tw compute-envs

Manage compute environments.

Compute environments define the execution platform where a pipeline runs. A compute environment is composed of the credentials, configuration, and storage options related to a particular computing platform. See Compute environments for more information on supported providers.

Run tw compute-envs -h to view the list of supported compute environment operations.

tw compute-envs add

Add a new compute environment.

tw compute-envs add

Run tw compute-envs add -h to view the list of supported providers.

Run tw compute-envs add <platform> -h to view the required and optional fields for your provider.

You must add the credentials for your provider before creating your compute environment.

Command:

tw compute-envs add aws-batch forge --name=my_aws_ce \
--credentials=<my_aws_creds_1> --region=eu-west-1 --max-cpus=256 \
--work-dir=s3://<bucket name> --wait=AVAILABLE

Example output:

New AWS-BATCH compute environment 'my_aws_ce' added at user workspace

This command will:

• Use Batch Forge to automatically manage the AWS Batch resource lifecycle (forge)
• Use the credentials previously added to the workspace (--credentials)
• Create the required AWS Batch resources in the AWS Ireland (eu-west-1) region
• Provision a maximum of 256 CPUs in the compute environment (--max-cpus)
• Use an existing S3 bucket to store the Nextflow work directory (--work-dir)
• Wait until the compute environment has been successfully created and is ready to use (--wait)

See the compute environment page for your provider for detailed information on Batch Forge and manual compute environment creation.

tw compute-envs add k8s

Add new Kubernetes compute environment.

tw compute-envs add k8s [OPTIONS]

Options

Option	Description	Required	Default
-n, --name	Compute environment name.	Yes
-w, --workspace	Workspace numeric identifier or reference in OrganizationName/WorkspaceName format (defaults to TOWER_WORKSPACE_ID environment variable)	No
-c, --credentials	Credentials identifier [default: workspace credentials].	No
--wait	Wait until given status or fail. Valid options: CREATING, AVAILABLE, DISABLED, DELETING, ERRORED, INVALID, DELETED.	No
-d, --description	Compute environment description.	No
--labels	Comma-separated list of labels.	No
--pre-run	Pre-run script.	No
--post-run	Post-run script.	No
--nextflow-config	Nextflow config	No
-e, --env	Add environment variables. By default are only added to the Nextflow head job process, if you want to add them to the process task prefix the name with 'compute:' or 'both:' if you want to make it available to both locations.	No
--storage-mount	Mount path for the PersistentVolumeClaim. Directory where the storage is mounted in containers. If absent, Platform defaults to /scratch.	No
--compute-account	Kubernetes service account for Nextflow-submitted pipeline jobs. Controls permissions for individual task pods. If absent, Platform defaults to default.	No
--pod-cleanup	Pod cleanup policy after job completion. ON_SUCCESS removes pods only on success. ALWAYS removes all pods. NEVER keeps all pods.	No
--head-pod-spec	Custom PodSpec YAML for the Nextflow head job pod. Provide path to YAML file with custom pod configuration.	No
--service-pod-spec	Custom PodSpec YAML for the compute environment service pod. Provide path to YAML file with custom pod configuration.	No
--work-dir	Nextflow work directory. Path where workflow intermediate files are stored on the shared storage.	Yes
-s, --server	Kubernetes control plane URL. The API server endpoint for cluster communication (e.g., https://your-k8s-host.com).	Yes
--namespace	Kubernetes namespace for workflow execution. Isolates resources within the cluster.	Yes
--ssl-cert	SSL certificate to authenticate the connection. Provide path to certificate file for secure cluster communication.	Yes
--head-account	Kubernetes service account for connecting to the cluster. Used by the Nextflow head job to authenticate with the Kubernetes API.	Yes
--storage-claim	PersistentVolumeClaim name for scratch storage. Must support ReadWriteMany access mode for shared workflow data.	Yes

tw compute-envs add aws-batch

Add new AWS Batch compute environment.

tw compute-envs add aws-batch

Use --secrets-kms-key to encrypt the temporary AWS Secrets Manager secrets that Platform creates for runs that use pipeline secrets with a customer-managed KMS key. Pass a key ARN or a key ID. When omitted, Platform uses the AWS-managed default key. The compute environment credentials and the execution role need additional KMS permissions on the key. See Pipeline secrets for the required policy statements.

tw compute-envs add aws-batch forge

Add new AWS Batch compute environment with automatic provisioning of compute resources.

tw compute-envs add aws-batch forge [OPTIONS]

