[{"data":1,"prerenderedAt":794},["ShallowReactive",2],{"/en-us/blog/how-we-built-gitlab-geo":3,"navigation-en-us":37,"banner-en-us":437,"footer-en-us":447,"blog-post-authors-en-us-Gabriel Mazetto":689,"blog-related-posts-en-us-how-we-built-gitlab-geo":703,"assessment-promotions-en-us":745,"next-steps-en-us":784},{"id":4,"title":5,"authorSlugs":6,"body":8,"categorySlug":9,"config":10,"content":14,"description":8,"extension":25,"isFeatured":12,"meta":26,"navigation":27,"path":28,"publishedDate":20,"seo":29,"stem":33,"tagSlugs":34,"__hash__":36},"blogPosts/en-us/blog/how-we-built-gitlab-geo.yml","How We Built Gitlab Geo",[7],"gabriel-mazetto",null,"engineering",{"slug":11,"featured":12,"template":13},"how-we-built-gitlab-geo",false,"BlogPost",{"title":15,"description":16,"authors":17,"heroImage":19,"date":20,"body":21,"category":9,"tags":22},"How we built GitLab Geo","Take a deep dive into the many architectural decisions we made while building GitLab Geo.",[18],"Gabriel Mazetto","https://res.cloudinary.com/about-gitlab-com/image/upload/v1749678985/Blog/Hero%20Images/how-we-built-geo-cover.jpg","2018-09-14","\n[Geo](https://docs.gitlab.com/ee/administration/geo/index.html), our solution for read-only mirrors of your GitLab instance, started with our co-founder [Dmitriy Zaporozhets](/company/team/#dzaporozhets)’ crazy idea of making not only the repositories, but the entire GitLab instance accessible from multiple geographical locations.\n\nAt that time (Q4 of 2015) there were only a few competitors trying to provide an *automatic mirroring* solution for repositories and/or issue trackers, and they were mostly built around an additional independent instance and a bunch of webhooks to replicate events. Also, in those cases, no other data was shared outside this asynchronous replication channel, and you had to set up the webhook per project and take care of the users yourself. Long story short: this was not practical for any instance with more than a couple of projects.\n\nWe also had a previous experience early that year [using DRBD to migrate 9 TB of data](/blog/moving-all-your-data/) from our dedicated co-location hosting to the AWS cloud,\nwhich didn't provide the scale, performance, or the UX we had in mind for the future.\n\nHere's the history of how we built Geo:\n\n## Phase 1: MVP\n\nGeo's first mission was to provide people who were located in satellite offices, or in distant locations, with fast access to the tools they need to get work done. The plan was not only to make it faster for Git clones to occur in remote offices but also to provide a fully functional read-only version of GitLab: all project issues, Git repositories, Wikis, etc. automatically synchronized from the primary with as little delay as possible.\n\nTo get there we made a few architectural decisions:\n\n#### 1. Use native database replication\n\nThis would allow us to replicate any user-visible information, user content, user and permissions, projects, any project relation to groups/namespaces, etc. Basically, any data ever written to the database in the primary node made readily available to the others, without any extra communication overhead in the webhooks.\n\nIt is also the most [Boring Solution](https://handbook.gitlab.com/handbook/values/#efficiency), as it uses proven technologies developed for databases in the past two decades. To simplify the endeavor we decided to support only PostgreSQL.\n\n#### 2. Use API calls to notify any secondary node of changes that should happen on disk\n\nThis is the second synchronization mechanism. If a new project is created or a repository updated, this notification lets any other node know they have this pending action, and should replicate the new data on disk.\n\n#### 3. Use Git itself to replicate the repositories\n\nWe investigated many alternatives to replicate our repositories, from using basic UNIX tools (like `rsync` or equivalent) to specific distributed file-systems features. We were aiming for a simple solution, as ideally we had to support the lowest common denominator, which is a Linux machine running the default filesystem (ext3 or 4). That limitation ruled out any distributed file-system based implementation.\n\nWe considered `rsync` and its variants as well, which could potentially work for our use case, but that would add significant CPU for each synchronization operation, and we expect it to increase as the repositories get bigger and bigger.\n\nBy using `rsync` we would need to grant more on-disk permissions than we were comfortable doing, and restricting its reach could be an engineering challenge in itself.\n\nThe same can be said for `scp` and its variants. In the end, we decided to use Git itself and benefit from its internal protocol. This was a no-brainer and very easy decision to make. We understood the protocol enough and we already had the required safeguards in place. All we needed was a slightly different authentication mechanism for the node-to-node synchronization.\n\n#### 4. Always push code to the primary, pull code from anywhere\n\nWhen we started Geo, there was no bundled Git support for having a multi-repository \"transactional\" replication, or information on how to implement one.\n\nWe figured out quickly that to implement something on that line it would require either a *global lock* or to implement a variant of [RAFT](https://raft.github.io/)/[PAXOS](https://en.wikipedia.org/wiki/Paxos_(computer_science)) on top of Git internal protocol.\n\nBoth solutions have their downsides and tradeoffs, and adding to that the time and effort to build it correctly, led us to opt for the simplest implementation: always push to the primary, notify secondaries that repository data changed, and have the secondaries fetch the changes. This is also in line with our motto of [Boring Solutions](https://handbook.gitlab.com/handbook/values/#efficiency).\n\nThe initial repository synchronization is no different than doing a `git clone \u003Cremote> --mirror`. The same idea goes for the repository updates, they behave very similarly to a `git fetch \u003Cremote> --prune`. The difference is that we need to replicate additional, internal metadata as well, that is not normally exposed to a regular user.\n\n![GitLab Geo - MVP Synchronization Architecture Diagram](https://about.gitlab.com/images/blogimages/how-we-built-geo/geo-architecture-mvp.png){: .medium.center}\n\n#### 5. Don’t replicate Redis data between nodes\n\nWe initially thought we could replicate Redis as well as the main database in order to share cached data, session information, etc. This would allow us to implement a Single Sign-On solution very easily, and by reusing the cache we would speed up the initial page load.\n\nAt that time Redis only supported **Leader** to **Follower** replication mode and even though it is usually super fast when used in a local network, the fact remains that replicating data across disparate geographical locations can add significant latency.\n\nThis additional latency would impact on the initial objective of simplifying the Single Sign-On implementation. If you simply log in on the primary node and get redirected to the secondary, chances are that the session information would still not be available on the secondary node due to the replication latency.\n\nThat would eventually fix itself by redirecting back and forth, but if the latency is significant enough, your browser will terminate the connection based on the redirect loop prevention feature. Another downside of this approach is that it creates a hard dependency on the primary node being online, or otherwise the secondary node would be inaccessible and/or completely broken.\n\nIn addition to all these issues, we needed an additional Redis instance that supports writing data to it, in order to persist Jobs to our Jobs system on the secondary node.\n\nSo it made sense, in the end, to give up on the idea of replicating Redis, and we started looking for a solution to the authentication problem.\n\n#### 6. Authenticate on the primary node only\n\nBecause we can’t write on the main database of secondary nodes, any auditing logs, brute force protection mechanism, password recovery tokens, etc. can’t have their data and state persisted inside secondary nodes. The only viable solution then is to authenticate on the primary and redirect the user to the secondary.\n\nThis decision also helped with the integration of any company-specific authentication systems. If a company uses internal authentication based on LDAP, CAS or SAML for instance, then they wouldn't have to replicate that system to the other location or configure firewall rule exceptions to accept traffic over the internet.\n\n#### 7. Implement Single Sign-On and Single Sign-Off using OAuth\n\nWith the previous Redis limitations in mind, we looked into alternatives to implement the authentication. We had to choose between either CAS or an OAuth-based one. As we already had OAuth Provider support inside GitLab, we decided to go with that.\n\nBasically, for any Geo node configured in the database we also have a corresponding OAuth application inside GitLab, and whenever a new user tries to log into a Geo node, they get redirected to the primary node and need to \"allow\" the \"Geo application\" to have access to their account credentials at the first login.\n\nThe shortcoming here is that if you are not logged in already and the primary goes down, you can't log in again until the primary node connectivity issue is fixed.\n\n#### 8. Build a read-only layer on the application side to prevent accidents\n\nWe needed this safeguard in place in case any required subsystem was misconfigured. With the read-only layer, we can prevent the instance from diverging from the primary in a non-recoverable way. It's also this layer that prevents anyone from pushing a repository change to the secondary node directly.\n\n#### 9. Don’t replicate any user attachments yet, just redirect to the primary\n\nInstead of trying to replicate user attachments at this stage, we decided to just rewrite the URLs pointing the resource to the primary node instead. This allowed us to iterate faster and still provide a decent experience to the end users.\n\nThey would still enjoy faster access to the repository data and have the web UI rendering the content from a closer location, with the exception of the issue/merge request attachments, avatars etc, which were still being fetched from the primary. But as they are also highly cachable the impact is minimal.\n\nThis was the initial foundation that allowed us to validate Geo as a viable solution. Later on, we took care of replicating the missing data as well.\n\n### Bonus trivia\n\nThe term **Geo** came only after a while, it was previously named as **GitLab RE** (*Read-Only Edition*), followed by **GitLab RO** (*Read Only*) before getting its final name: **GitLab Geo**.\n\n## Phase 2: First-generation synchronization mechanism\n\nWith the MVP implementation done, we needed to pave the way for a stable release. The first part we decided to improve was the notification mechanism for pending changes. During the MVP, we built a custom API endpoint and a buffered queue. That queue was also optimized to store only unique, idempotent events. If a project received three push events in the last few seconds, we only needed to store and process one event notification.\n\nWe decided that instead of building our own custom notification \"protocol\" and implementing some early optimizations, we should leverage existing GitLab internal capabilities: our own webhooks system.\n\n![GitLab Geo - First Generation Synchronization Architecture Diagram](https://about.gitlab.com/images/blogimages/how-we-built-geo/geo-architecture-first-gen.png){: .medium.center}\n\nBy taking that route, we would be forced to \"[drink our own champagne](https://en.wikipedia.org/wiki/Eating_your_own_dog_food#Alternative_terms)\" and as a result, improve our existing functionality. That decision actually resulted in improvements to our system-wide webhooks in a few ways. We added new system-wide webhook events, expanded the granularity of the information available, and fixed some performance issues.