SC116 — 16-Core, Large-Cache CFD HPC Node with 128 GB Memory for ANSYS Fluent

The SC116 node runs a single AMD Ryzen 9 9950X3D — 16 cores and 32 threads, boosting up to 5.7 GHz, backed by 128 MB of total L3 cache. It brings the same headline numbers as our top single-socket tier — 16 cores, a 5.7 GHz boost, 128 MB of cache — in a chip built from AMD's gaming-focused 3D V-Cache lineup rather than its workstation-class parts. Paired with 128 GB of RAM and 2 × 1.88 TB NVMe SSD, it's a strong all-around option for cases that want plenty of cores, a fast clock, and a large cache in one node.

Specifications

One detail worth knowing: the 9950X3D's 128 MB of cache is split unevenly across its two 8-core chiplets — one carries 96 MB of stacked 3D V-Cache, the other a standard 32 MB. Each chiplet can also run at a slightly different effective clock. On Windows, AMD's chipset driver steers latency-sensitive threads toward the cache-heavy chiplet; on a Linux HPC node, getting the same benefit for an MPI job typically means setting process affinity deliberately rather than relying on automatic scheduling.

What this node is built for

SC116 is a broad, well-rounded tier: plenty of cores, a fast clock, and a large total cache. It suits:

• Cases that want to scale across more cores without giving up per-core speed — 16 cores at up to 5.7 GHz outruns a lower-clocked many-core node on cases that fit this range.

• Large meshes with sizeable per-core working sets — 128 MB of total cache gives real headroom to keep active data on-chip, especially for the cores landing on the V-Cache chiplet.

• Parameter sweeps and design studies where turnaround matters most — more cores running fast means more iterations completed per hour.

• UDF-heavy and serial-bottlenecked solves — a high per-core clock handles the parts of a case that don't parallelize well.

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

Cores, clock speed & cache: choosing SC116

SC116 gives you 16 cores, a 5.7 GHz boost clock, and 128 MB of cache in one node — a strong combination for cases that don't fit neatly into a smaller high-clock tier or a lower-clocked many-core node. Keep in mind the asymmetric cache layout: a case that runs best on 8 or fewer cores can be pinned to the V-Cache chiplet for the full cache benefit, while a case that spreads across all 16 cores will have half its processes on the smaller-cache chiplet. If your workload is a single mesh partitioned evenly across all 16 cores, expect performance between what you'd see from a uniform 96 MB and a uniform 32 MB cache, rather than the full 128 MB working for every core at once.

As a rough guide, Fluent needs about 1 GB of RAM per million cells for a typical steady, single-phase case, so with 128 GB, SC116 comfortably covers meshes well beyond what a 32 GB node can hold.

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. That means faster loading of large case and data files, quicker autosaves, and room to keep multiple case variants or 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 the available cores (using algorithms such as METIS), each core solving its share of cells and exchanging boundary data each iteration via MPI. On SC116, that partitioning crosses two chiplets with different cache sizes, so how the job is pinned to cores can affect how evenly the work finishes — something we can help configure when we set up your run.

Why rent instead of buy

For engineers who want a high core count, a fast clock, and a large cache in one node, SC116 delivers all three — and you can still move to a many-core or workstation-class node when a job calls for it.

How it works — you're in full control

SC116 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 SC116 node.

2. Set up your simulation — load your geometry and case, configure your solver settings, and run ANSYS Fluent with all 16 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 SC116 different from our other 16-core node? Same headline numbers — 16 cores, up to 5.7 GHz, 128 MB of cache — but SC116's chip splits that cache unevenly across two chiplets and is built on a consumer platform (no ECC memory support). Our workstation-class 16-core tier carries its cache more evenly across all cores and supports ECC.

Does the split cache actually hurt performance? Only for jobs spread evenly across all 16 cores, where half the processes land on the smaller-cache chiplet. For cases that run well on 8 or fewer cores, pinning the job to the V-Cache chiplet gets the full cache benefit.

How large a mesh can it handle? With 128 GB of RAM, plan for meshes well 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.

What if I need even more cores or memory than this? We offer many-core and high-memory nodes for cases that scale past 16 cores or need more than 128 GB — we'll recommend the right tier, or a multi-node setup, for your specific case.

Is my data kept confidential? Yes.

Get started with SC116

If your cases want a high core count, a fast clock, and a large cache in one node, SC116 is built for exactly that.

Contact MR CFD to run your ANSYS Fluent case on SC116 — and we'll help you choose between clock speed, cache, core count, and memory for your simulation.sc109