Challenges:

• Re-Architecting a legacy & mission-critical applications
• Refactor legacy C application into container images for deployment onto the localized OpenShift cluster
• Integrating existing external tools
• Nexus Repository
• External SCM (corporate GitLab)
• Jenkins pipeline service into the CI/CD process needed for updating, delivering, and deploying crucial software components

Solution:

• Containerizing Application
• Application code assessment
• Refactored the application to core functionality to reduce image size
• Integrating GitLab
• Assessed current SCM practices, tools, and architecture
• Identified blockers with regards to network architecture with an external SCM
• Integrated the corporate SCM to be internal to the OpenShift cluster environment through the use of a Gogs Git server
• Integrating CI/CD Pipelines
• Utilized the included OpenShift certified image catalog to deploy a Jenkins pipeline server inside of the OpenShift cluster
• Teamed with Red Hat OpenShift Networking SMEs to identify networking roadblocks and solutions
• Integrated an external corporate images repository (Nexus) into the OpenShift cluster

Results:

• A complete, end-to-end CI/CD solution for a localized, security-restricted OpenShift environment
• Re-Architecture of a mission-critical legacy C application to work in a containerized, cloud-native environment
• Internalization and integration of the corporate SCM into the OpenShift cluster
• Deployment of an OpenShift-native Jenkins CI/CD Pipeline
• Integration of the corporate, external Nexus image repository
• Troubleshoot and overcome advanced security and networking issues to integrate all solutions in a self-contained, highly secure, and isolated OpenShift environment

The Client:
The railway offers the largest intermodal rail network in eastern North America

The Challenge:

The customer was building out a GPU cluster and was requesting assistance with this effort. To fully achieve their objective, they were looking for set-up, advisement and assistance with container builds to enable their team once deployed. The initial effort also included a graphical interface, procedural assistance with downloading Python images and pinning those versions. In addition, requested the use of Nvidia base images as a starting point and Gitlab optimization for this project.

The Results:

• Infrastructure Build out and Validation for Gitlab, Gitlab CI, Gitlab Docker Registry, R, Python packages
• Infrastructure Automation
• Built Docker-based development environment
• Provided options for both command line and GUI approaches
• Developed a plan for this type of workload in the future