\n\nWe've also improved the security of our webhooks implementation by adding ways of verifying that the notification came from a trusted source. Previously the only way to do that relied on whitelisting the originating IP address as a way to establish trust.\n\nThis security limitation was not present in the MVP version, as we reused the admin personal token as the authorization mechanism for the API, which is also not ideal, but better than previous webhook implementation.\n\nI consider this to be the first generation of the synchronization mechanism that was used in the wild. It had a few characteristics: it reacted almost like real-time for small updates, webhook was fast enough and parallelizable to be used on the scale we wanted to support.\n\nAs the very first version of Geo was only concerned with getting repositories available and in-sync, from one location to the other, that's where we focused all of our efforts. At that time, setting up a new Geo node required an almost identical clone of the primary to be available in advance. That included not only replicating the database but also *rsyncing* the repositories from one node to the other. For improved consistency, we required initially a *stop the world* phase in order to not lose changes made during the time between when the backup started and when the secondary node got completely set up.\n\nWhile this was still closer to a barebones solution, it already provided value for remote teams working together in a shared repository or simply in any project that needed to synchronize code between different locations. We had a few customers trying it out and evaluating the potential, but it was still not ready for production use as we were still missing a lot of functionality.\n\nThe *stop the world* phase previously mentioned got phased out later with the help of improved setup documentation. Much later, a good chunk of the initial cloning step got simplified by leveraging some improvements in the next-generation synchronization and by introducing a backfilling mechanism.\n\n### First-generation synchronization pitfalls\n\nWhile our first-generation solution worked fine for the highly active repositories, the use of webhooks as a notification mechanism had some really obvious drawbacks.\n\nIf, for any reason, the notification failed to be delivered, it had to be rescheduled and retried. Also because we were using our internal Jobs system to dispatch the webhooks, having a node go dark for a few hours meant our Jobs system would be busy retrying operations over an unreachable destination for at least that same amount of time.\n\nDepending on the volume of data and how long it has been accumulating changes, that could even fill up the Redis instance disk storage. If that ever happened we would have to resync the whole instance again and start from scratch.\n\nWe've improved the retrying mechanism with custom Geo logic to alleviate the problem, but it was clear to us that this was not going to be a viable solution for a Generally Available (*stable*) release.\n\nAlso because of backoff algorithm in the retrying logic, in conjunction with the asynchronicity aspect of the system, it could lead to important changes taking a lot of time to replicate, especially in less active projects. The busiest ones were less affected, as any update to the repository would get it to the current state rather than to the state when the update notification was issued. And because the project is receiving many updates during the day, it's expected to generate also many notification events.\n\nAny implementation misstep between sending the webhook or receiving and processing it on the other side could mean we would lose that information forever. This was again not a major issue with highly active projects, as it would eventually receive a new, valid update notification which would sync it to the current state, but the outliers could miss it until someone notices or another update arrives much later.\n\nWe also wanted to make Geo a viable Disaster Recovery solution in the long term, so missing updates without a way to recover from it was not an option.\n\n## Phase 3: Second-generation synchronization mechanism\n\nWe started looking for alternative ways of notifying the secondary nodes and also considered switching to other standalone queue systems instead. We were also worried about the lack of control over the order in which the operations would happen in a parallel and asynchronous replication system and on the effect it had on the data on disk.\n\nA few examples of situations that can happen because of the parallelism and the async nature of it:\n\n1. A project removal event can be processed before a project update for the same repository\n1. Renaming, creating a project with the new name and sending new content to it, if processed in an incorrect order, can lead to temporary data loss\n\nThere was also the case when the notification arrived before the database had replicated the required data. As an example, when the node receives the notification for new project creation, but the database doesn't have it yet.\n\nThat required the secondary node to keep a \"mailbox\" until the received events are ready to be processed. As they were basically Jobs, that meant keep retrying until the job succeeded.\n\nConsidering all the complexity we had brought to the application layer, we investigated a few standalone queue systems to which we could offload the burden, but decided ultimately to build an event queue mechanism in PostgreSQL instead, as it had three important advantages:\n\n#### 1. No extra dependencies\nWe were already replicating the database, so there is no need to install, configure and maintain another process, worry about backing up yet another component, integrate it in our Omnibus package, and provide support for our users.\n\n#### 2. No more delayed processing\nIf the event arrives on the other side, the data associated with it will already be there as well. We can also guarantee consistency with transactions and repeat less information than with the webhooks implementation.\n\n#### 3. Easy to retry/restore from backup or in a disaster situation\nWith a standalone queue system, to have a consistent backup solution you either need some sort of \u003Cabbr title=\"Write-Ahead Logging\">WAL\u003C/abbr> files that could help rebuild a consistent state between the systems or do backups in a \"stop the world\" way, otherwise, you may lose data.\n\n### Our implementation\n\nWe took inspiration from how other log-based replication systems work (like the database) and implemented it with a central table as the main source of truth and a few others to hold bookkeeping for specific event types. Any relevant information we used to ship with the webhook notification is now part of this implementation, with extras to support the missing replicable events.\n\nOn the secondary node, these new tables are read by a specific daemon (we call it the Geo Log Cursor), and as the name suggests, it holds a persistent pointer of the last processed event. This allows us to also report the state of replication and monitor if our replication is broken. We also made it highly available, so you can boot up one as **Active** and keep a few extras as **Standby**. If the Active daemon stops responding for a specified amount of time a new election starts and one of the Standbys takes place as the new Active.\n\nThe second part of the new system requires a persistent layer on the secondary node to keep any synchronization state and metadata. This was done by using another PostgreSQL instance.\n\nWe couldn’t reuse the same main instance, as we were replicating with *Streaming replication* mode. With *Streaming replication*, the whole instance is replicated, and you can’t perform any change in it. The alternative to being able to replicate and write in the same instance is to use *Logical replication* mode, but at that time, there was no official *Logical Replication* support available in the PostgreSQL versions we supported (PgLogical was also not a viable alternative back then).\n\nWith the new persistence layer (we call it the *Geo Tracking Database*), we had the foundations built to be able to actively compare the \"desired vs actual\" state, and find missing data on any secondary instance. We built a more robust backfilling mechanism based on that as well.\n\nQuerying between the two database instances (the replicated Secondary, and the Tracking Database), were made much faster and scalable by enabling Postgres FDW ([Foreign Data Wrapper](https://www.postgresql.org/docs/9.6/static/postgres-fdw.html)). That allowed us to query data using a few **LEFT JOIN** operations among the two instances, instead of pooling with multiple queries from the application layer against the two databases in isolation.\n\n![GitLab Geo - Second Generation Synchronization Architecture Diagram](https://about.gitlab.com/images/blogimages/how-we-built-geo/geo-architecture-second-gen.png){: .medium.center}\n\n### Other improvements\n\nAnother important shortcoming fixed was how we replicated the SSH Keys. This was technical debt we needed to pay since the first implementation. Historically, GitLab built the SSH authorization mechanism as with many other Git implementations, by writing each user-provided SSH Key to the `AuthorizedKeys` file on the server and pointing each one to our [gitlab-shell](https://gitlab.com/gitlab-org/gitlab-shell) application.\n\nThis implementation allowed us to authenticate the authorized users, and because we control how the Shell application is invoked, we can pass a specific key ID to it, that can be used later to identify the user on our database and authorize/deny operations to specific repositories.\n\nThe problem with this approach, in general, is that the bigger the user base is, the slower the initial request will be, as OpenSSH will have to perform a scan to the whole file (**O(N)** complexity). With Geo, that's not just about speed but any delay in updating this file either to add a new key or to revoke an existing one is very undesirable.\n\nWhen we decided to fix that we did for both Geo and GitLab Core by using an interesting feature present in newer versions of OpenSSH (6.9 and above), that allows overriding the `AuthorizedKeys` step, switching from reading the keys from a file to invoke a specified CLI instead (*O(1)* complexity). You can read more about it [in the documentation here](https://docs.gitlab.com/ee/administration/operations/fast_ssh_key_lookup.html#doc-nav).\n\nWe fixed another shortcoming around the repository synchronization, switching from Git over SSH protocol, to Git over HTTPS. The initial motivation was to simplify the setup steps, but that decision also allowed us to shape the synchronization differently when it was originated from a Geo node, vs a regular request. Internally we store additional metadata in the repository and also commits that may no longer exist in your regular branches, but were part of a previous merge request, or had user comments associated with them.\n\nBy also switching to full HTTP(S), it made it simpler to run our development instances locally with [GDK](https://gitlab.com/gitlab-org/gitlab-development-kit), which helped to improve our own internal development process as well.\n\n## Phase 4: Third-generation synchronization and the path to a Disaster Recovery solution\n\nWhile still working in Phase 3, we discovered another major limitation around how we stored files on disk. GitLab, for historical reasons, stored repositories and file attachments in a similar disk structure as the base URL routes. For group and project `gitlab-org/gitlab-ce` there would be a path on disk that would include `gitlab-org/gitlab-ce` as part of it. The same is true for file attachments.