Options

Option	Description	Required	Default
-n, --name	Compute environment name.	Yes
-w, --workspace	Workspace numeric identifier or reference in OrganizationName/WorkspaceName format (defaults to TOWER_WORKSPACE_ID environment variable)	No
-c, --credentials	Credentials identifier [default: workspace credentials].	No
--wait	Wait until given status or fail. Valid options: CREATING, AVAILABLE, DISABLED, DELETING, ERRORED, INVALID, DELETED.	No
-d, --description	Compute environment description.	No
--labels	Comma-separated list of labels.	No
--pre-run	Pre-run script.	No
--post-run	Post-run script.	No
--nextflow-config	Nextflow config	No
-e, --env	Add environment variables. By default are only added to the Nextflow head job process, if you want to add them to the process task prefix the name with 'compute:' or 'both:' if you want to make it available to both locations.	No
--create-efs	A OneZone EFS without backup will be created. EC2 instances can run on a different zone and inter-region transfer fees will be billed. If you want to remove transfer costs, restrict to only one subnet at advanced options.	No
--efs-id	Enter the EFS file system id e.g. fs-0123456789.	No
--efs-mount	Enter the EFS mount path [default: pipeline work directory].	No
--fsx-size	Enter the FSx storage capacity in GB (minimum 1,200 GB or increments of 2,400 GB).	No
--fsx-dns	Enter the FSx file system DNS name e.g. 'fs-0123456789.fsx.eu-west-1.amazonaws.com'.	No
--fsx-mount	Enter the FSx mount path [default: pipeline work directory].	No
--instance-types	EC2 instance types for compute resources. Comma-separated list of instance families or types. Use 'optimal' for automatic selection of M4, C4, and R4 instances.	No
--alloc-strategy	Instance allocation strategy. Controls how AWS Batch launches instances. BEST_FIT_PROGRESSIVE recommended for On-Demand. SPOT_CAPACITY_OPTIMIZED recommended for Spot instances.	No
--vpc-id	VPC identifier. The Virtual Private Cloud where compute resources will be deployed.	No
--subnets	VPC subnets for compute resources. Comma-separated list of subnet IDs for network isolation and internet access control.	No
--security-groups	Security group IDs for network access control. Comma-separated list defining firewall rules for EC2 compute nodes.	No
--ami-id	Custom AMI identifier. Must be AWS Linux 2 ECS-optimized image meeting compute resource specifications. If absent, Platform defaults to latest approved Amazon ECS-optimized AMI.	No
--key-pair	EC2 key pair name for SSH access. Enables remote access to compute nodes for debugging and maintenance.	No
--min-cpus	Minimum CPUs to keep provisioned. These CPUs remain active continuously and incur costs regardless of workload activity. If absent, Platform defaults to 0.	No
--boot-disk-size	Boot disk size in GB. Controls the root volume size for EC2 instances. If absent, Platform defaults to 50 GB.	No
--head-job-cpus	Number of CPUs allocated to the Nextflow head job. Controls the compute resources for the main workflow orchestration process.	No
--head-job-memory	Memory allocation for the Nextflow head job in megabytes. Determines available memory for workflow orchestration.	No
--head-job-role	IAM role ARN to grant fine-grained permissions to the Nextflow head job. Enables secure access to AWS resources.	No
--compute-job-role	IAM role ARN to grant fine-grained permissions to Nextflow compute jobs. Controls access for individual pipeline tasks.	No
--batch-execution-role	IAM role ARN for ECS task execution. Grants Amazon ECS containers permission to make AWS API calls on your behalf.	No
--ebs-blocksize	Initial EBS auto-expandable volume size in GB. Additional blocks of this size are added automatically when storage runs low. If absent, Platform defaults to 50 GB.	No
--bid-percentage	Maximum Spot instance price as percentage of On-Demand price. Controls cost ceiling for Spot instances. You pay the market price up to this maximum. If absent, Platform defaults to 100%.	No
--cli-path	AWS CLI installation path on EC2 instances. Specify custom path if AWS CLI is installed in non-standard location.	No
--secrets-kms-key	Customer-managed KMS key used to encrypt the temporary Secrets Manager secrets created for runs that use pipeline secrets. Accepts a key ARN or a key id. When omitted, the AWS-managed default Secrets Manager key is used.	No
--work-dir	Nextflow work directory. Path where workflow intermediate files are stored. Must be an S3 bucket path (e.g., s3://your-bucket/work).	Yes
-r, --region	AWS region where compute resources will be created (e.g., us-east-1, eu-west-1).	Yes
--max-cpus	Maximum CPUs provisioned by Batch Forge. Defines the upper limit for auto-scaling compute capacity.	Yes
--provisioning-model	Instance provisioning model. EC2 uses on-demand instances for reliability. SPOT uses interruptible instances for cost savings. Default: SPOT.	No	SPOT
--no-ebs-auto-scale	Disable EBS auto-expandable disk provisioning. When disabled, instances use fixed-size storage volumes.	No
--fusion	DEPRECATED - Use '--fusion-v2' instead.	No
--fusion-v2	Enable Fusion file system. Provides native access to S3 storage with low-latency I/O. Requires Wave containers.	No
--wave	Enable Wave containers. Allows access to private container repositories and on-demand container provisioning.	No
--fast-storage	Enable NVMe instance storage. Provides high-performance local storage for faster I/O operations. Requires Fusion file system.	No
--snapshots	Enable Fusion Snapshots. Automatically restores jobs interrupted by spot instance reclamation. Requires Fusion file system.	No
--fargate	Run Nextflow head job on Fargate. Enables serverless container execution for the orchestration process. Requires Fusion v2 and Spot provisioning model.	No
--gpu	Enable GPU instances. Provisions GPU-enabled EC2 instances for compute-intensive workloads requiring hardware acceleration.	No
--allow-buckets	Additional S3 buckets for read-write access. Comma-separated list of S3 bucket paths beyond the work directory. Format: s3://bucket-name or s3://bucket-name/path.	No
--preserve-resources	Preserve Batch Forge resources on deletion. Keeps AWS Batch compute environments and related resources when the compute environment is deleted from Seqera Platform.	No
--ecs-config	Custom ECS agent configuration file. Appends custom parameters to /etc/ecs/ecs.config on each cluster node. Provide path to configuration file.	No
--fusion-metrics-collection	Send Fusion metrics to Seqera for this compute environment. Fusion always generates the metrics; this only controls whether they are collected and sent to Seqera. Only valid when Fusion is enabled. If unset, Platform applies its default.	No

tw compute-envs add aws-batch manual

Add new AWS Batch compute environment using an existing environment.

tw compute-envs add aws-batch manual [OPTIONS]

Options

Option	Description	Required	Default
-n, --name	Compute environment name.	Yes
-w, --workspace	Workspace numeric identifier or reference in OrganizationName/WorkspaceName format (defaults to TOWER_WORKSPACE_ID environment variable)	No
-c, --credentials	Credentials identifier [default: workspace credentials].	No
--wait	Wait until given status or fail. Valid options: CREATING, AVAILABLE, DISABLED, DELETING, ERRORED, INVALID, DELETED.	No
-d, --description	Compute environment description.	No
--labels	Comma-separated list of labels.	No
--pre-run	Pre-run script.	No
--post-run	Post-run script.	No
--nextflow-config	Nextflow config	No
-e, --env	Add environment variables. By default are only added to the Nextflow head job process, if you want to add them to the process task prefix the name with 'compute:' or 'both:' if you want to make it available to both locations.	No
--head-job-cpus	Number of CPUs allocated to the Nextflow head job. Controls the compute resources for the main workflow orchestration process.	No
--head-job-memory	Memory allocation for the Nextflow head job in megabytes. Determines available memory for workflow orchestration.	No
--head-job-role	IAM role ARN to grant fine-grained permissions to the Nextflow head job. Enables secure access to AWS resources.	No
--compute-job-role	IAM role ARN to grant fine-grained permissions to Nextflow compute jobs. Controls access for individual pipeline tasks.	No
--batch-execution-role	IAM role ARN for ECS task execution. Grants Amazon ECS containers permission to make AWS API calls on your behalf.	No
--cli-path	Nextflow requires the AWS CLI installed in the Ec2 instances. Use this field to specify the path.	No
--secrets-kms-key	Customer-managed KMS key used to encrypt the temporary Secrets Manager secrets created for runs that use pipeline secrets. Accepts a key ARN or a key id. When omitted, the AWS-managed default Secrets Manager key is used.	No
--work-dir	Nextflow work directory. Path where workflow intermediate files are stored. Must be an S3 bucket path (e.g., s3://your-bucket/work).	Yes
-r, --region	AWS region where compute resources will be created (e.g., us-east-1, eu-west-1).	Yes
--head-queue	AWS Batch queue for the Nextflow head job. Should use on-demand instances for reliability.	Yes
--compute-queue	AWS Batch compute queue for running pipeline jobs. Nextflow submits tasks to this queue. Can be overridden in pipeline config.	Yes
--fusion-v2	Enable Fusion file system. Provides native access to S3 storage with low-latency I/O. Requires Wave containers.	No
--wave	Enable Wave containers. Allows access to private container repositories and on-demand container provisioning.	No
--fast-storage	Enable NVMe instance storage. Provides high-performance local storage for faster I/O operations. Requires Fusion file system.	No
--fusion-metrics-collection	Send Fusion metrics to Seqera for this compute environment. Fusion always generates the metrics; this only controls whether they are collected and sent to Seqera. Only valid when Fusion is enabled. If unset, Platform applies its default.	No

tw compute-envs add aws-cloud

Add new AWS Cloud compute environment.

tw compute-envs add aws-cloud [OPTIONS]

