Fleets¶

Fleets are groups of instances used to run dev environments, tasks, and services. Depending on the fleet configuration, instances can be interconnected clusters or standalone instances.

dstack supports two kinds of fleets:

Cloud fleets – dynamically provisioned through configured backends
SSH fleets – created using on-prem servers

Cloud fleets¶

When you call dstack apply to run a dev environment, task, or service, dstack reuses idle instances from an existing fleet. If none match the requirements, dstack creates a new cloud fleet.

For greater control over cloud fleet provisioning, create fleets explicitly using configuration files.

Apply a configuration¶

Define a fleet configuration as a YAML file in your project directory. The file must have a .dstack.yml extension (e.g. .dstack.yml or fleet.dstack.yml).

type: fleet
# The name is optional, if not specified, generated randomly
name: my-fleet

# Specify the number of instances
nodes: 2
# Uncomment to ensure instances are inter-connected
#placement: cluster

resources:
  gpu: 24GB

To create or update the fleet, pass the fleet configuration to dstack apply:

$ dstack apply -f examples/misc/fleets/.dstack.yml

Provisioning...
---> 100%

 FLEET     INSTANCE  BACKEND              GPU             PRICE    STATUS  CREATED 
 my-fleet  0         gcp (europe-west-1)  L4:24GB (spot)  $0.1624  idle    3 mins ago      
           1         gcp (europe-west-1)  L4:24GB (spot)  $0.1624  idle    3 mins ago

Once the status of instances changes to idle, they can be used by dev environments, tasks, and services.

Configuration options¶

Placement¶

To ensure instances are interconnected (e.g., for distributed tasks), set placement to cluster. This ensures all instances are provisioned in the same backend and region with optimal inter-node connectivity

AWS

dstack automatically enables the Elastic Fabric Adapter for all EFA-capable instance types . If the aws backend config has public_ips: false set, dstack enables the maximum number of interfaces supported by the instance. Otherwise, if instances have public IPs, only one EFA interface is enabled per instance due to AWS limitations.

The cluster placement is supported only for aws, azure, gcp, oci, and vultr backends.

Resources¶

When you specify a resource value like cpu or memory, you can either use an exact value (e.g. 24GB) or a range (e.g. 24GB.., or 24GB..80GB, or ..80GB).

type: fleet
# The name is optional, if not specified, generated randomly
name: my-fleet

nodes: 2

resources:
  # 200GB or more RAM
  memory: 200GB..
  # 4 GPUs from 40GB to 80GB
  gpu: 40GB..80GB:4
  # Disk size
  disk: 500GB

The gpu property allows specifying not only memory size but also GPU vendor, names and their quantity. Examples: nvidia (one NVIDIA GPU), A100 (one A100), A10G,A100 (either A10G or A100), A100:80GB (one A100 of 80GB), A100:2 (two A100), 24GB..40GB:2 (two GPUs between 24GB and 40GB), A100:40GB:2 (two A100 GPUs of 40GB).

Google Cloud TPU

To use TPUs, specify its architecture via the gpu property.

type: fleet
# The name is optional, if not specified, generated randomly
name: my-fleet

nodes: 2

resources:
  gpu: v2-8

Currently, only 8 TPU cores can be specified, supporting single TPU device workloads. Multi-TPU support is coming soon.

If you’re unsure which offers (hardware configurations) are available from the configured backends, use the dstack offer command to list them.

Blocks¶

For cloud fleets, blocks function the same way as in SSH fleets. See the Blocks section under SSH fleets for details on the blocks concept.

type: fleet

name: my-fleet

resources:
  gpu: NVIDIA:80GB:8

# Split into 4 blocks, each with 2 GPUs
blocks: 4

Idle duration¶

By default, fleet instances stay idle for 3 days and can be reused within that time. If the fleet is not reused within this period, it is automatically terminated.

To change the default idle duration, set idle_duration in the run configuration (e.g., 0s, 1m, or off for unlimited).

type: fleet
# The name is optional, if not specified, generated randomly
name: my-fleet

nodes: 2

# Terminate instances idle for more than 1 hour
idle_duration: 1h

resources:
  gpu: 24GB

Spot policy¶

By default, dstack uses on-demand instances. However, you can change that via the spot_policy property. It accepts spot, on-demand, and auto.

Retry policy¶

By default, if dstack fails to provision an instance or an instance is interrupted, no retry is attempted.

If you'd like dstack to do it, configure the retry property accordingly:

type: fleet
# The name is optional, if not specified, generated randomly
name: my-fleet

nodes: 1

resources:
  gpu: 24GB

retry:
  # Retry on specific events
  on_events: [no-capacity, interruption]
  # Retry for up to 1 hour
  duration: 1h

Cloud fleets are supported by all backends except kubernetes, vastai, and runpod.

Reference

Cloud fleets support many more configuration options, incl. backends, regions, max_price, and among others.

SSH fleets¶

If you have a group of on-prem servers accessible via SSH, you can create an SSH fleet.

Apply a configuration¶

Define a fleet configuration as a YAML file in your project directory. The file must have a .dstack.yml extension (e.g. .dstack.yml or fleet.dstack.yml).

type: fleet
# The name is optional, if not specified, generated randomly
name: my-fleet

# Uncomment if instances are interconnected
#placement: cluster

# SSH credentials for the on-prem servers
ssh_config:
  user: ubuntu
  identity_file: ~/.ssh/id_rsa
  hosts:
    - 3.255.177.51
    - 3.255.177.52

Requirements

1. Hosts should be pre-installed with Docker.

NVIDIAAMDIntel GaudiTenstorrent

2. Hosts with NVIDIA GPUs should also be pre-installed with CUDA 12.1 and NVIDIA Container Toolkit .

2. Hosts with AMD GPUs should also be pre-installed with AMDGPU-DKMS kernel driver (e.g. via native package manager or AMDGPU installer .)