\n\nKeeping both the database and disk in sync, even not considering Geo replication, means that at any time a project is renamed, several things have to be renamed on disk as well.\n\nThis is not only slow and error prone: what should we do if something fails to rename in the middle of the \"transaction?\" This is also problematic when replication comes into place as we are susceptible again to processing it in the correct order or risk a temporarily inconsistent state.\n\nWe tried to find a solution to problems around the order of execution of the events and we came up with three ideas:\n\n1. **Find or build a queue system that is guaranteed to process things in the same order they were scheduled**\n2. **Detect and recover from any replication failure or data corruption**\n3. **Make every replication operation idempotent, removing the queue-ordering requirement completely**\n\nThe first one was easily ruled out, as even if we switched to a queue system with that type of guarantee, it would be either slow due to the lack of parallelism in order to guarantee the order requirement, or will be extremely complex and hard to use as it would require extra care to have the same guarantees while also working in parallel.\n\nWe found no system that satisfied our needs, and even if we considered a standalone queue solution, we would lose the Postgres advantage from the previous generation, of having both the main database and the queue system always in sync.\n\nRuling out the first one, we considered the second idea of detecting and recovering from failures and data corruption as we concluded we needed it for *Disaster Recovery* anyway. Any robust *Disaster Recovery* solution needs to guarantee that the data it is holding is the exact one it's supposed to have. If, for any reason, that data gets corrupt or someone removes it from disk, it needs a way of detecting it and restores it to the desired state.\n\nTo achieve that, we built a robust verification mechanism that generates a checksum of the state of the repository and is stored in a separate table in the primary node. That table gets replicated to secondary nodes, where another checksum is also calculated (and stored in the Tracking Database). If both checksums match, we know the data is consistent. The checksum is recalculated automatically when an update event is processed, but can also be triggered manually.\n\n![Screen Capture - Repository Verification Status](https://about.gitlab.com/images/blogimages/how-we-built-geo/verification-status-primary.png){: .medium.center}\n\nWe used that mechanism to validate all repositories in `gitlab.com` when successfully [migrating from Azure to GCP](/blog/gcp-move-update/), last month.\n\nThe verification mechanism is not enough and while it gives us the guarantees we need, we can do better, which is why we also decided to implement the third idea as well, and make every replication operation idempotent in order to remove any situation where processing the incorrect order of events would put data in a temporarily inconsistent state.\n\nWe are calling that solution the [Hashed Storage](https://docs.gitlab.com/ee/administration/repository_storage_types.html). This is a complete rewrite of how GitLab stores files on disk. Instead of reusing the same paths as present in the URLs, we use the internal IDs to create a hash instead and derive the disk path from that hash. With the Hashed Storage, renaming a project or moving it to a new group requires only the database operations to be persisted, as the location on disk never changes.\n\n![Hashed Storage and Legacy Storage example](https://about.gitlab.com/images/blogimages/how-we-built-geo/hashed-storage-disk-path-example.png){: .medium.center}\n\nBy making the paths on disk immutable and non-conflicting, any `create`, `move` or `remove` operations can happen in any order, and they will never put the system in an inconsistent state. Also replicating a project rename or moving a project from one group/owner to another will require only the database change to be propagated to take full effect on a secondary node.\n\n## What to expect from Geo in the near future\n\nImplementing Geo has been an important effort at GitLab that involved many different areas. It is a crucial infrastructure feature that allowed us to migrate from one cloud provider to another. We also believe it's an important component to support the needs of many organizations today, from providing peace of mind regarding data safety in the events of a Disaster Recovery, to easing the burdens of distributed teams across the globe.\n\nWe've been using the feature ourselves and this allowed us to stress-test the biggest and most challenging GitLab installation, GitLab.com, making sure it will work just as fine for any other customer.\n\nOver the upcoming months we will be focusing on the following items:\n\n* Release a push proxy for Geo secondary nodes: [Pull and push from the same remote transparently](https://gitlab.com/groups/gitlab-org/-/epics/124)\n* Release [Hashed Storage as *Generally Available*](https://gitlab.com/groups/gitlab-org/-/epics/75)\n* Improve configuration: We want to reduce the steps and make it [simpler via automating most steps](https://gitlab.com/groups/gitlab-org/-/epics/367)\n* Improve the verification step: [Improve the signals we use for the checksum](https://gitlab.com/gitlab-org/gitlab-ee/issues/5196)\n* [Improve the Geo UX and UI](https://gitlab.com/groups/gitlab-org/-/epics/369)\n* Keep improving performance and reliability\n* Support replication of [GitLab Pages](https://gitlab.com/gitlab-org/gitlab-ee/issues/4611) and the internal [Docker Registry](https://gitlab.com/gitlab-org/gitlab-ee/issues/2870)\n\nCover photo by [NASA](https://unsplash.com/photos/Q1p7bh3SHj8) on [Unsplash](https://unsplash.com/)\n",[23,24],"features","inside GitLab","yml",{},true,"/en-us/blog/how-we-built-gitlab-geo",{"title":15,"description":16,"ogTitle":15,"ogDescription":16,"noIndex":12,"ogImage":19,"ogUrl":30,"ogSiteName":31,"ogType":32,"canonicalUrls":30},"https://about.gitlab.com/blog/how-we-built-gitlab-geo","https://about.gitlab.com","article","en-us/blog/how-we-built-gitlab-geo",[23,35],"inside-gitlab","EEvIpyqGZ4Rg_WzmLYb86AtNC7W_4DRSZhl0q7veZX0",{"data":38},{"logo":39,"freeTrial":44,"sales":49,"login":54,"items":59,"search":367,"minimal":398,"duo":417,"pricingDeployment":427},{"config":40},{"href":41,"dataGaName":42,"dataGaLocation":43},"/","gitlab logo","header",{"text":45,"config":46},"Get free trial",{"href":47,"dataGaName":48,"dataGaLocation":43},"https://gitlab.com/-/trial_registrations/new?glm_source=about.gitlab.com&glm_content=default-saas-trial/","free trial",{"text":50,"config":51},"Talk to sales",{"href":52,"dataGaName":53,"dataGaLocation":43},"/sales/","sales",{"text":55,"config":56},"Sign in",{"href":57,"dataGaName":58,"dataGaLocation":43},"https://gitlab.com/users/sign_in/","sign in",[60,87,182,187,288,348],{"text":61,"config":62,"cards":64},"Platform",{"dataNavLevelOne":63},"platform",[65,71,79],{"title":61,"description":66,"link":67},"The intelligent orchestration platform for DevSecOps",{"text":68,"config":69},"Explore our Platform",{"href":70,"dataGaName":63,"dataGaLocation":43},"/platform/",{"title":72,"description":73,"link":74},"GitLab Duo Agent Platform","Agentic AI for the entire software lifecycle",{"text":75,"config":76},"Meet GitLab Duo",{"href":77,"dataGaName":78,"dataGaLocation":43},"/gitlab-duo-agent-platform/","gitlab duo agent platform",{"title":80,"description":81,"link":82},"Why GitLab","See the top reasons enterprises choose GitLab",{"text":83,"config":84},"Learn more",{"href":85,"dataGaName":86,"dataGaLocation":43},"/why-gitlab/","why gitlab",{"text":88,"left":27,"config":89,"link":91,"lists":95,"footer":164},"Product",{"dataNavLevelOne":90},"solutions",{"text":92,"config":93},"View all Solutions",{"href":94,"dataGaName":90,"dataGaLocation":43},"/solutions/",[96,120,143],{"title":97,"description":98,"link":99,"items":104},"Automation","CI/CD and automation to accelerate deployment",{"config":100},{"icon":101,"href":102,"dataGaName":103,"dataGaLocation":43},"AutomatedCodeAlt","/solutions/delivery-automation/","automated software delivery",[105,109,112,116],{"text":106,"config":107},"CI/CD",{"href":108,"dataGaLocation":43,"dataGaName":106},"/solutions/continuous-integration/",{"text":72,"config":110},{"href":77,"dataGaLocation":43,"dataGaName":111},"gitlab duo agent platform - product menu",{"text":113,"config":114},"Source Code Management",{"href":115,"dataGaLocation":43,"dataGaName":113},"/solutions/source-code-management/",{"text":117,"config":118},"Automated Software Delivery",{"href":102,"dataGaLocation":43,"dataGaName":119},"Automated software delivery",{"title":121,"description":122,"link":123,"items":128},"Security","Deliver code faster without compromising security",{"config":124},{"href":125,"dataGaName":126,"dataGaLocation":43,"icon":127},"/solutions/application-security-testing/","security and compliance","ShieldCheckLight",[129,133,138],{"text":130,"config":131},"Application Security Testing",{"href":125,"dataGaName":132,"dataGaLocation":43},"Application security testing",{"text":134,"config":135},"Software Supply Chain Security",{"href":136,"dataGaLocation":43,"dataGaName":137},"/solutions/supply-chain/","Software supply chain security",{"text":139,"config":140},"Software Compliance",{"href":141,"dataGaName":142,"dataGaLocation":43},"/solutions/software-compliance/","software compliance",{"title":144,"link":145,"items":150},"Measurement",{"config":146},{"icon":147,"href":148,"dataGaName":149,"dataGaLocation":43},"DigitalTransformation","/solutions/visibility-measurement/","visibility and measurement",[151,155,159],{"text":152,"config":153},"Visibility & Measurement",{"href":148,"dataGaLocation":43,"dataGaName":154},"Visibility and Measurement",{"text":156,"config":157},"Value Stream Management",{"href":158,"dataGaLocation":43,"dataGaName":156},"/solutions/value-stream-management/",{"text":160,"config":161},"Analytics & Insights",{"href":162,"dataGaLocation":43,"dataGaName":163},"/solutions/analytics-and-insights/","Analytics and insights",{"title":165,"items":166},"GitLab for",[167,172,177],{"text":168,"config":169},"Enterprise",{"href":170,"dataGaLocation":43,"dataGaName":171},"/enterprise/","enterprise",{"text":173,"config":174},"Small Business",{"href":175,"dataGaLocation":43,"dataGaName":176},"/small-business/","small business",{"text":178,"config":179},"Public Sector",{"href":180,"dataGaLocation":43,"dataGaName":181},"/solutions/public-sector/","public sector",{"text":183,"config":184},"Pricing",{"href":185,"dataGaName":186,"dataGaLocation":43,"dataNavLevelOne":186},"/pricing/","pricing",{"text":188,"config":189,"link":191,"lists":195,"feature":275},"Resources",{"dataNavLevelOne":190},"resources",{"text":192,"config":193},"View all resources",{"href":194,"dataGaName":190,"dataGaLocation":43},"/resources/",[196,229,247],{"title":197,"items":198},"Getting started",[199,204,209,214,219,224],{"text":200,"config":201},"Install",{"href":202,"dataGaName":203,"dataGaLocation":43},"/install/","install",{"text":205,"config":206},"Quick start guides",{"href":207,"dataGaName":208,"dataGaLocation":43},"/get-started/","quick setup checklists",{"text":210,"config":211},"Learn",{"href":212,"dataGaLocation":43,"dataGaName":213},"https://university.gitlab.com/","learn",{"text":215,"config":216},"Product documentation",{"href":217,"dataGaName":218,"dataGaLocation":43},"https://docs.gitlab.com/","product documentation",{"text":220,"config":221},"Best practice videos",{"href":222,"dataGaName":223,"dataGaLocation":43},"/getting-started-videos/","best practice