = "https://azure.com/e/69724ebd82914a60857da6a3ace05a64" > Calculate cost< / a > < / td >
2022-01-26 15:12:36 +00:00
< / tr >
< tr >
< th scope = "row" > 50k< / th >
< td > < a href = "https://cloud.google.com/products/calculator/#id=8006396b-88ee-40cd-a1c8-77cdefa4d3c8" > Calculated cost< / a > < / td >
< td > < / td >
2022-04-08 12:08:48 +00:00
< td > < a href = "https://calculator.aws/#/estimate?id=e15926b1a3c7139e4faf390a3875ff807d2ab91c" > Calculated cost< / a > < / td >
2022-01-26 15:12:36 +00:00
< td > < / td >
2022-04-08 12:08:48 +00:00
< td > < a href = "https://azure.com/e/3f973040ebc14023933d35f576c89846" > Calculated cost< / a > < / td >
2022-02-22 18:17:57 +00:00
< / tr >
< / table >
2022-03-30 09:08:12 +00:00
## Recommended cloud providers and services
2022-02-22 18:17:57 +00:00
NOTE:
The following lists are non exhaustive. Generally, other cloud providers not listed
2022-06-21 00:08:43 +00:00
here likely work with the same specs, but this hasn't been validated.
2022-02-22 18:17:57 +00:00
Additionally, when it comes to other cloud provider services not listed here,
it's advised to be cautious as each implementation can be notably different
and should be tested thoroughly before production use.
Through testing and real life usage, the Reference Architectures are validated and supported on the following cloud providers:
< table >
< thead >
< tr >
< th > Reference Architecture< / th >
< th > GCP< / th >
< th > AWS< / th >
< th > Azure< / th >
< th > Bare Metal< / th >
< / tr >
< / thead >
< tbody >
< tr >
< td > Omnibus< / td >
< td > ✅< / td >
< td > ✅< / td >
< td > ✅< / td >
< td > ✅< / td >
< / tr >
< tr >
< td > Cloud Native Hybrid< / td >
< td > ✅< / td >
< td > ✅< / td >
< td > < / td >
< td > < / td >
< / tr >
< / tbody >
< / table >
Additionally, the following cloud provider services are validated and supported for use as part of the Reference Architectures:
< table >
< thead >
< tr >
< th > Cloud Service< / th >
< th > GCP< / th >
< th > AWS< / th >
< th > Bare Metal< / th >
< / tr >
< / thead >
< tbody >
< tr >
< td > Object Storage< / td >
< td > ✅ < a href = "https://cloud.google.com/storage" target = "_blank" > Cloud Storage< / a > < / td >
< td > ✅ < a href = "https://aws.amazon.com/s3/" target = "_blank" > S3< / a > < / td >
< td > ✅ < a href = "https://min.io/" target = "_blank" > MinIO< / a > < / td >
< / tr >
< tr >
< td > Database< / td >
< td > ✅ < a href = "https://cloud.google.com/sql" target = "_blank" rel = "noopener noreferrer" > Cloud SQL< / a > < / td >
< td > ✅ < a href = "https://aws.amazon.com/rds/" target = "_blank" rel = "noopener noreferrer" > RDS< / a > < / td >
< td > < / td >
< / tr >
< tr >
< td > Redis< / td >
< td > < / td >
2022-05-11 21:08:09 +00:00
< td > ✅ < a href = "https://aws.amazon.com/elasticache/" target = "_blank" rel = "noopener noreferrer" > ElastiCache< / a > < / td >
2022-01-26 15:12:36 +00:00
< td > < / td >
< / tr >
2022-02-22 18:17:57 +00:00
< / tbody >
2022-01-26 15:12:36 +00:00
< / table >
2022-02-22 18:17:57 +00:00
The following specific cloud provider services have been found to have issues in terms of either functionality or performance. As such, they either have caveats that should be considered or are not recommended:
- [Azure Blob Storage ](https://azure.microsoft.com/en-gb/services/storage/blobs/ ) has been found to have performance limits that can impact production use at certain times. For larger Reference Architectures the service may not be sufficient for production use and an alternative is recommended for use instead.
