Skip to main content
Home
Contact us
Watch Demo
Contact us
Watch Demo
Domino's logo

Who is Domino?

Domino Data Lab empowers the largest AI-driven enterprises to build and operate AI at scale. Domino’s Enterprise AI Platform provides an integrated experience encompassing model development, MLOps, collaboration, and governance. With Domino, global enterprises can develop better medicines, grow more productive crops, develop more competitive products, and more. Founded in 2013, Domino is backed by Sequoia Capital, Coatue Management, NVIDIA, Snowflake, and other leading investors.

Watch Demo

  • Platform

      • AI infrastructure
      • Data management
      • AI workbench
      • MLOps
      • AI governance
      • FinOps
      • Pricing
      • Security & compliance
      • What's new

  • Solutions

    • Industries

      • Life sciences
      • Finance
      • Public sector
      • Retail
      • Manufacturing

    • Use Cases

      • Generative AI
      • Cost-effective data science
      • Self-service data science
      • Model risk management
      • Cloud data science

  • Learn

      • Events
      • Blog
      • Podcast
      • Courses and certifications
      • Data Science Dictionary
      • Documentation
      • Support
      • Demo hub

  • Company

      • About
      • Why Domino
      • Careers
      • News and press
      • Partners
      • Customers
      • Contact us

© 2026 Domino Data Lab, Inc. Made in San Francisco.

  • Do not sell my personal information
  • Privacy policy
  • Terms and conditions
  • Security
  • Legal
HPC background

HPC and AI, orchestrated, optimized, and governed together

Built for data science, AI, and IT leaders running HPC and AI side by side. Keep the Slurm tools your teams already use. Stop paying twice for infrastructure that could be one system.
Watch demo
Talk to an expert

HPC solved scheduling decades ago. It never got a system of record.

No system of record

Researchers run massive Slurm simulations on isolated clusters with no audit trail and no way to reproduce a result.

No shared control plane

HPC and AI/ML run in separate worlds, through different tools. Every hand-off means manual coordination.

Resources optimized in isolation

Neither side can see the other's demand, so each cluster is sized for its own peak.

Everything HPC teams need, without leaving Domino

Slurm terminal

Native Slurm integration

Domino runs with real Slurm and Slinky, the NVIDIA-backed workload manager. Submit jobs using your existing scripts, with no rewrites. Multiple languages, including Mirai and ClusterMQ are supported out of the box. Domino Nexus extends this on-premises, for full hybrid and multi-cloud orchestration from one control plane.

Slurm jobs

Governed by design

Every Slurm job's accounting data flows into Domino's system of record. That means one audit trail, one Jobs monitoring view, and everything needed to reproduce a result. HPC stops being the ungoverned exception in an otherwise governed pipeline.

Slurm set up

Autoscaled, right-sized compute

Ephemeral nodesets can span multiple hardware types, including CPU, BigMem, and GPU. Admins set the scheduling policy and per-project limits once. Clusters exist only for the duration of the job, so you stop paying for a cluster that's either jammed or sitting idle.

See it on slurm
450x speedup
AI surrogate models cut a 12-day simulation down to 22sec with no loss in accuracy. [1]
5.5% → 1%
NVIDIA cut its own GPU waste, freeing compute for the workloads it actually needed. [2]
Over 70%
of researchers have failed to reproduce another scientist's results, and more than half couldn't reproduce their own. [3]

"By 2030, over 80% of all large enterprises will have executed a hybrid deployment model for AI workloads, up from less than 10% in 2026."

Gartner, Hype Cycle for Hybrid AI Infrastructure, 2026

Dennis Smith, 28 May 2026

"By 2028, more than 50% of enterprises will use public cloud selectively for bursty or experimental workloads while anchoring performance- and sovereignty-sensitive AI workloads in private and localized environments."

Gartner, The Future of AI Infrastructure: Designing for Sovereignty, Cost Optimization, and Performance

Enrique Castera, 24 June 2026

"By 2030, data center power limits will force 75% of I&O leaders to mandate energy-aware AI workload placement."

Gartner, AI Infrastructure Guide for Power-Constrained Data Centers

Daniel Bowers and Enrique Castera, 27 June 2026

Swipe

HPC on Domino, at a glance

Scheduler

Slurm and Slinky, the NVIDIA-backed standard workload manager

Frameworks

Bash, Python, R including Mirai and ClusterMQ, Julia, MATLAB, SAS, and other languages supported with no script rewrites required

Deployment

Cloud provider-independent, or on-premises

Compute

Ephemeral hardware including CPU, BigMem, and GPU as an example, autoscaled per job

Governance

Full audit trail on launch, stop, and restart events, plus reproducibility and ingestion into Domino's system of record

Resources

Domino named Visionary in 2026 Gartner Magic Quadrant for AI Platforms for Data Science Machine Learning

Download the report

Get the full Slurm setup walkthrough, step by step

See how it works

When HPC Meets AI: Closing the Governance Gap in Converged R&D

Read the white paper

FAQ

Close the gap between HPC and AI

See how Domino brings governance to the last ungoverned layer of R&D.

Watch demoTalk to an expert

[1] Published research, ocean circulation forecasting. arXiv preprint on AI surrogates for coastal simulation
[2] "Making GPU Clusters More Efficient with NVIDIA Data Center Monitoring Tools" — NVIDIA Developer Blog.
[3] Nature. Baker, M. (2016). "1,500 scientists lift the lid on reproducibility." Nature, 533, 452–454.