videos",{"text":225,"config":226},"Integrations",{"href":227,"dataGaName":228,"dataGaLocation":43},"/integrations/","integrations",{"title":230,"items":231},"Discover",[232,237,242],{"text":233,"config":234},"Customer success stories",{"href":235,"dataGaName":236,"dataGaLocation":43},"/customers/","customer success stories",{"text":238,"config":239},"Blog",{"href":240,"dataGaName":241,"dataGaLocation":43},"/blog/","blog",{"text":243,"config":244},"Remote",{"href":245,"dataGaName":246,"dataGaLocation":43},"https://handbook.gitlab.com/handbook/company/culture/all-remote/","remote",{"title":248,"items":249},"Connect",[250,255,260,265,270],{"text":251,"config":252},"GitLab Services",{"href":253,"dataGaName":254,"dataGaLocation":43},"/services/","services",{"text":256,"config":257},"Community",{"href":258,"dataGaName":259,"dataGaLocation":43},"/community/","community",{"text":261,"config":262},"Forum",{"href":263,"dataGaName":264,"dataGaLocation":43},"https://forum.gitlab.com/","forum",{"text":266,"config":267},"Events",{"href":268,"dataGaName":269,"dataGaLocation":43},"/events/","events",{"text":271,"config":272},"Partners",{"href":273,"dataGaName":274,"dataGaLocation":43},"/partners/","partners",{"backgroundColor":276,"textColor":277,"text":278,"image":279,"link":283},"#2f2a6b","#fff","Insights for the future of software development",{"altText":280,"config":281},"the source promo card",{"src":282},"https://res.cloudinary.com/about-gitlab-com/image/upload/v1758208064/dzl0dbift9xdizyelkk4.svg",{"text":284,"config":285},"Read the latest",{"href":286,"dataGaName":287,"dataGaLocation":43},"/the-source/","the source",{"text":289,"config":290,"lists":292},"Company",{"dataNavLevelOne":291},"company",[293],{"items":294},[295,300,306,308,313,318,323,328,333,338,343],{"text":296,"config":297},"About",{"href":298,"dataGaName":299,"dataGaLocation":43},"/company/","about",{"text":301,"config":302,"footerGa":305},"Jobs",{"href":303,"dataGaName":304,"dataGaLocation":43},"/jobs/","jobs",{"dataGaName":304},{"text":266,"config":307},{"href":268,"dataGaName":269,"dataGaLocation":43},{"text":309,"config":310},"Leadership",{"href":311,"dataGaName":312,"dataGaLocation":43},"/company/team/e-group/","leadership",{"text":314,"config":315},"Team",{"href":316,"dataGaName":317,"dataGaLocation":43},"/company/team/","team",{"text":319,"config":320},"Handbook",{"href":321,"dataGaName":322,"dataGaLocation":43},"https://handbook.gitlab.com/","handbook",{"text":324,"config":325},"Investor relations",{"href":326,"dataGaName":327,"dataGaLocation":43},"https://ir.gitlab.com/","investor relations",{"text":329,"config":330},"Trust Center",{"href":331,"dataGaName":332,"dataGaLocation":43},"/security/","trust center",{"text":334,"config":335},"AI Transparency Center",{"href":336,"dataGaName":337,"dataGaLocation":43},"/ai-transparency-center/","ai transparency center",{"text":339,"config":340},"Newsletter",{"href":341,"dataGaName":342,"dataGaLocation":43},"/company/contact/#contact-forms","newsletter",{"text":344,"config":345},"Press",{"href":346,"dataGaName":347,"dataGaLocation":43},"/press/","press",{"text":349,"config":350,"lists":351},"Contact us",{"dataNavLevelOne":291},[352],{"items":353},[354,357,362],{"text":50,"config":355},{"href":52,"dataGaName":356,"dataGaLocation":43},"talk to sales",{"text":358,"config":359},"Support portal",{"href":360,"dataGaName":361,"dataGaLocation":43},"https://support.gitlab.com","support portal",{"text":363,"config":364},"Customer portal",{"href":365,"dataGaName":366,"dataGaLocation":43},"https://customers.gitlab.com/customers/sign_in/","customer portal",{"close":368,"login":369,"suggestions":376},"Close",{"text":370,"link":371},"To search repositories and projects, login to",{"text":372,"config":373},"gitlab.com",{"href":57,"dataGaName":374,"dataGaLocation":375},"search login","search",{"text":377,"default":378},"Suggestions",[379,381,385,387,391,395],{"text":72,"config":380},{"href":77,"dataGaName":72,"dataGaLocation":375},{"text":382,"config":383},"Code Suggestions (AI)",{"href":384,"dataGaName":382,"dataGaLocation":375},"/solutions/code-suggestions/",{"text":106,"config":386},{"href":108,"dataGaName":106,"dataGaLocation":375},{"text":388,"config":389},"GitLab on AWS",{"href":390,"dataGaName":388,"dataGaLocation":375},"/partners/technology-partners/aws/",{"text":392,"config":393},"GitLab on Google Cloud",{"href":394,"dataGaName":392,"dataGaLocation":375},"/partners/technology-partners/google-cloud-platform/",{"text":396,"config":397},"Why GitLab?",{"href":85,"dataGaName":396,"dataGaLocation":375},{"freeTrial":399,"mobileIcon":404,"desktopIcon":409,"secondaryButton":412},{"text":400,"config":401},"Start free trial",{"href":402,"dataGaName":48,"dataGaLocation":403},"https://gitlab.com/-/trials/new/","nav",{"altText":405,"config":406},"Gitlab Icon",{"src":407,"dataGaName":408,"dataGaLocation":403},"https://res.cloudinary.com/about-gitlab-com/image/upload/v1758203874/jypbw1jx72aexsoohd7x.svg","gitlab icon",{"altText":405,"config":410},{"src":411,"dataGaName":408,"dataGaLocation":403},"https://res.cloudinary.com/about-gitlab-com/image/upload/v1758203875/gs4c8p8opsgvflgkswz9.svg",{"text":413,"config":414},"Get Started",{"href":415,"dataGaName":416,"dataGaLocation":403},"https://gitlab.com/-/trial_registrations/new?glm_source=about.gitlab.com/compare/gitlab-vs-github/","get started",{"freeTrial":418,"mobileIcon":423,"desktopIcon":425},{"text":419,"config":420},"Learn more about GitLab Duo",{"href":421,"dataGaName":422,"dataGaLocation":403},"/gitlab-duo/","gitlab duo",{"altText":405,"config":424},{"src":407,"dataGaName":408,"dataGaLocation":403},{"altText":405,"config":426},{"src":411,"dataGaName":408,"dataGaLocation":403},{"freeTrial":428,"mobileIcon":433,"desktopIcon":435},{"text":429,"config":430},"Back to pricing",{"href":185,"dataGaName":431,"dataGaLocation":403,"icon":432},"back to pricing","GoBack",{"altText":405,"config":434},{"src":407,"dataGaName":408,"dataGaLocation":403},{"altText":405,"config":436},{"src":411,"dataGaName":408,"dataGaLocation":403},{"title":438,"button":439,"config":444},"See how agentic AI transforms software delivery",{"text":440,"config":441},"Watch GitLab Transcend now",{"href":442,"dataGaName":443,"dataGaLocation":43},"/events/transcend/virtual/","transcend event",{"layout":445,"icon":446},"release","AiStar",{"data":448},{"text":449,"source":450,"edit":456,"contribute":461,"config":466,"items":471,"minimal":678},"Git is a trademark of Software Freedom Conservancy and our use of 'GitLab' is under license",{"text":451,"config":452},"View page source",{"href":453,"dataGaName":454,"dataGaLocation":455},"https://gitlab.com/gitlab-com/marketing/digital-experience/about-gitlab-com/","page source","footer",{"text":457,"config":458},"Edit this page",{"href":459,"dataGaName":460,"dataGaLocation":455},"https://gitlab.com/gitlab-com/marketing/digital-experience/about-gitlab-com/-/blob/main/content/","web ide",{"text":462,"config":463},"Please contribute",{"href":464,"dataGaName":465,"dataGaLocation":455},"https://gitlab.com/gitlab-com/marketing/digital-experience/about-gitlab-com/-/blob/main/CONTRIBUTING.md/","please contribute",{"twitter":467,"facebook":468,"youtube":469,"linkedin":470},"https://twitter.com/gitlab","https://www.facebook.com/gitlab","https://www.youtube.com/channel/UCnMGQ8QHMAnVIsI3xJrihhg","https://www.linkedin.com/company/gitlab-com",[472,519,573,617,644],{"title":183,"links":473,"subMenu":488},[474,478,483],{"text":475,"config":476},"View plans",{"href":185,"dataGaName":477,"dataGaLocation":455},"view plans",{"text":479,"config":480},"Why Premium?",{"href":481,"dataGaName":482,"dataGaLocation":455},"/pricing/premium/","why premium",{"text":484,"config":485},"Why Ultimate?",{"href":486,"dataGaName":487,"dataGaLocation":455},"/pricing/ultimate/","why ultimate",[489],{"title":490,"links":491},"Contact Us",[492,495,497,499,504,509,514],{"text":493,"config":494},"Contact sales",{"href":52,"dataGaName":53,"dataGaLocation":455},{"text":358,"config":496},{"href":360,"dataGaName":361,"dataGaLocation":455},{"text":363,"config":498},{"href":365,"dataGaName":366,"dataGaLocation":455},{"text":500,"config":501},"Status",{"href":502,"dataGaName":503,"dataGaLocation":455},"https://status.gitlab.com/","status",{"text":505,"config":506},"Terms of use",{"href":507,"dataGaName":508,"dataGaLocation":455},"/terms/","terms of use",{"text":510,"config":511},"Privacy statement",{"href":512,"dataGaName":513,"dataGaLocation":455},"/privacy/","privacy statement",{"text":515,"config":516},"Cookie preferences",{"dataGaName":517,"dataGaLocation":455,"id":518,"isOneTrustButton":27},"cookie preferences","ot-sdk-btn",{"title":88,"links":520,"subMenu":529},[521,525],{"text":522,"config":523},"DevSecOps platform",{"href":70,"dataGaName":524,"dataGaLocation":455},"devsecops platform",{"text":526,"config":527},"AI-Assisted Development",{"href":421,"dataGaName":528,"dataGaLocation":455},"ai-assisted development",[530],{"title":531,"links":532},"Topics",[533,538,543,548,553,558,563,568],{"text":534,"config":535},"CICD",{"href":536,"dataGaName":537,"dataGaLocation":455},"/topics/ci-cd/","cicd",{"text":539,"config":540},"GitOps",{"href":541,"dataGaName":542,"dataGaLocation":455},"/topics/gitops/","gitops",{"text":544,"config":545},"DevOps",{"href":546,"dataGaName":547,"dataGaLocation":455},"/topics/devops/","devops",{"text":549,"config":550},"Version Control",{"href":551,"dataGaName":552,"dataGaLocation":455},"/topics/version-control/","version control",{"text":554,"config":555},"DevSecOps",{"href":556,"dataGaName":557,"dataGaLocation":455},"/topics/devsecops/","devsecops",{"text":559,"config":560},"Cloud Native",{"href":561,"dataGaName":562,"dataGaLocation":455},"/topics/cloud-native/","cloud native",{"text":564,"config":565},"AI for Coding",{"href":566,"dataGaName":567,"dataGaLocation":455},"/topics/devops/ai-for-coding/","ai for coding",{"text":569,"config":570},"Agentic AI",{"href":571,"dataGaName":572,"dataGaLocation":455},"/topics/agentic-ai/","agentic ai",{"title":574,"links":575},"Solutions",[576,578,580,585,589,592,596,599,601,604,607,612],{"text":130,"config":577},{"href":125,"dataGaName":130,"dataGaLocation":455},{"text":119,"config":579},{"href":102,"dataGaName":103,"dataGaLocation":455},{"text":581,"config":582},"Agile development",{"href":583,"dataGaName":584,"dataGaLocation":455},"/solutions/agile-delivery/","agile delivery",{"text":586,"config":587},"SCM",{"href":115,"dataGaName":588,"dataGaLocation":455},"source code management",{"text":534,"config":590},{"href":108,"dataGaName":591,"dataGaLocation":455},"continuous integration & delivery",{"text":593,"config":594},"Value stream management",{"href":158,"dataGaName":595,"dataGaLocation":455},"value stream management",{"text":539,"config":597},{"href":598,"dataGaName":542,"dataGaLocation":455},"/solutions/gitops/",{"text":168,"config":600},{"href":170,"dataGaName":171,"dataGaLocation":455},{"text":602,"config":603},"Small business",{"href":175,"dataGaName":176,"dataGaLocation":455},{"text":605,"config":606},"Public sector",{"href":180,"dataGaName":181,"dataGaLocation":455},{"text":608,"config":609},"Education",{"href":610,"dataGaName":611,"dataGaLocation":455},"/solutions/education/","education",{"text":613,"config":614},"Financial services",{"href":615,"dataGaName":616,"dataGaLocation":455},"/solutions/finance/","financial