BullSequana XH3500 Supercomputer

The rapid evolution of AI, simulation, and hybrid quantum-classical workloads is pushing HPC infrastructure to its limits. Data centres now demand exascale-class performance, extreme density, and energy efficiency within constrained footprints. Rising GPU power peaks, thermal loads, and stricter security and compliance requirements make a new kind of supercomputing platform essential. BullSequana XH3500 meets these challenges with modular design, advanced cooling, and integrated hardware/software for future-ready workloads.

BullSequana XH3500 stands out with:

• Ultra-scalability & future readiness through modular architecture and backed with decades of field-proven supercomputing experiences.

• Best-in-class energy efficiency with 100% fanless Direct Liquid Cooling and smart power management, heat reuse ready.

• Maximum compute density & flexibility supporting CPUs, GPUs, and multiple interconnect technologies without vendor lock-in.

• Resilience and security by design, including secure boot, signed firmware, TPM, and compliance with EU standards.

• The platform uses full-rack Direct Liquid Cooling for complete heat capture and integrates an in-rack Coolant Distribution Unit for simplified scaling.

• Energy optimisation software like BullSequana ARGOS enables smart power capping, real-time carbon tracking, and workload-aware scheduling, ensuring every watt delivers maximum performance.

• Bull’s track record in energy efficiency has been proven by its leading position in GREEN500 listing consecutively.

Bull’s software stack maximises HPC performance and efficiency on BullSequana XH3500 through advanced management, monitoring, and optimisation tools.

• The Smart Management Centre (SMC) simplifies cluster operations with automated provisioning, real-time monitoring, and multi-layered security for thousands of nodes.

• Performance tools like BullSequana ARGOS deliver energy and workload profiling, while IBMS ensures optimal interconnect fabric health.

• Bull Slurm and OpenMPI provide validated, high-performance workload and communication management, and container solutions like Singularity enable secure AI and HPC workflows.

• For future-ready innovation, JARVICE accelerates AI deployment, and Qaptiva™ integrates quantum simulation into HPC environments, all backed by expert support for reliability and scalability.

Bull accelerates BullSequana XH3500’s value delivery through a complete ecosystem of solution components.

• These include ready-to-deploy service nodes, modular data centre designs, and high-performance interconnects for seamless scaling.

• Proven software stacks like Smart Management Centre simplify cluster operations, while BullSequana ARGOS optimises energy and performance.

• Beyond technology, Bull brings field-proven experience in configuring and stabilising complex workloads and applications at large scale, along with validated use cases that help clients deploy AI faster and achieve real-world impact rapidly.

Combined with expert services and a Centre of Excellence in Performance Programming (CEPP), these building blocks ensure peak performance, reduced time to market, and maximum ROI.