SC87

SC87 — 86-Core Granite Rapids CFD Node with Configurable Memory and Storage for Large-Scale ANSYS Fluent

The SC87 node is our flagship latest-generation single-socket platform for ANSYS Fluent, built on an 86-core Intel® Xeon® 6787P (Granite Rapids) with configurable memory and NVMe storage. It combines an exceptional core count with the wide bandwidth of the modern platform, and lets you choose the balance that fits your work — from 256 to 512 GB of fast DDR5 ECC RAM, and from 2 to 4 datacenter NVMe drives (960 GB to 1.92 TB each). When you want maximum modern parallelism, tailored to large, demanding cases, SC87 is the choice.

It is the flagship, high-core, configurable single-socket tier: the highest core count on our modern platform, a generous configurable memory range, and the wide bandwidth of Granite Rapids.

Specifications

SC87 uses a flagship Xeon 6787P on Intel's Granite Rapids platform: 86 cores at 2.0 GHz, with the wide multi-channel DDR5 bandwidth and PCIe connectivity of the current generation. Memory and storage are configurable — from 256 to 512 GB of DDR5 ECC RAM, and from 2 to 4 NVMe datacenter drives — so the node can be matched to your workload rather than fixed to one balance.

What's included with SC87

Every SC87 rental comes with extras to help you get the most from it:

• A comprehensive CFD course, free — a full training course to sharpen your simulation skills, included at no extra cost.

• One month of Claude AI Premium — AI assistance to support your engineering, scripting, and problem-solving.

• One hour of free CFD consultation — direct time with our CFD engineers to help set up, troubleshoot, or optimize your case.

These come standard with the node — practical support alongside the compute.

What this node is built for

SC87 is the flagship, high-core, configurable single-socket tier. It suits:

• Very large meshes solved fast — 86 modern Granite Rapids cores spread the work for an exceptional parallel speed-up.

• Demanding production CFD — when a large case needs both strong parallelism and fast turnaround.

• Heavy parameter studies and high-fidelity models — the core count and memory to run bigger, more detailed simulations.

• I/O-heavy transient runs — configurable NVMe (up to 4 drives) absorbs frequent timestep writes and large datasets.

With 86 modern cores and wide DDR5 bandwidth, Fluent partitions your mesh across all of them for a strong parallel speed-up, while your chosen memory and NVMe configuration keeps large cases in memory and the data flowing.

Configuring SC87: memory and storage

SC87's defining features are its core count and its flexibility. In CFD, cores determine how fast a case solves, memory sets the largest mesh you can run — roughly 1 GB of RAM per million cells for a typical steady, single-phase case — and storage holds your case and result data.

• Memory (256–512 GB): choose 256 GB for large cases, or 512 GB to hold very large meshes and memory-hungry physics (transient, multiphase) in memory.

• Storage (2–4 NVMe drives, 960 GB–1.92 TB each): choose fewer, smaller drives for lighter workloads, or more and larger drives for long transient runs and larger result sets.

Tell us your cell count, physics, and how much result data you expect, and we'll recommend the memory and storage configuration that fits — then match you to SC87 or another tier as appropriate. With 86 cores, SC87 is best matched to large cases that partition efficiently and use enough of that parallelism to pay off.

Fast NVMe storage

SC87 uses NVMe datacenter SSDs, which deliver markedly faster read/write than SATA. Loading large geometries and case files, writing solution data, and handling frequent autosaves all run quickly, so a fast solve is never held up by slow disk I/O. Because the drive count and capacity are configurable, you can size storage to your workload — from a responsive baseline to several terabytes for long transient simulations that write data at every timestep.

How parallel solving works on this node

When you run Fluent in parallel, the solver partitions your mesh across all 86 cores (using algorithms such as METIS), each core solving its share of cells and exchanging boundary data each iteration via MPI. On a single-socket node, that communication runs entirely on-chip rather than across sockets or a network, keeping parallel overhead low — an advantage of reaching such a high core count within one CPU. For large cases that fit in your chosen memory and partition well, those 86 modern cores deliver a substantial speed-up.

Why rent instead of buy

An 86-core, latest-generation server is a major investment to own and maintain. Renting SC87 gives you that flagship parallelism — configured to your workload — plus a free course, a month of Claude AI Premium, and a consultation hour, without the capital cost or the upkeep.

How it works — you're in full control

SC87 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 SC87 node.

2. Set up your simulation — load your geometry and case, configure your solver settings, and run ANSYS Fluent with all 86 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

What makes SC87 our flagship modern node?
It carries an exceptional 86-core count on Intel's latest Granite Rapids platform with DDR5 ECC memory — the highest core count on our modern single-socket line, for strong parallel throughput on large cases.

How do I choose the memory and storage?
Memory is configurable from 256 to 512 GB (roughly 1 GB per million cells for simpler steady cases, so 512 GB holds very large or complex meshes). Storage is configurable from 2 to 4 NVMe drives at 960 GB–1.92 TB each. Tell us your case size and result-data needs, and we'll recommend a configuration.

How large a mesh can it handle?
It depends on your chosen memory — plan around 1 GB per million cells for simpler steady single-phase cases (fewer for transient or multiphase physics). At 256 GB, expect very large meshes; at 512 GB, larger still. Send us your case for a precise answer.

Will my case use all 86 cores effectively?
It depends on size. Very high core counts pay off most on large meshes that partition efficiently; smaller cases may not scale all the way to 86 cores, where parallel overhead can eat into the gain. Send us your setup and we'll advise whether SC87 or a different tier is the better fit.

Is my data kept confidential?
Yes. Your geometry, case files, and results stay private to you, treated as strictly confidential — and because you run everything yourself over remote desktop, your work stays under your control throughout.

Get started with SC87

If you want flagship modern parallelism for large CFD cases — with a free CFD course, a month of Claude AI Premium, and a consultation hour included — SC87's 86-core Granite Rapids platform is built for exactly that.

Contact MR CFD for remote-desktop access to SC87 — and we'll help you configure the right memory and storage for your simulation.