services",{"title":188,"links":618},[619,621,623,625,628,630,632,634,636,638,640,642],{"text":200,"config":620},{"href":202,"dataGaName":203,"dataGaLocation":455},{"text":205,"config":622},{"href":207,"dataGaName":208,"dataGaLocation":455},{"text":210,"config":624},{"href":212,"dataGaName":213,"dataGaLocation":455},{"text":215,"config":626},{"href":217,"dataGaName":627,"dataGaLocation":455},"docs",{"text":238,"config":629},{"href":240,"dataGaName":241,"dataGaLocation":455},{"text":233,"config":631},{"href":235,"dataGaName":236,"dataGaLocation":455},{"text":243,"config":633},{"href":245,"dataGaName":246,"dataGaLocation":455},{"text":251,"config":635},{"href":253,"dataGaName":254,"dataGaLocation":455},{"text":256,"config":637},{"href":258,"dataGaName":259,"dataGaLocation":455},{"text":261,"config":639},{"href":263,"dataGaName":264,"dataGaLocation":455},{"text":266,"config":641},{"href":268,"dataGaName":269,"dataGaLocation":455},{"text":271,"config":643},{"href":273,"dataGaName":274,"dataGaLocation":455},{"title":289,"links":645},[646,648,650,652,654,656,658,662,667,669,671,673],{"text":296,"config":647},{"href":298,"dataGaName":291,"dataGaLocation":455},{"text":301,"config":649},{"href":303,"dataGaName":304,"dataGaLocation":455},{"text":309,"config":651},{"href":311,"dataGaName":312,"dataGaLocation":455},{"text":314,"config":653},{"href":316,"dataGaName":317,"dataGaLocation":455},{"text":319,"config":655},{"href":321,"dataGaName":322,"dataGaLocation":455},{"text":324,"config":657},{"href":326,"dataGaName":327,"dataGaLocation":455},{"text":659,"config":660},"Sustainability",{"href":661,"dataGaName":659,"dataGaLocation":455},"/sustainability/",{"text":663,"config":664},"Diversity, inclusion and belonging (DIB)",{"href":665,"dataGaName":666,"dataGaLocation":455},"/diversity-inclusion-belonging/","Diversity, inclusion and belonging",{"text":329,"config":668},{"href":331,"dataGaName":332,"dataGaLocation":455},{"text":339,"config":670},{"href":341,"dataGaName":342,"dataGaLocation":455},{"text":344,"config":672},{"href":346,"dataGaName":347,"dataGaLocation":455},{"text":674,"config":675},"Modern Slavery Transparency Statement",{"href":676,"dataGaName":677,"dataGaLocation":455},"https://handbook.gitlab.com/handbook/legal/modern-slavery-act-transparency-statement/","modern slavery transparency statement",{"items":679},[680,683,686],{"text":681,"config":682},"Terms",{"href":507,"dataGaName":508,"dataGaLocation":455},{"text":684,"config":685},"Cookies",{"dataGaName":517,"dataGaLocation":455,"id":518,"isOneTrustButton":27},{"text":687,"config":688},"Privacy",{"href":512,"dataGaName":513,"dataGaLocation":455},[690],{"id":691,"title":18,"body":8,"config":692,"content":694,"description":8,"extension":25,"meta":698,"navigation":27,"path":699,"seo":700,"stem":701,"__hash__":702},"blogAuthors/en-us/blog/authors/gabriel-mazetto.yml",{"template":693},"BlogAuthor",{"name":18,"config":695},{"headshot":696,"ctfId":697},"https://res.cloudinary.com/about-gitlab-com/image/upload/v1749678982/Blog/Author%20Headshots/brodock-headshot.jpg","brodock",{},"/en-us/blog/authors/gabriel-mazetto",{},"en-us/blog/authors/gabriel-mazetto","elgZFaYSswD9MZUjesPB6vCwvmu4bay0z__KaEaQWI8",[704,717,729],{"content":705,"config":715},{"title":706,"description":707,"authors":708,"heroImage":710,"date":711,"category":9,"tags":712,"body":714},"How IIT Bombay students are coding the future with GitLab","At GitLab, we often talk about how software accelerates innovation. But sometimes, you have to step away from the Zoom calls and stand in a crowded university hall to remember why we do this.",[709],"Nick Veenhof","https://res.cloudinary.com/about-gitlab-com/image/upload/v1750099013/Blog/Hero%20Images/Blog/Hero%20Images/blog-image-template-1800x945%20%2814%29_6VTUA8mUhOZNDaRVNPeKwl_1750099012960.png","2026-01-08",[259,611,713],"open source","The GitLab team recently had the privilege of judging the **iHack Hackathon** at **IIT Bombay's E-Summit**. The energy was electric, the coffee was flowing, and the talent was undeniable. But what struck us most wasn't just the code — it was the sheer determination of students to solve real-world problems, often overcoming significant logistical and financial hurdles to simply be in the room.\n\n\nThrough our [GitLab for Education program](https://about.gitlab.com/solutions/education/), we aim to empower the next generation of developers with tools and opportunity. Here is a look at what the students built, and how they used GitLab to bridge the gap between idea and reality.\n\n## The challenge: Build faster, build securely\n\nThe premise for the GitLab track of the hackathon was simple: Don't just show us a product; show us how you built it. We wanted to see how students utilized GitLab's platform — from Issue Boards to CI/CD pipelines — to accelerate the development lifecycle.\n\nThe results were inspiring.\n\n## The winners\n\n### 1st place: Team Decode — Democratizing Scientific Research\n\n**Project:** FIRE (Fast Integrated Research Environment)\n\nTeam Decode took home the top prize with a solution that warms a developer's heart: a local-first, blazing-fast data processing tool built with [Rust](https://about.gitlab.com/blog/secure-rust-development-with-gitlab/) and Tauri. They identified a massive pain point for data science students: existing tools are fragmented, slow, and expensive.\n\nTheir solution, FIRE, allows researchers to visualize complex formats (like NetCDF) instantly. What impressed the judges most was their \"hacker\" ethos. They didn't just build a tool; they built it to be open and accessible.\n\n**How they used GitLab:** Since the team lived far apart, asynchronous communication was key. They utilized **GitLab Issue Boards** and **Milestones** to track progress and integrated their repo with Telegram to get real-time push notifications. As one team member noted, \"Coordinating all these technologies was really difficult, and what helped us was GitLab... the Issue Board really helped us track who was doing what.\"\n\n![Team Decode](https://res.cloudinary.com/about-gitlab-com/image/upload/v1767380253/epqazj1jc5c7zkgqun9h.jpg)\n\n### 2nd place: Team BichdeHueDost — Reuniting to Solve Payments\n\n**Project:** SemiPay (RFID Cashless Payment for Schools)\n\nThe team name, BichdeHueDost, translates to \"Friends who have been set apart.\" It's a fitting name for a group of friends who went to different colleges but reunited to build this project. They tackled a unique problem: handling cash in schools for young children. Their solution used RFID cards backed by a blockchain ledger to ensure secure, cashless transactions for students.\n\n**How they used GitLab:** They utilized [GitLab CI/CD](https://about.gitlab.com/topics/ci-cd/) to automate the build process for their Flutter application (APK), ensuring that every commit resulted in a testable artifact. This allowed them to iterate quickly despite the \"flaky\" nature of cross-platform mobile development.\n\n![Team BichdeHueDost](https://res.cloudinary.com/about-gitlab-com/image/upload/v1767380253/pkukrjgx2miukb6nrj5g.jpg)\n\n### 3rd place: Team ZenYukti — Agentic Repository Intelligence\n\n**Project:** RepoInsight AI (AI-powered, GitLab-native intelligence platform)\n\nTeam ZenYukti impressed us with a solution that tackles a universal developer pain point: understanding unfamiliar codebases. What stood out to the judges was the tool's practical approach to onboarding and code comprehension: RepoInsight-AI automatically generates documentation, visualizes repository structure, and even helps identify bugs, all while maintaining context about the entire codebase.\n\n**How they used GitLab:** The team built a comprehensive CI/CD pipeline that showcased GitLab's security and DevOps capabilities. They integrated [GitLab's Security Templates](https://gitlab.com/gitlab-org/gitlab/-/tree/master/lib/gitlab/ci/templates/Security) (SAST, Dependency Scanning, and Secret Detection), and utilized [GitLab Container Registry](https://docs.gitlab.com/user/packages/container_registry/) to manage their Docker images for backend and frontend components. They created an AI auto-review bot that runs on merge requests, demonstrating an \"agentic workflow\" where AI assists in the development process itself.\n\n![Team ZenYukti](https://res.cloudinary.com/about-gitlab-com/image/upload/v1767380253/ymlzqoruv5al1secatba.jpg)\n\n## Beyond the code: A lesson in inclusion\n\nWhile the code was impressive, the most powerful moment of the event happened away from the keyboard.\n\nDuring the feedback session, we learned about the journey Team ZenYukti took to get to Mumbai. They traveled over 24 hours, covering nearly 1,800 kilometers. Because flights were too expensive and trains were booked, they traveled in the \"General Coach,\" a non-reserved, severely overcrowded carriage.\n\nAs one student described it:\n\n*\"You cannot even imagine something like this... there are no seats... people sit on the top of the train. This is what we have endured.\"*\n\nThis hit home. [Diversity, Inclusion, and Belonging](https://handbook.gitlab.com/handbook/company/culture/inclusion/) are core values at GitLab. We realized that for these students, the barrier to entry wasn't intellect or skill, it was access.\n\nIn that moment, we decided to break that barrier. We committed to reimbursing the travel expenses for the participants who struggled to get there. It's a small step, but it underlines a massive truth: **talent is distributed equally, but opportunity is not.**\n\n![hackathon class together](https://res.cloudinary.com/about-gitlab-com/image/upload/v1767380252/o5aqmboquz8ehusxvgom.jpg)\n\n### The future is bright (and automated)\n\nWe also saw incredible potential in teams like Prometheus, who attempted to build an autonomous patch remediation tool (DevGuardian), and Team Arrakis, who built a voice-first job portal for blue-collar workers using [GitLab Duo](https://about.gitlab.com/gitlab-duo/) to troubleshoot their pipelines.\n\nTo all the students who participated: You are the future. Through [GitLab for Education](https://about.gitlab.com/solutions/education/), we are committed to providing you with the top-tier tools (like GitLab Ultimate) you need to learn, collaborate, and change the world — whether you are coding from a dorm room, a lab, or a train carriage. **Keep shipping.**\n\n> :bulb: Learn more about the [GitLab for Education program](https://about.gitlab.com/solutions/education/).\n",{"slug":716,"featured":12,"template":13},"how-iit-bombay-students-code-future-with-gitlab",{"content":718,"config":727},{"title":719,"description":720,"authors":721,"heroImage":722,"date":723,"category":9,"tags":724,"body":726},"Artois University elevates research and curriculum with GitLab Ultimate for Education","Artois University's CRIL leveraged the GitLab for Education program to gain free access to Ultimate, transforming advanced research and computer science curricula.",[709],"https://res.cloudinary.com/about-gitlab-com/image/upload/v1750099203/Blog/Hero%20Images/Blog/Hero%20Images/blog-image-template-1800x945%20%2820%29_2bJGC5ZP3WheoqzlLT05C5_1750099203484.png","2025-12-10",[611,259,725],"product","Leading academic institutions face a critical challenge: how to provide thousands of students and researchers with industry-standard, **full-featured DevSecOps tools** without compromising institutional control. Many start with basic version control, but the modern curriculum demands integrated capabilities for planning, security, and advanced CI/CD.\n\nThe **GitLab for Education program** is designed to solve this by providing access to **GitLab Ultimate** for qualifying institutions, allowing them to scale their operations and elevate their academic offerings. \n\nThis article showcases a powerful success story from the **Centre de Recherche en Informatique de Lens (CRIL)**, a joint laboratory of **Artois University** and CNRS in France. After years of relying solely on GitLab Community Edition (CE), the university's move to GitLab Ultimate through the GitLab for Education program immediately unlocked advanced capabilities, transforming their teaching, research, and contribution workflows virtually overnight. This story demonstrates why GitLab Ultimate is essential for institutions seeking to deliver advanced computer science and research curricula.