- [Azure Database for PostgreSQL Server ](https://azure.microsoft.com/en-gb/services/postgresql/#overview ) (Single / Flexible) is not recommended for use due to notable performance issues or missing functionality.
2022-03-24 18:07:52 +00:00
- [AWS Aurora Database ](https://aws.amazon.com/rds/aurora/ ) is not recommended due to compatibility issues.
2022-02-22 18:17:57 +00:00
NOTE:
As a general rule we unfortunately don't recommend Azure Services at this time.
If required, we advise thorough testing is done at your intended scale
over a sustained period to validate if the service is suitable.
2020-05-15 00:08:06 +00:00
## Availability Components
2020-05-04 18:10:20 +00:00
2020-07-31 15:10:14 +00:00
GitLab comes with the following components for your use, listed from least to
most complex:
2020-05-04 18:10:20 +00:00
2020-09-07 15:09:04 +00:00
- [Automated backups ](#automated-backups )
- [Traffic load balancer ](#traffic-load-balancer )
- [Zero downtime updates ](#zero-downtime-updates )
- [Automated database failover ](#automated-database-failover )
- [Instance level replication with GitLab Geo ](#instance-level-replication-with-gitlab-geo )
2020-05-04 18:10:20 +00:00
2020-05-15 00:08:06 +00:00
As you implement these components, begin with a single server and then do
2020-05-04 18:10:20 +00:00
backups. Only after completing the first server should you proceed to the next.
2022-06-21 00:08:43 +00:00
Also, not implementing extra servers for GitLab doesn't necessarily mean that you have
2020-05-15 00:08:06 +00:00
more downtime. Depending on your needs and experience level, single servers can
2020-05-04 18:10:20 +00:00
have more actual perceived uptime for your users.
2021-03-02 15:10:57 +00:00
### Automated backups
2020-05-04 18:10:20 +00:00
> - Level of complexity: **Low**
> - Required domain knowledge: PostgreSQL, GitLab configurations, Git
This solution is appropriate for many teams that have the default GitLab installation.
With automatic backups of the GitLab repositories, configuration, and the database,
2020-05-15 00:08:06 +00:00
this can be an optimal solution if you don't have strict requirements.
2020-05-04 18:10:20 +00:00
[Automated backups ](../../raketasks/backup_restore.md#configuring-cron-to-make-daily-backups )
is the least complex to setup. This provides a point-in-time recovery of a predetermined schedule.
2021-01-29 12:09:29 +00:00
### Traffic load balancer **(PREMIUM SELF)**
2020-05-04 18:10:20 +00:00
> - Level of complexity: **Medium**
> - Required domain knowledge: HAProxy, shared storage, distributed systems
This requires separating out GitLab into multiple application nodes with an added
2022-06-21 00:08:43 +00:00
[load balancer ](../load_balancer.md ). The load balancer distributes traffic
2020-05-04 18:10:20 +00:00
across GitLab application nodes. Meanwhile, each application node connects to a
shared file server and database systems on the back end. This way, if one of the
application servers fails, the workflow is not interrupted.
[HAProxy ](https://www.haproxy.org/ ) is recommended as the load balancer.
2020-05-15 00:08:06 +00:00
With this added component you have a number of advantages compared
2020-05-04 18:10:20 +00:00
to the default installation:
- Increase the number of users.
- Enable zero-downtime upgrades.
- Increase availability.
2021-01-12 12:10:49 +00:00
For more details on how to configure a traffic load balancer with GitLab, you can refer
to any of the [available reference architectures ](#available-reference-architectures ) with more than 1,000 users.
2021-01-29 12:09:29 +00:00
### Zero downtime updates **(PREMIUM SELF)**
2020-05-09 00:09:39 +00:00
> - Level of complexity: **Medium**
> - Required domain knowledge: PostgreSQL, HAProxy, shared storage, distributed systems
2021-09-15 18:11:29 +00:00
GitLab supports [zero-downtime upgrades ](../../update/zero_downtime.md ).