Options

Option	Description	Required	Default
-n, --name	Compute environment name.	Yes
-w, --workspace	Workspace numeric identifier or reference in OrganizationName/WorkspaceName format (defaults to TOWER_WORKSPACE_ID environment variable)	No
-c, --credentials	Credentials identifier [default: workspace credentials].	No
--wait	Wait until given status or fail. Valid options: CREATING, AVAILABLE, DISABLED, DELETING, ERRORED, INVALID, DELETED.	No
-d, --description	Compute environment description.	No
--labels	Comma-separated list of labels.	No
--pre-run	Pre-run script.	No
--post-run	Post-run script.	No
--nextflow-config	Nextflow config	No
-e, --env	Add environment variables. By default are only added to the Nextflow head job process, if you want to add them to the process task prefix the name with 'compute:' or 'both:' if you want to make it available to both locations.	No
--sched-enabled	Enable the Seqera scheduler for this compute environment. Defaults to false if not specified.	No
--provisioning-model	Instance provisioning model used by the Seqera scheduler. Valid values: SPOT, SPOT_FIRST, ONDEMAND.	No
--sched-machine-types	EC2 instance types for compute nodes managed by the Seqera scheduler. Comma-separated list (e.g., m5.xlarge,c5.2xlarge). Leave empty to let the scheduler select the most cost-effective types.	No
--arm64	Enable ARM64 (Graviton) architecture EC2 instances to run compute jobs. Provides cost-effective compute with comparable performance to x86.	No
--boot-disk-size	EC2 instance boot disk size in GB. Controls the root volume size for compute instances. If absent, Platform defaults to 50 GB gp3 volume.	No
--ebs-encryption	Encrypt the boot EBS volume of provisioned instances. Defaults to false if not specified.	No
--ebs-kms-key	KMS key ARN used to encrypt the boot EBS volume. Only applied when EBS encryption is enabled (--ebs-encryption). When omitted, the account/region default EBS encryption key is used.	No
--secrets-kms-key	Customer-managed KMS key used to encrypt the temporary Secrets Manager secrets created for runs that use pipeline secrets. Accepts a key ARN or a key id. When omitted, the AWS-managed default Secrets Manager key is used.	No
--ec2-key-pair	EC2 key pair name for SSH access to running instances. The key pair must already exist in the specified region.	No
--image-id	AMI ID for launching EC2 instances. If omitted, Seqera-maintained default AMI is used. Use Seqera AMIs for best performance.	No
--instance-profile-arn	IAM instance profile ARN used by EC2 instances to assume roles. If unspecified, Seqera provisions an ARN with sufficient permissions.	No
--instance-type	EC2 instance type (e.g., t3.medium, m5.large). If omitted, a default instance type is used.	No
--security-groups	Security group IDs for network access control. Comma-separated list defining firewall rules for EC2 instances.	No
--subnet-id	DEPRECATED - Use '--subnet-ids' instead. VPC subnet ID for instance placement. Determines network isolation and internet access configuration.	No
--subnet-ids	VPC subnet IDs for instance placement. Comma-separated list; the first subnet is used for basic placement while Intelligent Compute may use all of them. Mutually exclusive with --subnet-id.	No
--vpc-id	VPC ID used to scope subnet and security-group selection. Determines the network in which EC2 instances are launched.	No
--work-dir	Nextflow work directory. Path where workflow intermediate files are stored. Must be an S3 bucket path (e.g., s3://your-bucket/work). Credentials must have read-write access.	Yes
-r, --region	AWS region where EC2 instances will be launched (e.g., us-east-1, eu-west-1).	Yes
--allow-buckets	S3 buckets that the compute environment can access. Comma-separated list of S3 bucket names or paths to grant read-write permissions for workflow data.	No
--fusion-metrics-collection	Send Fusion metrics to Seqera for this compute environment. Fusion always generates the metrics; this only controls whether they are collected and sent to Seqera. Only valid when Fusion is enabled. If unset, Platform applies its default.	No

Use --secrets-kms-key to encrypt the temporary AWS Secrets Manager secrets that Platform creates for runs that use pipeline secrets with a customer-managed KMS key. Pass a key ARN or a key ID. When omitted, Platform uses the AWS-managed default key. The instance profile role needs kms:Decrypt on the key. See Advanced options for the field description and the required policy statement.

tw compute-envs add eks

Add new Amazon EKS compute environment.

tw compute-envs add eks [OPTIONS]

Options

Option	Description	Required	Default
-n, --name	Compute environment name.	Yes
-w, --workspace	Workspace numeric identifier or reference in OrganizationName/WorkspaceName format (defaults to TOWER_WORKSPACE_ID environment variable)	No
-c, --credentials	Credentials identifier [default: workspace credentials].	No
--wait	Wait until given status or fail. Valid options: CREATING, AVAILABLE, DISABLED, DELETING, ERRORED, INVALID, DELETED.	No
-d, --description	Compute environment description.	No
--labels	Comma-separated list of labels.	No
--pre-run	Pre-run script.	No
--post-run	Post-run script.	No
--nextflow-config	Nextflow config	No
-e, --env	Add environment variables. By default are only added to the Nextflow head job process, if you want to add them to the process task prefix the name with 'compute:' or 'both:' if you want to make it available to both locations.	No
--storage-mount	Mount path for the PersistentVolumeClaim. Directory where the storage is mounted in containers. If absent, Platform defaults to /scratch.	No
--compute-account	Kubernetes service account for Nextflow-submitted pipeline jobs. Controls permissions for individual task pods. If absent, Platform defaults to default.	No
--pod-cleanup	Pod cleanup policy after job completion. ON_SUCCESS removes pods only on success. ALWAYS removes all pods. NEVER keeps all pods.	No
--head-pod-spec	Custom PodSpec YAML for the Nextflow head job pod. Provide path to YAML file with custom pod configuration.	No
--service-pod-spec	Custom PodSpec YAML for the compute environment service pod. Provide path to YAML file with custom pod configuration.	No
--work-dir	Nextflow work directory. Path where workflow intermediate files are stored on the shared storage.	Yes
-r, --region	AWS region where the EKS cluster is deployed (e.g., us-east-1, eu-west-1).	Yes
--cluster-name	Name of the AWS EKS (Elastic Kubernetes Service) cluster for workflow execution.	Yes
--namespace	Kubernetes namespace for workflow execution. Isolates resources within the cluster.	Yes
--head-account	Kubernetes service account for connecting to the cluster. Used by the Nextflow head job to authenticate with the Kubernetes API.	Yes
--storage-claim	PersistentVolumeClaim name for scratch storage. Must support ReadWriteMany access mode for shared workflow data.	No

tw compute-envs add slurm

Add new Slurm compute environment.

tw compute-envs add slurm [OPTIONS]

Options

Option	Description	Required	Default
-n, --name	Compute environment name.	Yes
-w, --workspace	Workspace numeric identifier or reference in OrganizationName/WorkspaceName format (defaults to TOWER_WORKSPACE_ID environment variable)	No
-c, --credentials	Credentials identifier [default: workspace credentials].	No
--wait	Wait until given status or fail. Valid options: CREATING, AVAILABLE, DISABLED, DELETING, ERRORED, INVALID, DELETED.	No
-d, --description	Compute environment description.	No
--labels	Comma-separated list of labels.	No
--pre-run	Pre-run script.	No
--post-run	Post-run script.	No
--nextflow-config	Nextflow config	No
-e, --env	Add environment variables. By default are only added to the Nextflow head job process, if you want to add them to the process task prefix the name with 'compute:' or 'both:' if you want to make it available to both locations.	No
--max-queue-size	Maximum number of jobs Nextflow can submit simultaneously to the Slurm queue. Controls job submission rate. If absent, Platform defaults to 100.	No
--head-job-options	Additional submit options for the Nextflow head job. Appended to the sbatch command for the main orchestration process.	No
--work-dir	Nextflow work directory on the cluster's shared file system. Must be an absolute path accessible from all compute nodes.	Yes
-u, --user-name	Username for SSH connection to the HPC cluster. Used to authenticate and launch pipeline execution on the head node.	No
-H, --host-name	Hostname or IP address of the HPC head node for SSH connection. Typically the cluster login node. Must be a fully qualified hostname, not a local IP address.	No
-p, --port	SSH port for cluster connection. If absent, Platform defaults to port 22.	No
-q, --head-queue	Slurm queue for launching the Nextflow head job. The queue where the main workflow orchestration process runs.	Yes
--compute-queue	Slurm queue for pipeline task submission. Nextflow submits individual jobs to this queue. Can be overridden in pipeline configuration.	No
--launch-dir	Directory where Nextflow executes. Must be an absolute path with read-write permissions (if absent, Platform defaults to the pipeline work directory).	No

tw compute-envs add lsf

Add new IBM LSF compute environment.

