This summary infographic compares JFrog and GitLab across several DevOps Stages and Categories. The methodology used to build this chart can be found here. More detailed comparisons and commentary on strenghts, gaps etc. are in sections below.
JFrog has transitioned from an artifact repository to a DevOps Platform that includes CI and CD capabilities through its acquisition of Shippable in Feb 2019. Recently in March 2020, JFrog announced the launch of its DevOps platform called 'JFrog Platform', a pre-integrated solution with a common UI across JFrog Pipelines, JFrog X-Ray and JFrog Source Composition Analysis products. This solution is backed by a common meta data model that facilitates information integration between these separate product. In addition to three primary products JFrog Artifactory, JFrog Pipelines and JFrog Xray, JFrog also provides other products such as JFrog Distribution, JFrog Mission Control and JFrog Container Registry.
JFrog Artifactory is a tool designed to store the binary output of the build process for use in distribution and deployment. Artifactory provides support for a number of package formats. JFrog Artifactory provides a single source of truth for build artifacts and works with JFrog Distribution to efficiently distribute large artifacts across the enterprise.
JFrog Pipelines is a CI-CD product that works well with its Artifactory repository. JFrog pipelines works through a combination of native steps (a set of higher order steps built on bash) and resources (inputs into or outputs from native steps, can be any type such as a build, integration etc.). JFrog pipelines is a functional CI-CD product, though it lacks several capabilities typically foudn in enterprise class products.
JFrog Xray is the security product that can be built-into various steps within a JFrog pipeline. Xray supports detecting security vulnerabilities in all dependent code and also provides license compliance capabilities.
Artifactory provides support for a number of package formats such as Maven, Debian, NPM, Helm, Ruby, Python, and Docker. Artifactory also stores a complete map of all the components that went into creating the artifact. This information feeds other products such as JFrog Xray. Artifacts can be efficiently distributed across remote sites using JFrog Distribution.
GitLab also offers the ability to store and distribute packages, but at the moment offers less package type compatibility than Artifactory does - Maven, Docker, NPM. GitLab strengths are in providing a single product for the full DevOps Lifecycle. In addition, GitLab CI-CD and Security Capabilities have better functionality and provide enterprise grade capabilities.
JFrog pipelines, through acquisition of Shippable, is a functional CI-CD product. JFrog Pipelines attempts to make it simpler to do CI-CD by building 'Native Steps'. This is akin to a prebuilt component or step in the CI-CD process that can be described in Yaml, thereby hiding all the low level complexity from the user. Some examples of Native Steps are Docker Build, Docker Push, NPM Build, NPM Publish, and XrayScan. JFrog Pipelines has several strengths and weaknesses. The main impact of its weaknesses are longer build times and lower collaboration.
JFrog Xray provides static application testing capabilities by scanning the application components for vulnerabilities against the VulnDB vulnerability database. Xray also provides security policy enforcement and capability to monitor for license compliance. Xray integrates with IDEs such as IntelliJ and allows developers to view security issues in the dev environment.
Note: This chart was developed by comparing the feature categories supported by GitLab and JFrog. For example, the ratio "5/7" for GitLab in Plan stage indicates support for 5 out of 7 feature categories within that DevOps Lifecycle Stage. We then applied certain % thresholds to color code the bars. In keeping with GitLab value of transparency, we applied this scoring methodology both to GitLab and JFrog capabilities, which is the reason you will see in some cases GitLab scores less than perfect scores. If you have questions about the analysis or additional inputs please feel free to submit an issue by clicking the link at the bottom of this page or writing a comment.