SC107 — 48-Core CFD HPC Node with 256 MB Cache and 12-Channel Memory for ANSYS Fluent

The SC107 node is the widest parallel tier in the lineup. It runs a single AMD EPYC 9455 — 48 cores and 96 threads at a 4.1 GHz all-core boost — with 256 MB of L3 cache and twelve DDR5 memory channels behind it. AMD lists CFD explicitly among this chip's target workloads, and the reason is the combination: a high core count backed by roughly 600 GB/s of memory bandwidth, which is what a heavily partitioned solve actually needs. Paired with 128 GB of RAM and 2 × 1.88 TB NVMe SSD, it's the node for cases that genuinely scale.

Specifications

Forty-eight cores is the widest spread we offer on one socket, and twelve memory channels keep them supplied rather than queuing for data — the ratio that decides whether a high core count actually converts into speed. The 4.1 GHz figure is the all-core boost, held with every rank loaded, not a single-core peak.

Sizing note: cores vs. memory on this node

SC107 carries the same 128 GB of RAM as the lower-core tiers, spread across four times as many cores. That works out to under 3 GB per core — fine for the meshes 128 GB can hold, but worth understanding: this node's strength is throughput on cases that fit in 128 GB, not capacity. A case big enough to keep 48 ranks genuinely busy while fitting in 128 GB is the sweet spot; if your mesh needs more memory than that, more cores won't help and our high-memory nodes are the right tool.

What this node is built for

SC107 is the maximum-throughput tier: the widest parallel spread with the bandwidth and cache to support it. It suits:

• Cases that scale well past 32 cores — 48 MPI ranks split a mesh finer than anything else here, and with real bandwidth behind 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 and would choke a narrower memory bus.

• Large parameter sweeps and design studies — 48 cores means substantially more work per hour, so you get through more design iterations per day.

• Long sustained runs — an all-core clock held under full load matters more across a multi-day transient than a peak that only appears on one core.

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

Choosing SC107

Choose SC107 when:

• Your case has the cell count to keep 48 partitions genuinely busy, and

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

The caveat matters more here than on any other node. Forty-eight ranks is a lot of partitions, and Fluent scales well only while each core still has enough cells to stay busy relative to the boundary data it exchanges with its neighbors. Split a moderate case 48 ways and the cores can spend more time communicating than solving — at which point a 16- or 24-core node, or even a high-clock 8-core tier, finishes first. This is the tier where right-sizing matters most, and it's worth checking your case's scaling before you book rather than assuming more cores is better.

As a rough guide, Fluent needs about 1 GB of RAM per million cells for a typical steady, single-phase case. With 128 GB, SC107 handles meshes into the tens to low hundreds of millions of cells — which at 48 ranks is enough cell count per core for good scaling on the larger end of that range.

Tell us your cell count, physics, and how your case scales with core count, and we'll point you to the right tier.

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 48 cores (using algorithms such as METIS), each core solving its share of cells and exchanging boundary data each iteration via MPI. On a single node, that communication runs through fast on-board channels rather than a network, keeping parallel overhead low — a real advantage over spreading the same 48 ranks across multiple networked machines. Between exchanges, each core pulls cell data from memory, and with twelve channels and 256 MB of shared L3, SC107's cores keep working rather than stalling as the partition count rises.

Why rent instead of buy

A 48-core server is a serious purchase for work that may only need it occasionally. 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

SC107 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 SC107 node.

2. Set up your simulation — load your geometry and case, configure your solver settings, and run ANSYS Fluent with all 48 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 SC107 different from our 32-core nodes? Half again as many cores on the same 12-channel platform with the same 256 MB cache, at a slightly lower all-core clock. If your case scales efficiently past 32 ranks, SC107 finishes it sooner; if it doesn't, the extra cores sit idle and a 32-core tier is the better value.

Will 48 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 48 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 48 cores? For cases that fit in 128 GB, yes — that's 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 for that.

Is 4.1 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. SC107's 4.1 GHz is the all-core number, held with all 48 cores working — which is what a real parallel solve sees.

How large a mesh can it handle? With 128 GB of RAM, plan for meshes into the tens to low hundreds of millions of cells for simpler steady single-phase cases, fewer for transient or multiphase physics (about 1 GB of RAM per million cells is a good starting estimate). Send us your case for a precise answer.

Is my data kept confidential? Yes.

Get started with SC107

If your cases genuinely scale across many cores and you want the widest parallel spread and fullest memory bandwidth we offer, SC107 is built for exactly that.

Contact MR CFD to run your ANSYS Fluent case on SC107 — and we'll help you confirm your case can use all 48 cores before you book.