tw compute-envs add lsf [OPTIONS]

Options

Option	Description	Required	Default
-n, --name	Compute environment name.	Yes
-w, --workspace	Workspace numeric identifier or reference in OrganizationName/WorkspaceName format (defaults to TOWER_WORKSPACE_ID environment variable)	No
-c, --credentials	Credentials identifier [default: workspace credentials].	No
--wait	Wait until given status or fail. Valid options: CREATING, AVAILABLE, DISABLED, DELETING, ERRORED, INVALID, DELETED.	No
-d, --description	Compute environment description.	No
--labels	Comma-separated list of labels.	No
--pre-run	Pre-run script.	No
--post-run	Post-run script.	No
--nextflow-config	Nextflow config	No
-e, --env	Add environment variables. By default are only added to the Nextflow head job process, if you want to add them to the process task prefix the name with 'compute:' or 'both:' if you want to make it available to both locations.	No
--max-queue-size	Maximum number of jobs Nextflow can submit simultaneously to the LSF queue. Controls job submission rate. If absent, Platform defaults to 100.	No
--head-job-options	Additional submit options for the Nextflow head job. Appended to the bsub command for the main orchestration process.	No
--unit-for-limits	Memory limit unit for LSF cluster. Must match LSF_UNIT_FOR_LIMITS in lsf.conf configuration file.	No
--per-job-mem-limit	Memory limit interpretation: per-job or per-process. Must match LSB_JOB_MEMLIMIT in lsf.conf configuration file.	No
--per-task-reserve	Memory reservation mode: per-task or per-host. Must match RESOURCE_RESERVE_PER_TASK in lsf.conf configuration file.	No
--work-dir	Nextflow work directory on the cluster's shared file system. Must be an absolute path accessible from all compute nodes.	Yes
-u, --user-name	Username for SSH connection to the HPC cluster. Used to authenticate and launch pipeline execution on the head node.	No
-H, --host-name	Hostname or IP address of the HPC head node for SSH connection. Typically the cluster login node. Must be a fully qualified hostname, not a local IP address.	No
-p, --port	SSH port for cluster connection. If absent, Platform defaults to port 22.	No
-q, --head-queue	LSF queue for launching the Nextflow head job. The queue where the main workflow orchestration process runs.	Yes
--compute-queue	LSF queue for pipeline task submission. Nextflow submits individual jobs to this queue. Can be overridden in pipeline configuration.	No
--launch-dir	Directory where Nextflow executes. Must be an absolute path with read-write permissions (if absent, Platform defaults to the pipeline work directory).	No

tw compute-envs add uge

Add new UNIVA grid engine compute environment.

tw compute-envs add uge [OPTIONS]

Options

Option	Description	Required	Default
-n, --name	Compute environment name.	Yes
-w, --workspace	Workspace numeric identifier or reference in OrganizationName/WorkspaceName format (defaults to TOWER_WORKSPACE_ID environment variable)	No
-c, --credentials	Credentials identifier [default: workspace credentials].	No
--wait	Wait until given status or fail. Valid options: CREATING, AVAILABLE, DISABLED, DELETING, ERRORED, INVALID, DELETED.	No
-d, --description	Compute environment description.	No
--labels	Comma-separated list of labels.	No
--pre-run	Pre-run script.	No
--post-run	Post-run script.	No
--nextflow-config	Nextflow config	No
-e, --env	Add environment variables. By default are only added to the Nextflow head job process, if you want to add them to the process task prefix the name with 'compute:' or 'both:' if you want to make it available to both locations.	No
--max-queue-size	Maximum number of jobs Nextflow can submit simultaneously to the Univa Grid Engine queue. Controls job submission rate. If absent, Platform defaults to 100.	No
--head-job-options	Additional submit options for the Nextflow head job. Appended to the submit command for the main orchestration process.	No
--work-dir	Nextflow work directory on the cluster's shared file system. Must be an absolute path accessible from all compute nodes.	Yes
-u, --user-name	Username for SSH connection to the HPC cluster. Used to authenticate and launch pipeline execution on the head node.	No
-H, --host-name	Hostname or IP address of the HPC head node for SSH connection. Typically the cluster login node. Must be a fully qualified hostname, not a local IP address.	No
-p, --port	SSH port for cluster connection. If absent, Platform defaults to port 22.	No
-q, --head-queue	Univa Grid Engine queue for launching the Nextflow head job. The queue where the main workflow orchestration process runs.	Yes
--compute-queue	Univa Grid Engine queue for pipeline task submission. Nextflow submits individual jobs to this queue. Can be overridden in pipeline configuration.	No
--launch-dir	Directory where Nextflow executes. Must be an absolute path with read-write permissions (if absent, Platform defaults to the pipeline work directory).	No

tw compute-envs add altair

Add new Altair PBS Pro compute environment.

tw compute-envs add altair [OPTIONS]

Options

Option	Description	Required	Default
-n, --name	Compute environment name.	Yes
-w, --workspace	Workspace numeric identifier or reference in OrganizationName/WorkspaceName format (defaults to TOWER_WORKSPACE_ID environment variable)	No
-c, --credentials	Credentials identifier [default: workspace credentials].	No
--wait	Wait until given status or fail. Valid options: CREATING, AVAILABLE, DISABLED, DELETING, ERRORED, INVALID, DELETED.	No
-d, --description	Compute environment description.	No
--labels	Comma-separated list of labels.	No
--pre-run	Pre-run script.	No
--post-run	Post-run script.	No
--nextflow-config	Nextflow config	No
-e, --env	Add environment variables. By default are only added to the Nextflow head job process, if you want to add them to the process task prefix the name with 'compute:' or 'both:' if you want to make it available to both locations.	No
--max-queue-size	Maximum number of jobs Nextflow can submit simultaneously to the Altair PBS queue. Controls job submission rate. If absent, Platform defaults to 100.	No
--head-job-options	Additional submit options for the Nextflow head job. Appended to the submit command for the main orchestration process.	No
--work-dir	Nextflow work directory on the cluster's shared file system. Must be an absolute path accessible from all compute nodes.	Yes
-u, --user-name	Username for SSH connection to the HPC cluster. Used to authenticate and launch pipeline execution on the head node.	No
-H, --host-name	Hostname or IP address of the HPC head node for SSH connection. Typically the cluster login node. Must be a fully qualified hostname, not a local IP address.	No
-p, --port	SSH port for cluster connection. If absent, Platform defaults to port 22.	No
-q, --head-queue	Altair PBS queue for launching the Nextflow head job. The queue where the main workflow orchestration process runs.	Yes
--compute-queue	Altair PBS queue for pipeline task submission. Nextflow submits individual jobs to this queue. Can be overridden in pipeline configuration.	No
--launch-dir	Directory where Nextflow executes. Must be an absolute path with read-write permissions (if absent, Platform defaults to the pipeline work directory).	No

tw compute-envs add moab

Add new MOAB compute environment.

tw compute-envs add moab [OPTIONS]

