SC106 — 64-Core CFD HPC Node with 256 MB Cache and 12-Channel Memory for ANSYS Fluent

The SC106 node is the widest parallel tier we offer. It runs a single AMD EPYC 9555 — 64 cores and 128 threads at a 4.2 GHz all-core boost — with 256 MB of L3 cache and twelve DDR5 memory channels behind it. AMD lists CFD and HPC explicitly among this chip's target workloads. Paired with 128 GB of RAM and 2 × 1.88 TB NVMe SSD, it's the node for large cases that genuinely scale across many cores.

Sixty-four cores in one socket also means no network between ranks: what would otherwise be a small cluster runs entirely on-board, with MPI traffic crossing fast internal channels rather than an interconnect.

Specifications

Sixty-four cores is the widest spread available on one socket here, and twelve memory channels — roughly 600 GB/s — keep them supplied rather than queuing for data. Notably, the 4.2 GHz all-core boost is higher than our 48-core node's, so you're not paying for the extra cores with clock speed. That 4.2 GHz is the figure held with every rank loaded, not a single-core peak.

Sizing note: this is a throughput node, not a capacity node

SC106 carries the same 128 GB of RAM as the 8-core tiers, spread across 64 cores — about 2 GB per core. That's the key thing to understand before choosing it. This node is built to solve cases that fit in 128 GB as fast as possible, not to hold larger ones.

It also means the case needs to be big enough to justify 64 partitions while still fitting in that memory. As a rough guide, Fluent needs about 1 GB of RAM per million cells for a typical steady single-phase case, so the sweet spot here is a mesh in the tens to low hundreds of millions of cells — large enough that each of the 64 ranks has real work, small enough to fit. If your mesh needs more than 128 GB, more cores won't help; our high-memory nodes are the right tool.

What this node is built for

SC106 is the maximum-throughput tier. It suits:

• Large cases that scale well past 32 or 48 cores — 64 MPI ranks split a mesh finer than anything else here, and with the bandwidth to back them, that converts into shorter wall-clock time.

• Bandwidth-heavy physics — multiphase, combustion, many transported scalars, coupled solvers: cases that move a lot of data per iteration.

• Time-critical turnaround — when a run that would take days on a smaller node needs to finish overnight.

• Large parameter sweeps and design studies — 64 cores means substantially more work per hour, so you get through far more design iterations per day.

• Cluster-scale jobs without a cluster — 64 ranks on one socket, with no network hop between them.

• Transient, LES, and DES runs writing heavy data — NVMe storage keeps large data files and frequent autosaves from slowing down the solve.

Choosing SC106

Choose SC106 when:

• Your case has the cell count to keep 64 partitions genuinely busy, and

• It fits within 128 GB, and

• Your solve is limited by throughput rather than single-thread speed or memory capacity.

All three conditions matter. Fluent scales well only while each core still has enough cells to stay busy relative to the boundary data it exchanges with its neighbors — and at 64 ranks, that threshold is high. Split a moderate case this many ways and the cores will spend more time communicating than solving, at which point a 16- or 24-core node finishes first, and a high-clock 8-core tier may beat both. More cores is not automatically faster, and on this node the gap between a well-sized case and a poorly-sized one is the largest in the lineup.

Send us your case before booking this tier — we'd rather point you to a smaller node that solves it faster than rent you cores your case can't use.

Fast NVMe storage

The two 1.88 TB NVMe SSDs give substantially higher read/write throughput than SATA SSDs, plus nearly 3.76 TB of combined capacity — which matters most on the highest-throughput node in the lineup, where a fast solve shouldn't be held up waiting on disk. That means faster loading of large case and data files, quicker autosaves, and room to keep a full DOE sweep's output on disk without constantly clearing space.

How parallel solving works on this node

When you run Fluent in parallel, the solver partitions your mesh across all 64 cores (using algorithms such as METIS), each core solving its share of cells and exchanging boundary data each iteration via MPI. Running 64 ranks on a single socket is a real structural advantage over spreading the same job across multiple networked machines: all that communication stays on fast internal channels instead of crossing a network, so parallel overhead stays low at a rank count where it would otherwise start to dominate. Between exchanges, twelve memory channels and 256 MB of shared L3 keep the cores fed rather than stalling.

Why rent instead of buy

A 64-core server is a substantial purchase for work that may only need it on occasion. Renting gives you the throughput for the jobs that call for it, and lets you drop back to a smaller tier for everything else.

How it works — you're in full control

SC106 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 SC106 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

How is SC106 different from our 48-core node? A third more cores on the same 12-channel platform with the same 256 MB cache — and a slightly higher all-core clock (4.2 vs. 4.1 GHz). If your case scales efficiently past 48 ranks, SC106 is the faster node outright; if it doesn't, the extra cores sit idle.

Will 64 cores always be faster? No, and this is the node where that's most likely to bite. Fluent scales well only while each core has enough cells to stay busy relative to the data it exchanges with its neighbors. A moderate case split 64 ways can spend more time on communication than on solving. Send us your case and we'll check the scaling before you book.

Is 128 GB enough memory for 64 cores? For cases that fit in 128 GB, yes — that's exactly what this node is sized for. It's a throughput node, not a capacity node. If your mesh needs more than 128 GB, adding cores won't help; our high-memory nodes are the right tier.

Is 4.2 GHz slow compared to a 5.7 GHz node? On a single thread, yes. But 5.7 GHz figures are single-core peaks that come down under a full parallel load. SC106's 4.2 GHz is the all-core number, held with all 64 cores working — which is what a real parallel solve sees.

Is one 64-core node better than several smaller networked nodes? For a single case, usually yes: MPI traffic between 64 ranks stays on-board instead of crossing a network, which keeps parallel overhead lower than the same rank count spread across machines.

Is my data kept confidential? Yes.

Get started with SC106

If your cases are large, genuinely scale across many cores, and fit within 128 GB, SC106 gives you cluster-scale parallelism in a single node.

Contact MR CFD to run your ANSYS Fluent case on SC106 — and we'll help you confirm your case can use all 64 cores before you book.