\n\n## GitLab Ultimate unlocked: Managing scale and driving academic value\n\n**Artois University's** self-managed GitLab instance is a large-scale operation, supporting nearly **3,000 users** across approximately **19,000 projects**, primarily serving computer science students and researchers. While GitLab Community Edition was robust, the upgrade to GitLab Ultimate provided the sophisticated tooling necessary for managing this scale and facilitating advanced university-level work.\n\n***\"We can see the difference,\" says Daniel Le Berre, head of research at CRIL and the instance maintainer. \"It's a completely different product. Each week reveals new features that directly enhance our productivity and teaching.\"***\n\nThe institution joined the GitLab for Education program specifically because it covers both **instructional and non-commercial research use cases** and offers full access to Ultimate's features, removing significant cost barriers.\n\n### Key GitLab Ultimate benefits for students and researchers\n\n* **Advanced project management at scale:** Master's students now benefit from **GitLab Ultimate's project planning features**. This enables them to structure, track, and manage complex, long-term research projects using professional methodologies like portfolio management and advanced issue tracking that seamlessly roll up across their thousands of projects.\n\n* **Enhanced visibility:** Features like improved dashboards and code previews directly in Markdown files dramatically streamline tracking and documentation review, reducing administrative friction for both instructors and students managing large project loads.\n\n## Comprehensive curriculum: From concepts to continuous delivery\n\nGitLab Ultimate is deeply integrated into the computer science curriculum, moving students beyond simple `git` commands to practical **DevSecOps implementation**.\n\n* **Git fundamentals:** Students begin by visualizing concepts using open-source tools to master Git concepts.\n\n* **Full CI/CD implementation:** Students use GitLab CI for rigorous **Test-Driven Development (TDD)** in their software projects. They learn to build, test, and perform quality assurance using unit and integration testing pipelines—core competency made seamless by the integrated platform.\n\n* **DevSecOps for research and documentation:** The university teaches students that DevSecOps principles are vital for all collaborative work. Inspired by earlier work in Delft, students manage and produce critical research documentation (PDFs from Markdown files) using GitLab, incorporating quality checks like linters and spell checks directly in the CI pipeline. This ensures high-quality, reproducible research output.\n\n* **Future-proofing security skills:** The GitLab Ultimate platform immediately positions the institution to incorporate advanced DevSecOps features like SAST and DAST scanning as their research and development code projects grow, ensuring students are prepared for industry security standards.\n\n## Accelerating open source contributions with GitLab Duo\n\nAccess to the full GitLab platform, including our AI capabilities, has empowered students to make impactful contributions to the wider open source community faster than ever before.\n\nTwo Master's students recently completed direct contributions to the GitLab product, adding the **ORCID identifier** into user profiles. Working on GitLab.com, they leveraged **GitLab Duo's AI chat and code suggestions** to navigate the codebase efficiently.\n\n***\"This would not have been possible without GitLab Duo,\" Daniel Le Berre notes. \"The AI features helped students, who might have lacked deep codebase knowledge, deliver meaningful contributions in just two weeks.\"***\n\nThis demonstrates how providing students with cutting-edge tools **accelerates their learning and impact**, allowing them to translate classroom knowledge into real-world contributions immediately.\n\n## Empowering open research and institutional control\n\nThe stability of the self-managed instance at Artois University is key to its success. This model guarantees **institutional control and stability** — a critical factor for long-term research preservation.\n\nThe institution's expertise in this area was recently highlighted in a major 2024 study led by CRIL, titled: \"[Higher Education and Research Forges in France - Definition, uses, limitations encountered and needs analysis](https://hal.science/hal-04208924v4)\" ([Project on GitLab](https://gitlab.in2p3.fr/coso-college-codes-sources-et-logiciels/forges-esr-en)). The research found that the vast majority of public forges in French Higher Education and Research relied on **GitLab**. This finding underscores the consensus among academic leaders that self-hosted solutions are essential for **data control and longevity**, especially when compared to relying on external, commercial forges.\n\n## Unlock GitLab Ultimate for your institution today\n\nThe success story of **Artois University's CRIL** proves the transformative power of the GitLab for Education program. By providing **free access to GitLab Ultimate**, we enable large-scale institutions to:\n\n1.  **Deliver a modern, integrated DevSecOps curriculum.**\n\n2.  **Support advanced, collaborative research projects with Ultimate planning features.**\n\n3.  **Empower students to make AI-assisted open source contributions.**\n\n4.  **Maintain institutional control and data longevity.**\n\nIf your academic institution is ready to equip its students and researchers with the complete DevSecOps platform and its most advanced features, we invite you to join the program.\n\nThe program provides **free access to GitLab Ultimate** for qualifying instructional and non-commercial research use cases.\n\n**Apply now [online](https://about.gitlab.com/solutions/education/join/).**\n",{"slug":728,"featured":27,"template":13},"artois-university-elevates-curriculum-with-gitlab-ultimate-for-education",{"content":730,"config":743},{"category":9,"tags":731,"body":734,"date":735,"updatedDate":736,"heroImage":737,"authors":738,"title":741,"description":742},[732,733,106],"tutorial","git","\nEnterprise teams are increasingly migrating from Azure DevOps to GitLab to gain strategic advantages and accelerate secure software delivery. \n\n\n- GitLab comes with integrated controls, policies, and [compliance frameworks](https://docs.gitlab.com/user/compliance/compliance_frameworks/) that allow organizations to implement software delivery standards at scale. This is especially important for regulated industries.\n\n- [Security testing](https://docs.gitlab.com/user/application_security/) is embedded in the pipeline and results show in the developer workflow, including static application security testing (SAST), source code analysis (SCA), dynamic application security testing (DAST), infrastructure-as-code scanning (IaC), container scanning, and API scanning.\n\n- [AI capabilities](https://about.gitlab.com/gitlab-duo-agent-platform/) across the full software delivery lifecycle include advanced agent orchestration and customizable flows to support how your organizational teams work.\n\n\nGitLab's open-source, open-core approach, flexible deployment options such as single-tenant dedicated and self-managed, and truly unified platform eliminate integration complexity and security gaps. \n\n\nFor teams facing mounting pressure to accelerate delivery while strengthening security posture and maintaining regulatory compliance, GitLab represents not just a migration but a platform evolution.\n\n\nMigrating from Azure DevOps to GitLab can seem like a daunting task, but with the right approach and tools, it can be a smooth and efficient process. This guide will walk you through the steps needed to successfully migrate your projects, repositories, and pipelines from Azure DevOps to GitLab.\n\n\n## Overview\n\nGitLab provides both [Congregate](https://gitlab.com/gitlab-org/professional-services-automation/tools/migration/congregate/) (maintained by [GitLab Professional Services](https://about.gitlab.com/professional-services/) organization) and [a built-in Git repository import](https://docs.gitlab.com/user/project/import/repo_by_url/) for migrating projects from Azure DevOps (ADO). These options support repository-by-repository or bulk migration and preserve git commit history, branches, and tags. With Congregate and professional services tools, we support additional assets such as wikis, work items, CI/CD variables, container images, packages, pipelines, and more (see this [feature matrix](https://gitlab.com/gitlab-org/professional-services-automation/tools/migration/congregate/-/blob/master/customer/ado-migration-features-matrix.md)). Use this guide to plan and execute your migration and complete post-migration follow-up tasks.\n\n\nEnterprises migrating from ADO to GitLab commonly follow a multi-phase approach:\n\n\n- Migrate repositories from ADO to GitLab using Congregate or GitLab's built-in repository migration.\n\n- Migrate pipelines from Azure Pipelines to GitLab CI/CD.\n\n- Migrate remaining assets such as boards, work items, and artifacts to GitLab Issues, Epics, and the Package and Container Registries.\n\n\nHigh-level migration phases:\n\n\n```mermaid\ngraph LR\n    subgraph Prerequisites\n        direction TB\n        A[\"Set up identity provider (IdP) and\u003Cbr/>provision users\"]\n        A --> B[\"Set up runners and\u003Cbr/>third-party integrations\"]\n        B --> I[\"Users enablement and\u003Cbr/>change management\"]\n    end\n    \n    subgraph MigrationPhase[\"Migration phase\"]\n        direction TB\n        C[\"Migrate source code\"]\n        C --> D[\"Preserve contributions and\u003Cbr/> format history\"]\n        D --> E[\"Migrate work items and\u003Cbr/>map to \u003Ca href=\"https://docs.gitlab.com/topics/plan_and_track/\">GitLab Plan \u003Cbr/>and track work\"]\n    end\n    \n    subgraph PostMigration[\"Post-migration steps\"]\n        direction TB\n        F[\"Create or translate \u003Cbr/>ADO pipelines to GitLab CI\"]\n        F --> G[\"Migrate other assets\u003Cbr/>packages and container images\"]\n        G --> H[\"Introduce \u003Ca href=\"https://docs.gitlab.com/user/application_security/secure_your_application/\">security\u003C/a> and\u003Cbr/>SDLC improvements\"]\n    end\n    \n    Prerequisites --> MigrationPhase\n    MigrationPhase --> PostMigration\n\n    style A fill:#FC6D26\n    style B fill:#FC6D26\n    style I fill:#FC6D26\n    style C fill:#8C929D\n    style D fill:#8C929D\n    style E fill:#8C929D\n    style F fill:#FFA500\n    style G fill:#FFA500\n    style H fill:#FFA500\n```\n\n\n## Planning your migration\n\n\n**To plan your migration, ask these questions:**\n\n\n- How soon do we need to complete the migration?\n\n- Do we understand what will be migrated?\n\n- Who will run the migration?\n\n- What organizational structure do we want in GitLab?\n\n- Are there any constraints, limitations, or pitfalls that need to be taken into account?\n\n\nDetermine your timeline, as it will largely dictate your migration approach. Identify champions or groups familiar with both ADO and GitLab platforms (such as early adopters) to help drive adoption and provide guidance.\n\n\n**Inventory what you need to migrate:**\n\n\n- The number of repositories, pull requests, and contributors\n\n- The number and complexity of work items and pipelines\n\n- Repository sizes and dependency relationships\n\n- Critical integrations and runner requirements (agent pools with specific capabilities)\n\n\nUse GitLab Professional Services's [Evaluate](https://gitlab.com/gitlab-org/professional-services-automation/tools/utilities/evaluate#beta-azure-devops) tool to produce a complete inventory of your entire Azure DevOps organization, including repositories, PR counts, contributor lists, number of pipelines, work items, CI/CD variables and more. If you're working with the GitLab Professional Services team, share this report with your engagement manager or technical architect to help plan the migration.