Options

Option	Description	Required	Default
-n, --name	Compute environment name.	Yes
-w, --workspace	Workspace numeric identifier or reference in OrganizationName/WorkspaceName format (defaults to TOWER_WORKSPACE_ID environment variable)	No
-c, --credentials	Credentials identifier [default: workspace credentials].	No
--wait	Wait until given status or fail. Valid options: CREATING, AVAILABLE, DISABLED, DELETING, ERRORED, INVALID, DELETED.	No
-d, --description	Compute environment description.	No
--labels	Comma-separated list of labels.	No
--pre-run	Pre-run script.	No
--post-run	Post-run script.	No
--nextflow-config	Nextflow config	No
-e, --env	Add environment variables. By default are only added to the Nextflow head job process, if you want to add them to the process task prefix the name with 'compute:' or 'both:' if you want to make it available to both locations.	No
--max-queue-size	Maximum number of jobs Nextflow can submit simultaneously to the Moab queue. Controls job submission rate. If absent, Platform defaults to 100.	No
--head-job-options	Additional submit options for the Nextflow head job. Appended to the submit command for the main orchestration process.	No
--work-dir	Nextflow work directory on the cluster's shared file system. Must be an absolute path accessible from all compute nodes.	Yes
-u, --user-name	Username for SSH connection to the HPC cluster. Used to authenticate and launch pipeline execution on the head node.	No
-H, --host-name	Hostname or IP address of the HPC head node for SSH connection. Typically the cluster login node. Must be a fully qualified hostname, not a local IP address.	No
-p, --port	SSH port for cluster connection. If absent, Platform defaults to port 22.	No
-q, --head-queue	Moab queue for launching the Nextflow head job. The queue where the main workflow orchestration process runs.	Yes
--compute-queue	Moab queue for pipeline task submission. Nextflow submits individual jobs to this queue. Can be overridden in pipeline configuration.	No
--launch-dir	Directory where Nextflow executes. Must be an absolute path with read-write permissions (if absent, Platform defaults to the pipeline work directory).	No

tw compute-envs add gke

Add new Google GKE compute environment.

tw compute-envs add gke [OPTIONS]

Options

Option	Description	Required	Default
-n, --name	Compute environment name.	Yes
-w, --workspace	Workspace numeric identifier or reference in OrganizationName/WorkspaceName format (defaults to TOWER_WORKSPACE_ID environment variable)	No
-c, --credentials	Credentials identifier [default: workspace credentials].	No
--wait	Wait until given status or fail. Valid options: CREATING, AVAILABLE, DISABLED, DELETING, ERRORED, INVALID, DELETED.	No
-d, --description	Compute environment description.	No
--labels	Comma-separated list of labels.	No
--pre-run	Pre-run script.	No
--post-run	Post-run script.	No
--nextflow-config	Nextflow config	No
-e, --env	Add environment variables. By default are only added to the Nextflow head job process, if you want to add them to the process task prefix the name with 'compute:' or 'both:' if you want to make it available to both locations.	No
--storage-mount	Mount path for the PersistentVolumeClaim. Directory where the storage is mounted in containers. If absent, Platform defaults to /scratch.	No
--compute-account	Kubernetes service account for Nextflow-submitted pipeline jobs. Controls permissions for individual task pods. If absent, Platform defaults to default.	No
--pod-cleanup	Pod cleanup policy after job completion. ON_SUCCESS removes pods only on success. ALWAYS removes all pods. NEVER keeps all pods.	No
--head-pod-spec	Custom PodSpec YAML for the Nextflow head job pod. Provide path to YAML file with custom pod configuration.	No
--service-pod-spec	Custom PodSpec YAML for the compute environment service pod. Provide path to YAML file with custom pod configuration.	No
--work-dir	Nextflow work directory. Path where workflow intermediate files are stored on the shared storage.	Yes
-r, --region	Google Cloud region or zone where the GKE cluster is deployed (e.g., us-central1, europe-west1).	Yes
--cluster-name	Name of the Google GKE (Google Kubernetes Engine) cluster for workflow execution.	Yes
--namespace	Kubernetes namespace for workflow execution. Isolates resources within the cluster.	Yes
--head-account	Kubernetes service account for connecting to the cluster. Used by the Nextflow head job to authenticate with the Kubernetes API.	Yes
--storage-claim	PersistentVolumeClaim name for scratch storage. Must support ReadWriteMany access mode for shared workflow data.	No

tw compute-envs add google-batch

Add new Google Batch compute environment.

tw compute-envs add google-batch [OPTIONS]

Options

Option	Description	Required	Default
-n, --name	Compute environment name.	Yes
-w, --workspace	Workspace numeric identifier or reference in OrganizationName/WorkspaceName format (defaults to TOWER_WORKSPACE_ID environment variable)	No
-c, --credentials	Credentials identifier [default: workspace credentials].	No
--wait	Wait until given status or fail. Valid options: CREATING, AVAILABLE, DISABLED, DELETING, ERRORED, INVALID, DELETED.	No
-d, --description	Compute environment description.	No
--labels	Comma-separated list of labels.	No
--pre-run	Pre-run script.	No
--post-run	Post-run script.	No
--nextflow-config	Nextflow config	No
-e, --env	Add environment variables. By default are only added to the Nextflow head job process, if you want to add them to the process task prefix the name with 'compute:' or 'both:' if you want to make it available to both locations.	No
--network	Google Cloud VPC network name or URI. Required when using network tags or subnets.	No
--subnetwork	Google Cloud VPC subnetwork name or URI. Must be in the same region as the compute environment location.	No
--network-tags	Comma-separated list of network tags applied to VMs for firewall rule targeting. Tags must be lowercase, use only letters, numbers, and hyphens (1-63 chars). Requires --network.	No
--use-private-address	Do not attach a public IP address to VM instances. When enabled, only Google internal services are accessible. Requires Cloud NAT for external access.	No
--boot-disk-size	Boot disk size in GB. Controls the root volume size for compute instances. If absent, Platform defaults to 50 GB.	No
--boot-disk-image	Custom boot disk image for compute job VMs. Accepts: projects/{PROJECT}/global/images/{IMAGE}, projects/{PROJECT}/global/images/family/{FAMILY}, or a Batch image name (e.g., batch-debian).	No
--head-job-cpus	Number of CPUs allocated to the Nextflow head job. Controls the compute resources for the main workflow orchestration process.	No
--head-job-memory	Memory allocation for the Nextflow head job in megabytes. Value must be a multiple of 256 MiB and from 0.5 GB to 8 GB per CPU.	No
--service-account-email	Google Cloud service account email for pipeline execution. Grants fine-grained IAM permissions to Nextflow jobs.	No
--head-job-machine-type	GCP machine type for the Nextflow head job (e.g., n2-standard-4). Mutually exclusive with --head-job-template.	No
--head-job-template	Google Compute Engine instance template for the Nextflow head job. Specify either the template name (if in the same project) or the fully qualified reference (projects/PROJECT_ID/global/instanceTemplates/TEMPLATE_NAME). Mutually exclusive with --head-job-machine-type.	No
--compute-jobs-machine-type	Comma-separated list of GCP machine types for compute jobs (e.g., n2-standard-8,c2-standard-4). Supports wildcard families (e.g., n2-*). Mutually exclusive with --compute-job-template.	No
--compute-job-template	Google Compute Engine instance template for pipeline compute jobs. Specify either the template name (if in the same project) or the fully qualified reference (projects/PROJECT_ID/global/instanceTemplates/TEMPLATE_NAME). Mutually exclusive with --compute-jobs-machine-type.	No
--work-dir	Nextflow work directory. Path where workflow intermediate files are stored. Must be a Google Cloud Storage bucket path (e.g., gs://your-bucket/work).	Yes
-l, --location	Google Cloud region where job executions are deployed to Google Batch API (e.g., us-central1, europe-west1).	Yes
--spot	Use Spot virtual machines. Enables cost-effective preemptible instances for compute workloads. Spot VMs may be interrupted when capacity is needed.	No
--fusion-v2	Enable Fusion file system. Provides native access to Google Cloud Storage with low-latency I/O. Requires Wave containers.	No
--fusion-snapshots	Enable Fusion Snapshots (beta). Allows Fusion to restore jobs interrupted by Spot VM reclamation. Requires Fusion v2.	No
--wave	Enable Wave containers. Allows access to private container repositories and on-demand container provisioning.	No
--fusion-metrics-collection	Send Fusion metrics to Seqera for this compute environment. Fusion always generates the metrics; this only controls whether they are collected and sent to Seqera. Only valid when Fusion is enabled. If unset, Platform applies its default.	No

tw compute-envs add google-cloud

Add new Google Cloud compute environment.

