![[WIP] Solution HPC DLC en rack Châssis de serveurs lame Cloe 10](/web/image/product.image/603/image_1024/%5BWIP%5D%20Solution%20HPC%20DLC%20en%20rack%C2%A0%20Ch%C3%A2ssis%20de%20serveurs%20lame%20Cloe%2010?unique=7ca8fb5)
![[WIP] Solution HPC DLC en rack Châssis de serveurs lame Cloe 10](/web/image/product.image/603/image_128/%5BWIP%5D%20Solution%20HPC%20DLC%20en%20rack%C2%A0%20Ch%C3%A2ssis%20de%20serveurs%20lame%20Cloe%2010?unique=7ca8fb5)
Why is HPC important?
Engineering simulation
CFD, crash testing and structural analysis run at full fidelity. Aerospace, automotive and industrial teams explore the whole design space instead of rationing runs, and prototype far less.
Life sciences
Genome sequencing, molecular dynamics and drug screening depend on sustained parallel compute. HPC compresses pipelines that would take months on conventional infrastructure into days.
Energy and earth sciences
Seismic imaging, reservoir modelling, weather forecasting and climate projection all process vast datasets against hard deadlines, forecasts are worthless if they arrive late.
Rack solution
fully configured for HPC at scale
Redefine Your HPC Infrastructure at Scale
This rack solution, built on the Cloe 10 platform, is a turnkey HPC Infrastructure engineered for the most demanding workloads. From fluid simulations to genomics and compute-intensive tasks, this solution delivers uncompromising performance, thermal efficiency, and operational simplicity, all within a single, fully integrated rack.
Scalable from Day One
This ecosystem is designed to grow with your business. If you are deploying a single rack or scaling to a multi-rack HPC supercomputer with thousands of CPUs, we provide a proven, end-to-end solution backed by deep partnerships with major CPU makers, like Intel and AMD.
As an energy-efficient server manufacturer, we hold every component, configuration, and deployment to validation standards for HPC infrastructure. The result: proven compatibility, optimized performance, and reliability you can build mission-critical workloads on.
Turnkey deployment | Multi-rack scalability
Integrated liquid cooling | Flexible Network configuration
Peak performance, sustained
Liquid-cooled components run well within their thermal limits, keeping servers stable and delivering consistently high performance.
Lower power, same output
Fewer fans, lower speeds, same performance: liquid-cooled servers match air-cooled output while drawing less power.
Quiet operation
Air cooling means banks of high-speed, high-noise fans. 2CRSi's liquid cooling cuts fan count, and noise, dramatically.
Decades of cooling expertise
Our liquid cooling supplier brings 20 years of cooling expertise across desktop PCs and data centers, backed by 2CRSi's own 10+ years in the field.
Built to run and last
Liquid cooling systems are monitored continuously and need little maintenance, with component warranties from 2CRSi and our cooling partners.
Ease of use
2CRSi servers with liquid cooling kits mount straight into a rack or tie into a building's water system, with quick, easy, dry-break disconnects.
Networking solutions
Interconnect is a design choice, not a fixed option. We build the fabric to your job profile: NDR/XDR InfiniBand for tightly-coupled MPI, or 25 to 400G Ethernet with RoCE, in non-blocking fat-tree, or over-subscribed where the budget goes further. Topology, cabling and transceivers designed, integrated and validated in our factory.
Storage solutions
Block, object or file, sized to your I/O pattern rather than to a catalogue. NVMe scratch for checkpointing and metadata-heavy workloads, HDD tiers for capacity, and parallel filesystems that scale bandwidth with the node count. Privileged partnerships across Lustre, BeeGFS and parallel NFS solutions, benchmarked before the racks leave us.
Coolant Distribution Unit
The in-rack liquid-to-liquid CDU delivers 100 kW of cooling capacity at a 5°C approach temperature, bringing direct-to-chip liquid cooling to high-density racks without requiring facility-level infrastructure changes.
The unit runs on standard single-phase AC 220–240 V / 50–60 Hz power, simplifying deployment in existing data center environments. Built for continuous operation, it features 1+1 redundant power supplies and pumps, so a single component failure never interrupts cooling to the rack.
Deploying a multi-rack HPC supercomputer with hundreds or thousands CPUs? We can propose more optimized configurations, including in-row CDUs supporting up to 1350 kW.
In-Rack CDU
In-Row CDU (for large scale deployment)