\n\n\nMigration timing is primarily driven by pull request count, repository size, and amount of contributions (e.g. comments in PR, work items, etc). For example, 1,000 small repositories with few PRs and limited contributors can migrate much faster than a smaller set of repositories containing tens of thousands of PRs and thousands of contributors. Use your inventory data to estimate effort and plan test runs before proceeding with production migrations.\n\n\nCompare inventory against your desired timeline and decide whether to migrate all repositories at once or in batches. If teams cannot migrate simultaneously, batch and stagger migrations to align with team schedules. For example, in Professional Services engagements, we organize migrations into waves of 200-300 projects to manage complexity and respect API rate limits, both in [GitLab](https://docs.gitlab.com/security/rate_limits/) and [ADO](https://learn.microsoft.com/en-us/azure/devops/integrate/concepts/rate-limits?view=azure-devops).\n\n\nGitLab's built-in [repository importer](https://docs.gitlab.com/user/project/import/repo_by_url/) migrates Git repositories (commits, branches, and tags) one-by-one. Congregate is designed to preserve pull requests (known in GitLab as merge requests), comments, and related metadata where possible; the simple built-in repository import focuses only on the Git data (history, branches, and tags).\n\n\n**Items that typically require separate migration or manual recreation:**\n\n\n- Azure Pipelines - create equivalent GitLab CI/CD pipelines (consult with [CI/CD YAML](https://docs.gitlab.com/ci/yaml/) and/or with [CI/CD components](https://docs.gitlab.com/ci/components/)). Alternatively, consider using AI-based pipeline conversion available in Congregate.\n\n- Work items and boards - map to GitLab Issues, Epics, and Issue Boards.\n\n- Artifacts, container images (ACR) - migrate to GitLab Package Registry or Container Registry.\n\n- Service hooks and external integrations - recreate in GitLab.\n\n- [Permissions models](https://docs.gitlab.com/user/permissions/) differ between ADO and GitLab; review and plan permissions mapping rather than assuming exact preservation.\n\n\nReview what each tool (Congregate vs. built-in import) will migrate and choose the one that fits your needs. Make a list of any data or integrations that must be migrated or recreated manually.\n\n\n**Who will run the migration?**\n\n\nMigrations are typically run by a GitLab group owner or instance administrator, or by a designated migrator who has been granted the necessary permissions on the destination group/project. Congregate and the GitLab import APIs require valid authentication tokens for both Azure DevOps and GitLab.\n\n\n- Decide whether a group owner/admin will perform the migrations or whether you will grant a specific team/person delegated access.\n\n- Ensure the migrator has correctly configured personal access tokens (Azure DevOps and GitLab) with the scopes required by your chosen migration tool (for example, api/read_repository scopes and any tool-specific requirements). \n\n- Test tokens and permissions with a small pilot migration.\n\n**Note:** Congregate leverages file-based import functionality for ADO migrations and requires instance administrator permissions to run ([see our documentation](https://docs.gitlab.com/user/project/settings/import_export/#migrate-projects-by-uploading-an-export-file)). If you are migrating to GitLab.com, consider engaging Professional Services. For more information, see the [Professional Services Full Catalog](https://about.gitlab.com/professional-services/catalog/). Non-admin account cannot preserve contribution attribution!\n\n\n**What organizational structure do we want in GitLab?**\n\nWhile it's possible to map ADO structure directly to GitLab structure, it's recommended to rationalize and simplify the structure during migration. Consider how teams will work in GitLab and design the structure to facilitate collaboration and access management. Here is a way to think about mapping ADO structure to GitLab structure:\n\n\n```mermaid\ngraph TD\n    subgraph GitLab\n        direction TB\n        A[\"Top-level Group\"]\n        B[\"Subgroup (optional)\"]\n        C[\"Projects\"]\n        A --> B\n        A --> C\n        B --> C\n    end\n\n    subgraph AzureDevOps[\"Azure DevOps\"]\n        direction TB\n        F[\"Organizations\"]\n        G[\"Projects\"]\n        H[\"Repositories\"]\n        F --> G\n        G --> H\n    end\n\n    style A fill:#FC6D26\n    style B fill:#FC6D26\n    style C fill:#FC6D26\n    style F fill:#8C929D\n    style G fill:#8C929D\n    style H fill:#8C929D\n```\n\nRecommended approach:\n\n\n- Map each ADO organization to a GitLab group (or a small set of groups), not to many small groups. Avoid creating a GitLab group for every ADO team project. Use migration as an opportunity to rationalize your GitLab structure.\n\n- Use subgroups and project-level permissions to group related repositories.\n\n- Manage access to sets of projects by using GitLab groups and group membership (groups and subgroups) rather than one group per team project.\n\n- Review GitLab [permissions](https://docs.gitlab.com/ee/user/permissions.html) and consider [SAML Group Links](https://docs.gitlab.com/user/group/saml_sso/group_sync/) to implement an enterprise RBAC model for your GitLab instance (or a GitLab.com namespace).\n\n\n**ADO Boards and work items: State of migration**\n\n\nIt's important to understand how work items migrate from ADO into GitLab Plan (issues, epics, and boards).\n\n\n- ADO Boards and work items map to GitLab Issues, Epics, and Issue Boards. Plan how your workflows and board configurations will translate.\n\n- ADO Epics and Features become GitLab Epics.\n\n- Other work item types (e.g., user stories, tasks, bugs) become project-scoped issues.\n\n- Most standard fields are preserved; selected custom fields can be migrated when supported.\n\n- Parent-child relationships are retained so Epics reference all related issues.\n\n- Links to pull requests are converted to merge request links to maintain development traceability.\n\n\nExample: Migration of an individual work item to a GitLab Issue, including field accuracy and relationships:\n\n\n![Example: Migration of an individual work item to a GitLab Issue](https://res.cloudinary.com/about-gitlab-com/image/upload/v1764769188/ztesjnxxfbwmfmtckyga.png)\n\n\nBatching guidance:\n\n\n- If you need to run migrations in batches, use your new group/subgroup structure to define batches (for example, by ADO organization or by product area).\n\n- Use inventory reports to drive batch selection and test each batch with a pilot migration before scaling.\n\n\n**Pipelines migration**\n\n\nCongregate [recently introduced](https://gitlab.com/gitlab-org/professional-services-automation/tools/migration/congregate/-/merge_requests/1298) AI-powered conversion for multi-stage YAML pipelines from Azure DevOps to GitLab CI/CD. This automated conversion works best for simple, single-file pipelines and is designed to provide a working starting point rather than a production-ready `.gitlab-ci.yml` file. The tool generates a functionally equivalent GitLab pipeline that you can then refine and optimize for your specific needs.\n\n\n- Converts Azure Pipelines YAML to `.gitlab-ci.yml` format automatically.\n\n- Best suited for straightforward, single-file pipeline configurations.\n\n- Provides a boilerplate to accelerate migration, not a final production artifact.\n\n- Requires review and adjustment for complex scenarios, custom tasks, or enterprise requirements.\n\n- Does not support Azure DevOps classic release pipelines — [convert these to multi-stage YAML](https://learn.microsoft.com/en-us/azure/devops/pipelines/release/from-classic-pipelines?view=azure-devops) first.\n\n\nRepository owners should review the [GitLab CI/CD documentation](https://docs.gitlab.com/ci/) to further optimize and enhance their pipelines after the initial conversion.\n\n\nExample of converted pipelines:\n\n\n```yml \n\n# azure-pipelines.yml\n\ntrigger:\n  - main\n\nvariables:\n  imageName: myapp\n\nstages:\n  - stage: Build\n    jobs:\n      - job: Build\n        pool:\n          vmImage: 'ubuntu-latest'\n        steps:\n          - checkout: self\n\n          - task: Docker@2\n            displayName: Build Docker image\n            inputs:\n              command: build\n              repository: $(imageName)\n              Dockerfile: '**/Dockerfile'\n              tags: |\n                $(Build.BuildId)\n\n  - stage: Test\n    jobs:\n      - job: Test\n        pool:\n          vmImage: 'ubuntu-latest'\n        steps:\n          - checkout: self\n\n          # Example: run tests inside the container\n          - script: |\n              docker run --rm $(imageName):$(Build.BuildId) npm test\n            displayName: Run tests\n\n  - stage: Push\n    jobs:\n      - job: Push\n        pool:\n          vmImage: 'ubuntu-latest'\n        steps:\n          - checkout: self\n\n          - task: Docker@2\n            displayName: Login to ACR\n            inputs:\n              command: login\n              containerRegistry: '\u003Cyour-acr-service-connection>'\n\n          - task: Docker@2\n            displayName: Push image to ACR\n            inputs:\n              command: push\n              repository: $(imageName)\n              tags: |\n                $(Build.BuildId)\n\n```\n\n```yaml\n\n# .gitlab-ci.yml\n\nvariables:\n  imageName: myapp\n\nstages:\n  - build\n  - test\n  - push\n\nbuild:\n  stage: build\n  image: docker:latest\n  services:\n    - docker:dind\n  script:\n    - docker build -t $imageName:$CI_PIPELINE_ID -f $(find . -name Dockerfile) .\n  only:\n    - main\n\ntest:\n  stage: test\n  image: docker:latest\n  services:\n    - docker:dind\n  script:\n    - docker run --rm $imageName:$CI_PIPELINE_ID npm test\n  only:\n    - main\n\npush:\n  stage: push\n  image: docker:latest\n  services:\n    - docker:dind\n  before_script:\n    - docker login -u $CI_REGISTRY_USER -p $CI_REGISTRY_PASSWORD $CI_REGISTRY\n  script:\n    - docker tag $imageName:$CI_PIPELINE_ID $CI_REGISTRY/$CI_PROJECT_PATH/$imageName:$CI_PIPELINE_ID\n    - docker push $CI_REGISTRY/$CI_PROJECT_PATH/$imageName:$CI_PIPELINE_ID\n  only:\n    - main\n\n```\n\n**Final checklist:**\n\n\n- Decide timeline and batch strategy.\n\n- Produce a full inventory of repositories, PRs, and contributors.\n\n- Choose Congregate or the built-in import based on scope (PRs and metadata vs. Git data only).\n\n- Decide who will run migrations and ensure tokens/permissions are configured.\n\n- Identify assets that must be migrated separately (pipelines, work items, artifacts, and hooks) and plan those efforts.\n\n- Run pilot migrations, validate results, then scale according to your plan.\n\n\n## Running your migrations\n\n\nAfter planning, execute migrations in stages, starting with trial runs. Trial migrations help surface org-specific issues early and let you measure duration, validate outcomes, and fine-tune your approach before production.\n\n\nWhat trial migrations validate:\n\n\n- Whether a given repository and related assets migrate successfully (history, branches, tags; plus MRs/comments if using Congregate)\n\n- Whether the destination is usable immediately (permissions, runners, CI/CD variables, integrations)\n\n- How long each batch takes, to set schedules and stakeholder expectations\n\n\nDowntime guidance:\n\n\n- GitLab's built-in Git import and Congregate do not inherently require downtime.\n\n- For production waves, freeze changes in ADO (branch protections or read-only) to avoid missed commits, PR updates, or work items created mid-migration.\n\n- Trial runs do not require freezes and can be run anytime.\n\n\nBatching guidance:\n\n\n- Run trial batches back-to-back to shorten elapsed time; let teams validate results asynchronously.\n\n- Use your planned group/subgroup structure to define batches and respect API rate limits.