tw compute-envs add google-cloud [OPTIONS]

Options

Option	Description	Required	Default
-n, --name	Compute environment name.	Yes
-w, --workspace	Workspace numeric identifier or reference in OrganizationName/WorkspaceName format (defaults to TOWER_WORKSPACE_ID environment variable)	No
-c, --credentials	Credentials identifier [default: workspace credentials].	No
--wait	Wait until given status or fail. Valid options: CREATING, AVAILABLE, DISABLED, DELETING, ERRORED, INVALID, DELETED.	No
-d, --description	Compute environment description.	No
--labels	Comma-separated list of labels.	No
--pre-run	Pre-run script.	No
--post-run	Post-run script.	No
--nextflow-config	Nextflow config	No
-e, --env	Add environment variables. By default are only added to the Nextflow head job process, if you want to add them to the process task prefix the name with 'compute:' or 'both:' if you want to make it available to both locations.	No
--sched-enabled	Enable the Seqera scheduler for this compute environment. Defaults to false if not specified.	No
--provisioning-model	Instance provisioning model used by the Seqera scheduler. Valid values: SPOT, SPOT_FIRST, ONDEMAND.	No
--sched-machine-types	Compute Engine machine types for compute nodes managed by the Seqera scheduler. Comma-separated list (e.g., n2-standard-4,c2-standard-8). Leave empty to let the scheduler select the most cost-effective types.	No
--arm64	Enable ARM64 (Axion) architecture instances to run compute jobs. Provides efficient compute for compatible workloads.	No
--boot-disk-size	Boot disk size in GB for Compute Engine instances. Uses pd-standard disk type. If absent, Platform defaults to 50 GB.	No
--gpu	Enable GPU-enabled instances for compute jobs. When enabled, Deep Learning VM base images with CUDA are automatically selected.	No
--image-id	Image ID defining the operating system and pre-installed software for Compute Engine instances. Supports Ubuntu LTS Google public images. For GPU instances, Deep Learning VM base images with CUDA are automatically selected.	No
--instance-type	Compute Engine machine type (e.g., n1-standard-1, n2-standard-2). If omitted, a default machine type is used.	No
--network	Google Cloud VPC network name or URI. Required when using subnetworks or network tags. When omitted, the project's 'default' network is used.	No
--subnetworks	Google Cloud VPC subnetworks for instance placement. Comma-separated list of names or URIs in the same region as the compute environment; the first is used for basic placement while Intelligent Compute may use all of them. Requires --network.	No
--network-tags	Comma-separated list of network tags applied to VMs for firewall rule targeting. Tags must be lowercase, use only letters, numbers, and hyphens (1-63 chars). Requires --network.	No
--use-private-address	Do not attach a public IP address to VM instances. When enabled, only Google internal services are accessible. Requires Cloud NAT for external access.	No
--work-dir	Nextflow work directory. Path where workflow intermediate files are stored. Must be a Google Cloud Storage bucket path (e.g., gs://your-bucket/work). Credentials must have read-write access.	Yes
-r, --region	Google Cloud region where compute instances will be launched (e.g., us-central1, europe-west1).	Yes
-z, --zone	Google Cloud zone within the region (e.g., us-central1-a). If omitted, defaults to the first zone alphabetically in the region.	Yes
--fusion-metrics-collection	Send Fusion metrics to Seqera for this compute environment. Fusion always generates the metrics; this only controls whether they are collected and sent to Seqera. Only valid when Fusion is enabled. If unset, Platform applies its default.	No

Use the networking options to place VMs in an existing VPC instead of the project's default network:

tw compute-envs add google-cloud --name=my_gcp_ce \
--credentials=<my_gcp_creds> --region=europe-west2 --zone=europe-west2-a \
--work-dir=gs://<bucket name> \
--network=my-vpc --subnetworks=my-subnet-a,my-subnet-b \
--network-tags=allow-ssh --use-private-address

• --subnetworks and --network-tags require --network. The CLI rejects the command before contacting Platform if --network is missing or a tag is not lowercase letters, numbers, and hyphens.
• Subnetworks must be in the compute environment region. VMs are placed in the first listed subnetwork, and Intelligent Compute may use all of them.
• With --use-private-address, VMs have no public IP address, so the VPC must provide outbound access through Cloud NAT and Private Google Access.

See Advanced options for the equivalent fields in the Platform UI.

tw compute-envs add azure-batch

Add new Azure Batch compute environments.

tw compute-envs add azure-batch

tw compute-envs add azure-batch forge

Add new Azure Batch compute environment with automatic provisioning of compute resources.

tw compute-envs add azure-batch forge [OPTIONS]

