# Welcome

## Introduction to Strong Compute

Our vision is to be the command and control for compute—what Kubernetes/VMware should be if they were great for AI. So we've built the Strong Compute Instant Super Computer (ISC).

## Instant Super Computer

Strong Compute’s ISC is designed for efficient AI model training, packaged into the [ISC Control Plane](https://docs.strongcompute.com/getting-started/1.-access-strong-compute) (website) and [ISC Portal](https://docs.strongcompute.com/basic-concepts/isc-commands-cli) (terminal). 

Access interactive [Container](https://docs.strongcompute.com/basic-concepts/workstations-images-and-containers) environments, and launch [Experiments](https://docs.strongcompute.com/basic-concepts/launching-experiments) for cluster-scale training. You bring the project, we bring the compute.

## Get started

New users of Strong Compute can follow the [Guided Onboarding](https://docs.strongcompute.com/getting-started/in-app-onboarding) checklist in the ISC Control Plane, or follow the [Getting Started](https://docs.strongcompute.com/getting-started/1.-access-strong-compute) guide in these docs to get started.

#### Academic research grantees

Strong Compute research grantees also need to set up [VPN](https://docs.strongcompute.com/getting-started/vpn-sydney-cluster-only) access to the Strong Compute private cluster.

## Benefits of Strong Compute

* **Multi-Cloud Backend:** Large range of compute nodes with H100, A100, A6000, L40, L4, and other GPU types in 1x, 4x, and 8x GPU shapes.
* **Large cluster training:** Training clusters with up to 48 GPUs.
* **Robust Checkpointing:** Easy to use tools for atomic checkpointing and assisted training resumption.
* **Lightning Fast Dataset Transfer:** Achieves 30GB/second (target 100GB/sec) moving containers, datasets, and training artifacts on and off training clusters.
* **Cost Control UI:** Helps tightly scope and manage expenses.