2. Hosts with Intel Gaudi accelerators should be pre-installed with Gaudi software and drivers. This should include the drivers, hl-smi, and Habana Container Runtime.

2. Hosts with Tenstorrent accelerators should be pre-installed with Tenstorrent software. This should include the drivers, tt-smi, and HugePages.

3. The user specified should have passwordless sudo access.

To create or update the fleet, pass the fleet configuration to dstack apply:

$ dstack apply -f examples/misc/fleets/.dstack.yml

Provisioning...
---> 100%

 FLEET     INSTANCE  GPU             PRICE  STATUS  CREATED 
 my-fleet  0         L4:24GB (spot)  $0     idle    3 mins ago      
           1         L4:24GB (spot)  $0     idle    3 mins ago

When you apply, dstack connects to the specified hosts using the provided SSH credentials, installs the dependencies, and configures these hosts as a fleet.

Once the status of instances changes to idle, they can be used by dev environments, tasks, and services.

Configuration options¶

Placement¶

If the hosts are interconnected (i.e. share the same network), set placement to cluster. This is required if you'd like to use the fleet for distributed tasks.

Network¶

By default, dstack automatically detects the network shared by the hosts. However, it's possible to configure it explicitly via the network property.

Blocks¶

By default, a single job occupies the entire instance, so if the instance has 8 GPUs, the job will use all of them.

To make it more efficient, you can set the blocks property to specify how many blocks you’d like the instance to be divided into, allowing multiple jobs to use these blocks concurrently.

type: fleet
name: my-fleet

ssh_config:
  user: ubuntu
  identity_file: ~/.ssh/id_rsa
  hosts:
    - hostname: 3.255.177.51
      blocks: 4
    - hostaname: 3.255.177.52
      # As many as possible, according to numbers of GPUs and CPUs
      blocks: auto
    - hostaname: 3.255.177.53
      # Do not sclice. This is the default value, may be omitted
      blocks: 1

For instance, with 8 GPUs, 128 CPUs, and 2TB of memory, setting blocks to 8 would assign 1 GPU, 16 CPUs, and 256 GB of memory to each block. These blocks can be used concurrently, and a single job can occupy multiple blocks if needed.

GPUs and CPUs must be divisible by the number of blocks. All resources (GPU, CPU, memory) are split proportionally, except disk storage, which is shared.

You can also set blocks to auto, which automatically sets the number of blocks to match the number of GPUs.

Environment variables¶

If needed, you can specify environment variables that will be used by dstack-shim and passed to containers.

For example, these variables can be used to configure a proxy:

type: fleet
name: my-fleet

env:
  - HTTP_PROXY=http://proxy.example.com:80
  - HTTPS_PROXY=http://proxy.example.com:80
  - NO_PROXY=localhost,127.0.0.1

ssh_config:
  user: ubuntu
  identity_file: ~/.ssh/id_rsa
  hosts:
    - 3.255.177.51
    - 3.255.177.52

Proxy jump¶

If fleet hosts are behind a head node (aka "login node"), configure proxy_jump:

type: fleet
name: my-fleet

ssh_config:
  user: ubuntu
  identity_file: ~/.ssh/worker_node_key
  hosts:
    - 3.255.177.51
    - 3.255.177.52
  proxy_jump:
    hostname: 3.255.177.50
    user: ubuntu
    identity_file: ~/.ssh/head_node_key

To be able to attach to runs, both explicitly with dstack attach and implicitly with dstack apply, you must either add a front node key (~/.ssh/head_node_key) to an SSH agent or configure a key path in ~/.ssh/config:

Host 3.255.177.50
    IdentityFile ~/.ssh/head_node_key

where Host must match ssh_config.proxy_jump.hostname or ssh_config.hosts[n].proxy_jump.hostname if you configure head nodes on a per-worker basis.

Reference

For all SSH fleet configuration options, refer to the reference.

Troubleshooting¶

Resources

Once the fleet is created, double-check that the GPU, memory, and disk are detected correctly.

If the status does not change to idle after a few minutes or the resources are not displayed correctly, ensure that all host requirements are satisfied.

If the requirements are met but the fleet still fails to be created correctly, check the logs at /root/.dstack/shim.log on the hosts for error details.

Manage fleets¶

List fleets¶

The dstack fleet command lists fleet instances and their status:

$ dstack fleet

 FLEET     INSTANCE  BACKEND              GPU             PRICE    STATUS  CREATED 
 my-fleet  0         gcp (europe-west-1)  L4:24GB (spot)  $0.1624  idle    3 mins ago      
           1         gcp (europe-west-1)  L4:24GB (spot)  $0.1624  idle    3 mins ago

Delete fleets¶

When a fleet isn't used by a run, you can delete it by passing the fleet configuration to dstack delete:

$ dstack delete -f cluster.dstack.yaml
Delete the fleet my-gcp-fleet? [y/n]: y
Fleet my-gcp-fleet deleted

Alternatively, you can delete a fleet by passing the fleet name to dstack fleet delete. To terminate and delete specific instances from a fleet, pass -i INSTANCE_NUM.

What's next?

Read about dev environments, tasks, and services