SC122 — Dual-Socket 64-Core Intel CFD HPC Node with 768 GB Memory and 7.5 TB Storage for ANSYS Fluent

The SC122 node completely redefines memory boundaries for complex simulation. Powered by two Intel Xeon Gold 6530 processors, it delivers a highly capable 64 physical cores and 128 threads with a 4.0 GHz boost clock. But the defining feature of this node is its colossal memory footprint: an unprecedented 768 GB of RAM. Paired with two 3.75 TB SATA SSDs, providing roughly 7.5 TB of usable storage, this machine is a specialized heavyweight for the largest, most memory-intensive single-node solves.

By pairing a solid 64-core Intel compute foundation with 768 GB of memory, SC122 provides one of the highest memory-to-core ratios available. It is engineered for enterprise-scale meshes and hyper-complex physics models that simply cannot be loaded—let alone solved—on standard high-performance workstations.

Specifications

Compared to our standard 64-core or 128-core nodes, SC122 shifts the investment entirely into memory capacity. With 768 GB spread across 64 cores, it offers a staggering 12 GB of RAM per core. This extreme ratio ensures that no matter how complex your physics are, or how massive your single-phase mesh is, the solver will never run out of memory before it runs out of compute.

What this node is built for

SC122 is the extreme-memory tier. It suits:

• Enterprise-scale monolithic meshes — 768 GB of RAM opens the door to colossal meshes (approaching 700+ million cells) that must be held in a single machine's memory architecture.

• Extreme memory-per-cell physics — Dense combustion networks, complex discrete phase models (DPM), melting/solidification, and heavy multiphase flows where each cell requires immense amounts of RAM to calculate.

• Intel-optimized workflows — Proprietary User Defined Functions (UDFs) and specific compiling habits that rely on Intel’s highly refined AVX-512 instruction sets and Intel MPI.

• Massive meshing operations — Generating hundreds of millions of poly-hexcore cells in Fluent Meshing requires huge overhead memory; SC122 powers through massive meshing tasks without hanging.

Understanding the dual-socket layout

A two-socket machine is not simply a bigger one-socket machine. Each processor has its own memory controllers and its own local RAM, and a core reaching data attached to the other socket pays a latency penalty crossing the interconnect between them.

For Fluent this is manageable and worth knowing about. Partitioning the mesh so each socket works mostly on locally-held data — and pinning MPI ranks accordingly — keeps that cross-socket traffic low. Run without any NUMA awareness and you can lose some of the bandwidth advantage the second socket provides. We configure this for you as part of the setup, so in practice it's a detail we handle rather than one you have to manage.

Choosing SC122

Choose SC122 when:

• Your case requires more than 384 GB of memory to simply open and initialize, or

• You are generating a colossal mesh in Fluent Meshing and need massive memory overhead, or

• Your physics are so memory-intensive that you require 12 GB of RAM per core to avoid solver starvation, and

• You prefer the proven stability of Intel Xeon architecture.

If your case fits in 256 GB and you want peak parallel speed, a 128-core node (like SC92) will cut your solve times much faster. If you need peak NVMe write speeds or 22 TB of storage for heavy transients, you should look at our massive-storage tiers. SC122’s sole, unbeatable advantage is its 768 GB memory pool—it runs the cases that crash everything else.

Storage: balanced capacity for steady-state

Two 3.75 TB SATA SSDs provide 7.5 TB of reliable storage. While it does not feature the 22 TB bulk of our massive arrays or the instantaneous peak speed of NVMe, 7.5 TB is perfectly balanced for monolithic steady-state runs, massive meshing outputs, and moderate transient workflows. It provides plenty of space to house colossal .cas and .dat files generated by 700-million cell models without inflating the cost with unnecessary storage hardware.

How parallel solving works on this node

When you run Fluent in parallel, the solver partitions your mesh across 64 Intel cores. Because of the 768 GB memory pool, you can pass incredibly large chunks of the mesh to each core. Intel's architecture handles the MPI cross-talk beautifully across the dual sockets, while high-bandwidth DDR5 memory feeds the massive data chunks to the processing pipelines. The result is the ability to reliably chew through ultra-large-scale problems that are impossible to partition on smaller nodes.

Why rent instead of buy

How it works — you're in full control

SC122 is yours to drive directly. You connect to the node over a secure remote desktop connection and run everything yourself, exactly as you would on a local workstation:

1. Connect via remote desktop — you get full remote access to the SC122 node.

2. Set up your simulation — load your geometry and case, configure your solver settings, and run ANSYS Fluent with all 64 cores at your command.

3. Run and monitor in real time — start, pause, adjust, and watch the solution as it progresses, with complete control throughout.

4. Keep your results — post-process on the node or download your solution and data files whenever you're ready.

You have full authority over the node for the duration of your rental: we provide the hardware and the support, and you run the simulations your way.

Frequently asked questions

Why 768 GB of RAM for only 64 cores?

Not all CFD problems are compute-bound; many are memory-bound. If you have a 500-million cell model, a 64-core node with 256 GB of RAM will simply fail to allocate memory and crash. SC122 is designed specifically to hold massive cell counts or extreme physics that require up to 12 GB of RAM per core, allowing the 64 cores to successfully calculate the solution.

How is SC122 different from the AMD-based SC89?

Both are 64-core nodes, but SC122 uses Intel Xeon architecture and doubles the memory of the SC89 (768 GB vs 384 GB). While SC89 uses faster NVMe storage, SC122 is the undisputed champion of pure RAM capacity.

Is 7.5 TB of SATA storage enough for massive meshes?

Yes, for steady-state and standard workflows. A 500-million cell case file might be tens of gigabytes in size. 7.5 TB can hold hundreds of these massive files. It only becomes a bottleneck if you are running highly frequent autosaves on an ultra-long transient campaign.

How large a mesh can it handle?

Assuming roughly 1 GB of RAM per million cells for a standard steady-state run, 768 GB theoretically opens the door to meshes approaching 700+ million cells, software scaling limits permitting. Send us your case for a precise evaluation.

Is my data kept confidential?

Yes.

Get started with SC122

If your simulation is crashing due to memory exhaustion, or you are generating meshes so large they require an enterprise-scale RAM footprint just to load, SC122 is the ultimate capacity solution.

Contact MR CFD to run your ANSYS Fluent case on SC122 — and we'll help you provision the ultimate memory environment for your largest workflows.