Options

Option	Description	Required	Default
-n, --name	Compute environment name.	Yes
-w, --workspace	Workspace numeric identifier or reference in OrganizationName/WorkspaceName format (defaults to TOWER_WORKSPACE_ID environment variable)	No
-c, --credentials	Credentials identifier [default: workspace credentials].	No
--wait	Wait until given status or fail. Valid options: CREATING, AVAILABLE, DISABLED, DELETING, ERRORED, INVALID, DELETED.	No
-d, --description	Compute environment description.	No
--labels	Comma-separated list of labels.	No
--pre-run	Pre-run script.	No
--post-run	Post-run script.	No
--nextflow-config	Nextflow config	No
-e, --env	Add environment variables. By default are only added to the Nextflow head job process, if you want to add them to the process task prefix the name with 'compute:' or 'both:' if you want to make it available to both locations.	No
--head-vm-type	Azure VM size for the head pool (dual pool mode). If absent, defaults to Standard_D2s_v3.	No
--head-vm-count	Number of VMs in the head pool (dual pool mode). If absent, defaults to 1.	No
--head-no-auto-scale	Disable autoscaling for the head pool (dual pool mode).	No
--head-boot-disk-size	Boot disk size in GB for the head pool nodes (dual pool mode). Overrides --boot-disk-size for this pool.	No
--worker-vm-type	Azure VM size for the worker pool (dual pool mode). If absent, defaults to Standard_D4s_v3.	No
--worker-vm-count	Max number of VMs in the worker pool (dual pool mode).	No
--worker-no-auto-scale	Disable autoscaling for the worker pool (dual pool mode).	No
--worker-boot-disk-size	Boot disk size in GB for the worker pool nodes (dual pool mode). Overrides --boot-disk-size for this pool.	No
--managed-identity-head-client-id	Head job managed identity client ID (UUID). The user-assigned managed identity used by the Nextflow launcher (head job).	No
--managed-identity-pool-client-id	Compute job managed identity client ID (UUID). The user-assigned managed identity used by compute tasks running on Batch pool nodes.	No
--managed-identity-head-resource-id	Head job managed identity resource ID. Full Azure resource ID of the user-assigned managed identity for the head job. Required in Forge mode when head job managed identity client ID is specified. Format: /subscriptions/{sub}/resourceGroups/{rg}/providers/Microsoft.ManagedIdentity/userAssignedIdentities/{name}	No
--managed-identity-pool-resource-id	Compute job managed identity resource ID. Full Azure resource ID of the user-assigned managed identity for compute jobs. Required in Forge mode when compute job managed identity client ID is specified. Format: /subscriptions/{sub}/resourceGroups/{rg}/providers/Microsoft.ManagedIdentity/userAssignedIdentities/{name}	No
--delete-jobs-on-completion	Delete Azure Batch jobs when the workflow completes successfully. Failed jobs are always preserved. Default: false.	No
--delete-tasks-on-completion	Delete individual Azure Batch tasks when they complete successfully. Failed tasks are preserved. Default: true.	No
--terminate-jobs-on-completion	Terminate Azure Batch jobs when all tasks complete. Default: true.	No
--token-duration	Duration of the SAS (shared access signature) token for Azure Blob Storage access. If absent, Platform defaults to 12h.	No
--job-max-wall-clock-time	Maximum elapsed time for an Azure Batch job before automatic termination. Accepts duration syntax (e.g., '7d', '1d12h', '168h'). Defaults to 7d. Maximum: 180 days.	No
--work-dir	Nextflow work directory. Path where workflow intermediate files are stored. Must be an Azure Blob Storage path.	Yes
-l, --location	Azure region where compute resources will be deployed (e.g., eastus, westeurope).	Yes
--dual-pool	Enable dual pool mode with separate head and worker pools. Head pool runs the Nextflow launcher on a small VM; worker pool scales independently for pipeline tasks.	No
--vm-type	Azure VM size for compute pool (single pool mode). Must be a valid Azure Batch VM type. If absent, Platform defaults to Standard_D4_v3.	No
--vm-count	Number of VMs in the Batch pool (single pool mode). With autoscaling enabled, this is the maximum capacity. Pool scales to zero when unused.	No
--no-auto-scale	Disable pool autoscaling (single pool mode). When disabled, pool maintains fixed VM count and does not scale based on workload.	No
--boot-disk-size	Boot disk size in GB for pool nodes. Applies to all pools. In dual pool mode, per-pool values (--head-boot-disk-size, --worker-boot-disk-size) take precedence. If absent, Azure's default is used.	No
--preserve-resources	Preserve Azure Batch pool resources on deletion. Keeps the compute pool and related resources when the compute environment is deleted from Seqera Platform.	No
--registry-credentials	Container registry credentials for private registries. Comma-separated list of credential names to access private Docker registries.	No
--fusion-v2	Enable Fusion file system. Provides native access to Azure Blob Storage with low-latency I/O. Requires Wave containers.	No
--wave	Enable Wave containers. Allows access to private container repositories and on-demand container provisioning.	No
--subnet-id	Azure VNet subnet resource ID for private network isolation. Requires Entra (service principal) credentials. Format: /subscriptions/{sub}/resourceGroups/{rg}/providers/Microsoft.Network/virtualNetworks/{vnet}/subnets/{subnet}	No
--fusion-metrics-collection	Send Fusion metrics to Seqera for this compute environment. Fusion always generates the metrics; this only controls whether they are collected and sent to Seqera. Only valid when Fusion is enabled. If unset, Platform applies its default.	No

tw compute-envs add azure-batch manual

Add new Azure Batch compute environment using an existing environment.

tw compute-envs add azure-batch manual [OPTIONS]

Options

Option	Description	Required	Default
-n, --name	Compute environment name.	Yes
-w, --workspace	Workspace numeric identifier or reference in OrganizationName/WorkspaceName format (defaults to TOWER_WORKSPACE_ID environment variable)	No
-c, --credentials	Credentials identifier [default: workspace credentials].	No
--wait	Wait until given status or fail. Valid options: CREATING, AVAILABLE, DISABLED, DELETING, ERRORED, INVALID, DELETED.	No
-d, --description	Compute environment description.	No
--labels	Comma-separated list of labels.	No
--pre-run	Pre-run script.	No
--post-run	Post-run script.	No
--nextflow-config	Nextflow config	No
-e, --env	Add environment variables. By default are only added to the Nextflow head job process, if you want to add them to the process task prefix the name with 'compute:' or 'both:' if you want to make it available to both locations.	No
--managed-identity-head-client-id	Head job managed identity client ID (UUID). The user-assigned managed identity used by the Nextflow launcher (head job).	No
--managed-identity-pool-client-id	Compute job managed identity client ID (UUID). The user-assigned managed identity used by compute tasks running on Batch pool nodes.	No
--delete-jobs-on-completion	Delete Azure Batch jobs when the workflow completes successfully. Failed jobs are always preserved. Default: false.	No
--delete-tasks-on-completion	Delete individual Azure Batch tasks when they complete successfully. Failed tasks are preserved. Default: true.	No
--terminate-jobs-on-completion	Terminate Azure Batch jobs when all tasks complete. Default: true.	No
--token-duration	Duration of the SAS (shared access signature) token for Azure Blob Storage access. If absent, Platform defaults to 12h.	No
--job-max-wall-clock-time	Maximum elapsed time for an Azure Batch job before automatic termination. Accepts duration syntax (e.g., '7d', '1d12h', '168h'). Defaults to 7d. Maximum: 180 days.	No
--work-dir	Nextflow work directory. Path where workflow intermediate files are stored. Must be an Azure Blob Storage path.	Yes
-l, --location	Azure region where compute resources will be deployed (e.g., eastus, westeurope).	Yes
--compute-pool-name	Pre-configured Azure Batch pool for the Nextflow head job. When used with --worker-pool, this pool handles only the launcher. Must include azcopy command-line tool.	Yes
--worker-pool	Pre-configured Azure Batch pool for pipeline worker tasks. When specified, the head job runs on --compute-pool-name and worker tasks run on this pool. Must be different from the head pool.	No
--fusion-v2	Enable Fusion file system. Provides native access to Azure Blob Storage with low-latency I/O. Requires Wave containers.	No
--wave	Enable Wave containers. Allows access to private container repositories and on-demand container provisioning.	No
--subnet-id	Azure VNet subnet resource ID for private network isolation. Requires Entra (service principal) credentials. Format: /subscriptions/{sub}/resourceGroups/{rg}/providers/Microsoft.Network/virtualNetworks/{vnet}/subnets/{subnet}	No
--fusion-metrics-collection	Send Fusion metrics to Seqera for this compute environment. Fusion always generates the metrics; this only controls whether they are collected and sent to Seqera. Only valid when Fusion is enabled. If unset, Platform applies its default.	No

tw compute-envs add azure-cloud

Add new Azure Cloud compute environment.

tw compute-envs add azure-cloud [OPTIONS]

