Scientific computing · Deep SIML Labs

Simulate more.
Discover faster.

We develop GPU methods that fit larger biological simulations into a fixed memory budget. Our research explores how simulation and optimization can guide the next experiment in biology and materials.

Research publications →
Cellular signaling benchmarkHow many cells fit in 6 GB?
SIML CUDA
8,100,736
Triton
6,022,368

MCF-7 signaling model · One RTX 3060
Six seconds of simulated time

See the experiment

Demonstrated results

What we measured on the GPU.

One study measures interacting cellular models. The other measures binary lattice updates. Each uses its own baseline and answers a different question.

Biological Simulation

GPU capacity for cellular signaling

A GPU can run out of memory before a model reaches the population a researcher wants to study. On one RTX 3060, SIML CUDA advanced 8,100,736 interacting MCF-7 signaling patches through six simulated seconds. Triton advanced 6,022,368 under the same 6 GB budget and accuracy requirements.

Computational Systems

GPU updates for binary lattices

On one RTX 3060, SIML updated 256 small binary lattices 12.08× as fast as a separately tuned UInt8 CUDA control once the data were on the GPU. A separate capacity test kept one connected lattice of 84.26 billion binary variables updating continuously for ten minutes.

Our technology

From simulation to experiment.

SIML StudioInvited beta for reviewing compact enum mappings and exporting code.Explore Studio →

How we work

A result should guide the next test.

  1. 01Question

    Define the constraint and the outcome that would matter.

  2. 02Model

    Specify the system, its assumptions, and a reference.

  3. 03Simulate

    Advance the model under a declared compute budget.

  4. 04Compare

    Match the workload and score both implementations.

  5. 05Optimize

    Refine the representation or the execution method.

  6. 06Validate

    Check numerical agreement. Physical validation of proposed materials is planned.

Technical evaluations

What limits your simulation?

These are potential joint evaluations. Bring a workload, its current implementation, and the outcome that would make a difference to your team.

GPU memory
How many states or interacting cells can each implementation advance under the same memory cap and accuracy requirement?
Kernel throughput
Can a specialized representation improve resident updates once setup, transfers, and end-to-end runtime are measured separately?
Model selection
Which numerical method or candidate configuration deserves a longer simulation or a physical experiment?
Discuss a Technical Evaluation

Partnerships

Work with the lab.

We welcome research groups and scientific computing teams interested in comparing methods, extending a model, or designing a joint validation study.

Discuss a research collaboration