Compute icons

The 29 compute icons in Cloud, described

Autoscaleautoscale

Autoscale — more instances when there is more to do, fewer when there is less.

Bare metalbare-metal

Bare metal — an actual physical server, dedicated rather than a share of one.

Cloudcloud

The cloud — hosted infrastructure on somebody else's computers, and a great many of them.

Cloud addcloud-add

Cloud add — connect a new cloud account or provision a new cloud resource.

Cloud alertcloud-alert

Cloud alert — something is wrong up in the cloud, an error or a degraded service.

Cloud backupcloud-backup

A cloud with an arrow pointing back on it — a backup kept in the cloud, ready to restore.

Cloud checkcloud-check

Cloud check — the cloud side is healthy, connected and in sync.

Cloud configcloud-config

A cloud with sliders on it — cloud settings, the account and region configured.

Cloud offcloud-off

Cloud off — disconnected from the cloud, working offline or locally.

Cloud synccloud-sync

A cloud with a sync loop under it — keep this copy and the cloud copy the same.

Computecompute

Compute — processing cycles for hire, CPU and GPU instances that do the work.

CPUcpu

The CPU, the processor where general computation runs before work is handed to accelerators.

GPUgpu

A GPU — the graphics card or accelerator where the heavy matrix maths of AI actually runs.

GPU clustergpu-cluster

A GPU cluster — many accelerators working together as one training fleet.

Hybrid cloudhybrid-cloud

Hybrid cloud — some of it on your own premises, some of it in a provider's cloud.

Memorymemory-ram

Memory — the RAM modules that decide how much a machine can hold at once.

Multi-cloudmulti-cloud

Multi-cloud — more than one provider at once, workloads across several clouds.

Cluster namespacenamespace-cluster

A cluster namespace — one name that spans many machines in a fleet.

Nodenode

A node — one machine, instance or member in a cluster of many.

Node clusternode-cluster

A node cluster — machines grouped together to act as one system.

Private cloudprivate-cloud

A private cloud — cloud infrastructure that is yours alone, walled and dedicated.

Provisionprovision

Provision — make the resource exist, spin it up and allocate it.

Rackrack

A rack — machines stacked in a datacentre cabinet, one above the other.

Scale downscale-down

Scale down — give it less capacity, shrink to smaller or fewer instances.

Scale upscale-up

Scale up — give it more capacity, grow to bigger or more instances.

Serverserver

A server — the machine that answers requests, the backend host.

TPUtpu

A TPU — a chip grown for tensors, a matrix accelerator built for training.

Virtual machinevm

A virtual machine — a whole computer that is really software running on a host.

Warm poolwarm-pool

A warm pool — instances started early and kept on standby so nobody waits.

In code, each is one import — import { Autoscale } from "@iconmind/react/icons/autoscale" — and the same name in Vue, Svelte, Solid, Preact, React Native, Astro, Blade and Flutter.

Narrow it by tag