TrinityX: a hugely scalable, Open Source management system for HPC, AI and Data cluster environments
Our development of TrinityX is the accumulation of having installed and designed HPC, AI and Data solutions for over sixteen years. It is the end result of having dealt with many complex Linux environments consisting of one up to thausands of nodes and signalling what could be improved upon. Designed from the ground up to provide all services required in a modern high performance system, TrinityX allows full customisation of your cluster installation.
We use the term “ecosystem” because TrinityX is more than just a cluster management suite.
On top of all the required services to orchestrate the function, administration, and use of cluster, TrinityX delivers a complete user environment and support for a large variety of HPC or HPC-adjacent platforms and tools. Due to its modern, modular design, TrinityX is endlessly adaptable to your particular needs.
Luna is a baremetal provisioning tool that employs the BitTorrent protocol to provision nodes. As such, every booting node helps the others to boot. It delivers full images of operating systems and not a ‘recipe’ on how to configure one. It also dramatically speeds up installation time, and reduces administrative efforts.
Within ClusterVision, we enable our engineers and developers to work on personal projects that often lead to new insights and innovative solutions. Luna was originally developed by one of our engineers in his free time and was later incorporated into TrinityX. We encourage this to further advance the technology behind HPC.
- Scalable to tens of thousands of nodes
- PDSH parallel command
- Shared storage for essential configuration files and home directories
- Our lightning-fast provisioning tool that can get 500 nodes up and running in under 5 minutes
- Full hardware integration, e.g. Infiniband, Omnipath, PXE, Intel Broadwell, IPMI, switches, OpenPOWER, and more
- Complete HPC user environment (modules environment, scientific libraries, compilers, profilers, debuggers)
- NPI libraries: OpenMPI, Intel MPI, and optional mpich/mvapich/mvapich2
- The Simple Linux Utility for Resource Management (SLURM) preconfigured to make full use of a cluster
- Full HPC performance using the optional Docker-based application containerisation
- High availability for controllers, storage, and login nodes
- Parallel file system support: Lustre, IBM Spectrum Scale (GPFS), and BeeGFS
- An integrated authentication system that can be plugged to existing back-ends with minor tweaks
- A comprehensive monitoring and metering system to keep track of critical events and resource usage
- Node-to-switch integration, automatic discovery
Now featuring Ansible for central configuration management and streamlined automation of tasks like software updates and implementation of services. A powerful tool to increase efficiency.
TrinityX is a fully open-source HPC platform. At ClusterVision, we believe free exchange of ideas leads to a better result. In this spirit, we rely on open-source tools when building software features and have opened up our own codebases for collaboration on GitHub. You can find the repository for TrinityX on GitHub.