2022-04-13 09:10:26 +00:00
Single GitLab nodes can be updated with only a [few minutes of downtime ](../../update/index.md#upgrade-based-on-installation-method ).
2020-07-29 12:09:45 +00:00
To avoid this, we recommend to separate GitLab into several application nodes.
2022-06-21 00:08:43 +00:00
As long as at least one of each component is online and capable of handling the instance's usage load, your team's productivity is not interrupted during the update.
2020-05-09 00:09:39 +00:00
2021-01-28 12:09:54 +00:00
### Automated database failover **(PREMIUM SELF)**
2020-05-04 18:10:20 +00:00
> - Level of complexity: **High**
2021-06-14 15:09:48 +00:00
> - Required domain knowledge: PgBouncer, Patroni, shared storage, distributed systems
2020-05-04 18:10:20 +00:00
By adding automatic failover for database systems, you can enable higher uptime
2020-05-15 00:08:06 +00:00
with additional database nodes. This extends the default database with
2020-05-04 18:10:20 +00:00
cluster management and failover policies.
2021-06-14 15:09:48 +00:00
[PgBouncer in conjunction with Patroni ](../postgresql/replication_and_failover.md )
2020-06-11 09:08:16 +00:00
is recommended.
2020-05-04 18:10:20 +00:00
2021-01-28 12:09:54 +00:00
### Instance level replication with GitLab Geo **(PREMIUM SELF)**
2020-05-04 18:10:20 +00:00
> - Level of complexity: **Very High**
> - Required domain knowledge: Storage replication
2020-09-15 15:10:08 +00:00
[GitLab Geo ](../geo/index.md ) allows you to replicate your GitLab
2020-05-04 18:10:20 +00:00
instance to other geographical locations as a read-only fully operational instance
that can also be promoted in case of disaster.
2020-11-18 12:09:16 +00:00
## Deviating from the suggested reference architectures
2021-12-15 12:10:49 +00:00
As a general guideline, the further away you move from the Reference Architectures,
2022-06-21 00:08:43 +00:00
the harder it is to get support for it. With any deviation, you're introducing
a layer of complexity that adds challenges to finding out where potential
2020-11-18 12:09:16 +00:00
issues might lie.
The reference architectures use the official GitLab Linux packages (Omnibus
2021-06-23 12:07:58 +00:00
GitLab) or [Helm Charts ](https://docs.gitlab.com/charts/ ) to install and configure the various components. The components are
2020-11-18 12:09:16 +00:00
installed on separate machines (virtualized or bare metal), with machine hardware
requirements listed in the "Configuration" column and equivalent VM standard sizes listed
in GCP/AWS/Azure columns of each [available reference architecture ](#available-reference-architectures ).
Running components on Docker (including Compose) with the same specs should be fine, as Docker is well known in terms of support.
However, it is still an additional layer and may still add some support complexities, such as not being able to run `strace` easily in containers.
Other technologies, like [Docker swarm ](https://docs.docker.com/engine/swarm/ )
are not officially supported, but can be implemented at your own risk. In that
2022-06-21 00:08:43 +00:00
case, GitLab Support is not able to help you.
2020-11-18 12:09:16 +00:00
2021-06-23 12:07:58 +00:00
## Supported modifications for lower user count HA reference architectures
2020-06-16 12:09:00 +00:00
2021-06-23 12:07:58 +00:00
The reference architectures for user counts [3,000 ](3k_users.md ) and up support High Availability (HA).
2020-06-16 12:09:00 +00:00
2021-06-23 12:07:58 +00:00
In the specific case you have the requirement to achieve HA but have a lower user count, select modifications to the [3,000 user ](3k_users.md ) architecture are supported.
2020-06-16 12:09:00 +00:00
2021-06-23 12:07:58 +00:00
For more details, [refer to this section in the architecture's documentation ](3k_users.md#supported-modifications-for-lower-user-counts-ha ).