\n\n\nRecommended steps:\n\n\n1. Create a test destination in GitLab for trials:\n\n\n  - GitLab.com: create a dedicated group/namespace (for example, my-org-sandbox)\n\n  - Self-managed: create a top-level group or a separate test instance if needed\n\n\n2. Prepare authentication:\n\n\n  - Azure DevOps PAT with required scopes.\n\n  - GitLab Personal Access Token with api and read_repository (plus admin access for file-based imports used by Congregate).\n\n\n3. Run trial migrations:\n\n\n  - Repos only: use GitLab's built-in import (Repo by URL)\n\n  - Repos + PRs/MRs and additional assets: use Congregate\n\n\n4. Post-trial follow-up:\n\n\n  - Verify repo history, branches, tags; merge requests (if migrated), issues/epics (if migrated), labels, and relationships.\n\n  - Check permissions/roles, protected branches, required approvals, runners/tags, variables/secrets, integrations/webhooks.\n\n  - Validate pipelines (`.gitlab-ci.yml`) or converted pipelines where applicable.\n\n\n5. Ask users to validate functionality and data fidelity.\n\n6. Resolve issues uncovered during trials and update your runbooks.\n\n7. Network and security:\n\n\n  - If your destination uses IP allow lists, add the IPs of your migration host and any required runners/integrations so imports can succeed.\n\n\n8. Run production migrations in waves:\n\n\n  - Enforce change freezes in ADO during each wave.\n\n  - Monitor progress and logs; retry or adjust batch sizes if you hit rate limits.\n\n\n9. Optional: remove the sandbox group or archive it after you finish.\n\n\n\u003Cfigure class=\"video_container\">\n  \u003Ciframe src=\"https://www.youtube.com/embed/ibIXGfrVbi4?si=ZxOVnXjCF-h4Ne0N\" frameborder=\"0\" allowfullscreen=\"true\">\u003C/iframe>\n\u003C/figure>\n\n\n## Terminology reference for GitLab and Azure DevOps\n\n| GitLab                                                           | Azure DevOps                                 | Similarities & Key Differences                                                                                                                                          |\n| ---------------------------------------------------------------- | -------------------------------------------- | --------------------------------------------------------------------------------------------------------------------------------------------------------------------- |\n| Group                                                            | Organization                                 | Top-level namespace, membership, policies. ADO org contains Projects; GitLab Group contains Subgroups and Projects.                                                   |\n| Group or Subgroup                                                | Project                                      | Logical container, permissions boundary. ADO Project holds many repos; GitLab Groups/Subgroups organize many Projects.                                                |\n| Project (includes a Git repo)                                    | Repository (inside a Project)                | Git history, branches, tags. In GitLab, a \"Project\" is the repo plus issues, CI/CD, wiki, etc. One repo per Project.                                                  |\n| Merge Request (MR)                                               | Pull Request (PR)                            | Code review, discussions, approvals. MR rules include approvals, required pipelines, code owners.                                                                     |\n| Protected Branches, MR Approval Rules, Status Checks             | Branch Policies                              | Enforce reviews and checks. GitLab combines protections + approval rules + required status checks.                                                                    |\n| GitLab CI/CD                                                     | Azure Pipelines                              | YAML pipelines, stages/jobs, logs. ADO also has classic UI pipelines; GitLab centers on .gitlab-ci.yml.                                                               |\n| .gitlab-ci.yml                                                   | azure-pipelines.yml                          | Defines stages/jobs/triggers. Syntax/features differ; map jobs, variables, artifacts, and triggers.                                                                   |\n| Runners (shared/specific)                                        | Agents / Agent Pools                         | Execute jobs on machines/containers. Target via demands (ADO) vs tags (GitLab). Registration/scoping differs.                                                         |\n| CI/CD Variables (project/group/instance), Protected/Masked       | Pipeline Variables, Variable Groups, Library | Pass config/secrets to jobs. GitLab supports group inheritance and masking/protection flags.                                                                          |\n| Integrations, CI/CD Variables, Deploy Keys                       | Service Connections                          | External auth to services/clouds. Map to integrations or variables; cloud-specific helpers available.                                                                 |\n| Environments & Deployments (protected envs)                      | Environments (with approvals)                | Track deploy targets/history. Approvals via protected envs and manual jobs in GitLab.                                                                                 |\n| Releases (tag + notes)                                           | Releases (classic or pipelines)              | Versioned notes/artifacts. GitLab Release ties to tags; deployments tracked separately.                                                                               |\n| Job Artifacts                                                    | Pipeline Artifacts                           | Persist job outputs. Retention/expiry configured per job or project.                                                                                                  |\n| Package Registry (NuGet/npm/Maven/PyPI/Composer, etc.)           | Azure Artifacts (NuGet/npm/Maven, etc.)      | Package hosting. Auth/namespace differ; migrate per package type.                                                                                                     |\n| GitLab Container Registry                                        | Azure Container Registry (ACR) or others     | OCI images. GitLab provides per-project/group registries.                                                                                                             |\n| Issue Boards                                                     | Boards                                       | Visualize work by columns. GitLab boards are label-driven; multiple boards per project/group.                                                                         |\n| Issues (types/labels), Epics                                     | Work Items (User Story/Bug/Task)             | Track units of work. Map ADO types/fields to labels/custom fields; epics at group level.                                                                              |\n| Epics, Parent/Child Issues                                       | Epics/Features                               | Hierarchy of work. Schema differs; use epics + issue relationships.                                                                                                   |\n| Milestones and Iterations                                        | Iteration Paths                              | Time-boxing. GitLab Iterations (group feature) or Milestones per project/group.                                                                                       |\n| Labels (scoped labels)                                           | Area Paths                                   | Categorization/ownership. Replace hierarchical areas with scoped labels.                                                                                              |\n| Project/Group Wiki                                               | Project Wiki                                 | Markdown wiki. Backed by repos in both; layout/auth differ slightly.                                                                                                  |\n| Test reports via CI, Requirements/Test Management, integrations  | Test Plans/Cases/Runs                        | QA evidence/traceability. No 1:1 with ADO Test Plans; often use CI reports + issues/requirements.                                                                     |\n| Roles (Owner/Maintainer/Developer/Reporter/Guest) + custom roles | Access levels + granular permissions         | Control read/write/admin. Models differ; leverage group inheritance and protected resources.                                                                          |\n| Webhooks                                                         | Service Hooks                                | Event-driven integrations. Event names/payloads differ; reconfigure endpoints.                                                                                        |\n| Advanced Search                                                  | Code Search                                  | Full-text repo search. Self-managed GitLab may need Elasticsearch/OpenSearch for advanced features.                                                                   |\n","2025-12-03","2026-01-16","https://res.cloudinary.com/about-gitlab-com/image/upload/v1749658924/Blog/Hero%20Images/securitylifecycle-light.png",[739,740],"Evgeny Rudinsky","Michael Leopard","Guide: Migrate from Azure DevOps to GitLab","Learn how to carry out the full migration from Azure DevOps to GitLab using GitLab Professional Services migration tools — from planning and execution to post-migration follow-up tasks.",{"featured":27,"template":13,"slug":744},"migration-from-azure-devops-to-gitlab",{"promotions":746},[747,761,772],{"id":748,"categories":749,"header":751,"text":752,"button":753,"image":758},"ai-modernization",[750],"ai-ml","Is AI achieving its promise at scale?","Quiz will take 5 minutes or less",{"text":754,"config":755},"Get your AI maturity score",{"href":756,"dataGaName":757,"dataGaLocation":241},"/assessments/ai-modernization-assessment/","modernization assessment",{"config":759},{"src":760},"https://res.cloudinary.com/about-gitlab-com/image/upload/v1772138786/qix0m7kwnd8x2fh1zq49.png",{"id":762,"categories":763,"header":764,"text":752,"button":765,"image":769},"devops-modernization",[725,557],"Are you just managing tools or shipping innovation?",{"text":766,"config":767},"Get your DevOps maturity score",{"href":768,"dataGaName":757,"dataGaLocation":241},"/assessments/devops-modernization-assessment/",{"config":770},{"src":771},"https://res.cloudinary.com/about-gitlab-com/image/upload/v1772138785/eg818fmakweyuznttgid.png",{"id":773,"categories":774,"header":776,"text":752,"button":777,"image":781},"security-modernization",[775],"security","Are you trading speed for security?",{"text":778,"config":779},"Get your security maturity score",{"href":780,"dataGaName":757,"dataGaLocation":241},"/assessments/security-modernization-assessment/",{"config":782},{"src":783},"https://res.cloudinary.com/about-gitlab-com/image/upload/v1772138786/p4pbqd9nnjejg5ds6mdk.png",{"header":785,"blurb":786,"button":787,"secondaryButton":792},"Start building faster today","See what your team can do with the intelligent orchestration platform for DevSecOps.\n",{"text":788,"config":789},"Get your free trial",{"href":790,"dataGaName":48,"dataGaLocation":791},"https://gitlab.com/-/trial_registrations/new?glm_content=default-saas-trial&glm_source=about.gitlab.com/","feature",{"text":493,"config":793},{"href":52,"dataGaName":53,"dataGaLocation":791},1772652077647]