Options

Option	Description	Required	Default
-n, --name	Compute environment name.	Yes
-w, --workspace	Workspace numeric identifier or reference in OrganizationName/WorkspaceName format (defaults to TOWER_WORKSPACE_ID environment variable)	No
-c, --credentials	Credentials identifier [default: workspace credentials].	No
--wait	Wait until given status or fail. Valid options: CREATING, AVAILABLE, DISABLED, DELETING, ERRORED, INVALID, DELETED.	No
-d, --description	Compute environment description.	No
--labels	Comma-separated list of labels.	No
--pre-run	Pre-run script.	No
--post-run	Post-run script.	No
--nextflow-config	Nextflow config	No
-e, --env	Add environment variables. By default are only added to the Nextflow head job process, if you want to add them to the process task prefix the name with 'compute:' or 'both:' if you want to make it available to both locations.	No
--sched-enabled	Enable the Seqera scheduler for this compute environment. Defaults to false if not specified.	No
--provisioning-model	Instance provisioning model used by the Seqera scheduler. Valid values: SPOT, SPOT_FIRST, ONDEMAND.	No
--sched-machine-types	Azure VM sizes for compute nodes managed by the Seqera scheduler. Comma-separated list (e.g., Standard_D4s_v3,Standard_E4s_v3). Leave empty to let the scheduler select the most cost-effective sizes.	No
--data-collection-endpoint	Azure Monitor data collection endpoint URL for log ingestion. Used to route logs to Log Analytics workspace.	No
--data-collection-rule-id	Azure Monitor data collection rule ID. Defines how logs are processed and routed to destination workspaces.	No
--instance-type	Azure virtual machine size (e.g., Standard_D2s_v3, Standard_E4s_v3). If omitted, a default VM size is used.	No
--log-table-name	Custom table name in Log Analytics workspace for storing compute environment logs. Enables organized log management.	No
--log-workspace-id	Azure Log Analytics workspace ID for monitoring compute environment activity and logs.	No
--managed-identity-client-id	User-assigned managed identity client ID for authentication. Used with managed identity resource ID for VM access control.	No
--managed-identity-id	User-assigned managed identity resource ID. Provides VMs with Azure resource access without storing credentials.	No
--network-id	Azure virtual network resource ID. Defines the network where VMs will be deployed for network isolation and connectivity.	No
--subnets	Subnet names within the virtual network for VM placement. Comma-separated list scoping where VMs are launched inside the network specified by --network-id.	No
--subscription-id	Azure subscription ID where resources will be created. Used to specify the billing and access control boundary.	No
--work-dir	Nextflow work directory. Path where workflow intermediate files are stored. Must be an Azure Blob Storage path (e.g., az://your-container/work). Credentials must have read-write access.	Yes
-r, --region	Azure region where virtual machines will be deployed (e.g., eastus, westeurope).	Yes
--resource-group	Azure resource group for organizing and managing virtual machines. The resource group must already exist in the subscription.	Yes
--fusion-metrics-collection	Send Fusion metrics to Seqera for this compute environment. Fusion always generates the metrics; this only controls whether they are collected and sent to Seqera. Only valid when Fusion is enabled. If unset, Platform applies its default.	No

tw compute-envs add seqera-compute

Add new Seqera Compute environment.

tw compute-envs add seqera-compute [OPTIONS]

Options

Option	Description	Required	Default
-n, --name	Compute environment name.	Yes
-w, --workspace	Workspace numeric identifier or reference in OrganizationName/WorkspaceName format (defaults to TOWER_WORKSPACE_ID environment variable)	No
-c, --credentials	Credentials identifier [default: workspace credentials].	No
--wait	Wait until given status or fail. Valid options: CREATING, AVAILABLE, DISABLED, DELETING, ERRORED, INVALID, DELETED.	No
-d, --description	Compute environment description.	No
--labels	Comma-separated list of labels.	No
--pre-run	Pre-run script.	No
--post-run	Post-run script.	No
--nextflow-config	Nextflow config	No
-e, --env	Add environment variables. By default are only added to the Nextflow head job process, if you want to add them to the process task prefix the name with 'compute:' or 'both:' if you want to make it available to both locations.	No
--work-dir	Work directory suffix relative to the S3 bucket that will be created by Seqera Compute.	No
-r, --region	AWS region.	Yes
--instance-type-size	Studios instance size, controlling compute resources and capabilities. Options: SMALL, MEDIUM, LARGE. Free-tier organizations are limited to SMALL.	No

tw compute-envs update

Update a compute environment.

tw compute-envs update [OPTIONS]

Options

Option	Description	Required	Default
-i, --id	Compute environment unique identifier.	Yes
-n, --name	Compute environment name.	Yes
-w, --workspace	Workspace numeric identifier or reference in OrganizationName/WorkspaceName format (defaults to TOWER_WORKSPACE_ID environment variable)	No
--new-name	New compute environment name.	No
-d, --description	Compute environment description.	No

tw compute-envs delete

Delete a compute environment.

tw compute-envs delete [OPTIONS]

Options

Option	Description	Required	Default
-i, --id	Compute environment unique identifier.	Yes
-n, --name	Compute environment name.	Yes
-w, --workspace	Workspace numeric identifier or reference in OrganizationName/WorkspaceName format (defaults to TOWER_WORKSPACE_ID environment variable)	No
--wait	Wait until the compute environment is fully deleted.	No

Command:

Command:

tw compute-envs delete --name=my_aws_ce

Example output:

Compute environment '1sxCxvxfx8xnxdxGxQxqxH' deleted at user workspace

tw compute-envs view

View compute environment details.

tw compute-envs view [OPTIONS]

Options

Option	Description	Required	Default
-i, --id	Compute environment unique identifier.	Yes
-n, --name	Compute environment name.	Yes
-w, --workspace	Workspace numeric identifier or reference in OrganizationName/WorkspaceName format (defaults to TOWER_WORKSPACE_ID environment variable)	No

tw compute-envs list

List compute environments.

tw compute-envs list [OPTIONS]

Options

Option	Description	Required	Default
-w, --workspace	Workspace numeric identifier or reference in OrganizationName/WorkspaceName format (defaults to TOWER_WORKSPACE_ID environment variable)	No

tw compute-envs export

Export compute environment configuration as a JSON file.

tw compute-envs export [OPTIONS] [FILENAME]

Arguments

Argument	Description	Required
FILENAME	File name and path for the exported compute environment configuration.	No

Options

Option	Description	Required	Default
-i, --id	Compute environment unique identifier.	Yes
-n, --name	Compute environment name.	Yes
-w, --workspace	Workspace numeric identifier or reference in OrganizationName/WorkspaceName format (defaults to TOWER_WORKSPACE_ID environment variable)	No

Export a compute environment configuration as JSON for scripting and reproducibility.

Command:

tw compute-envs export --name=my_aws_ce my_aws_ce_v1.json

Example output:

Compute environment exported into 'my_aws_ce_v1.json'

tw compute-envs import

Import a compute environment configuration from a JSON file.

tw compute-envs import [OPTIONS] <FILENAME>

Arguments

Argument	Description	Required
FILENAME	File path containing the compute environment configuration.	Yes

Options

Option	Description	Required	Default
-n, --name	Compute environment name.	Yes
-w, --workspace	Workspace numeric identifier or reference in OrganizationName/WorkspaceName format (defaults to TOWER_WORKSPACE_ID environment variable)	No
-c, --credentials	Credentials identifier [default: workspace credentials].	No
--wait	Wait until given status or fail. Valid options: CREATING, AVAILABLE, DISABLED, DELETING, ERRORED, INVALID, DELETED.	No
-d, --description	Compute environment description.	No
--labels	Comma-separated list of labels.	No
--overwrite	Overwrite the compute environment if it already exists.	No	false

Import a compute environment into a workspace from a previously exported JSON file.

Command:

tw compute-envs import --name=my_aws_ce_v1 ./my_aws_ce_v1.json

Example output:

New AWS-BATCH compute environment 'my_aws_ce_v1' added at user workspace

tw compute-envs primary

Manage the primary compute environment.

tw compute-envs primary

tw compute-envs primary get

Get the primary compute environment.

tw compute-envs primary get [OPTIONS]

Options

Option	Description	Required	Default
-w, --workspace	Workspace numeric identifier or reference in OrganizationName/WorkspaceName format (defaults to TOWER_WORKSPACE_ID environment variable)	No

tw compute-envs primary set

Set a compute environment as primary.

tw compute-envs primary set [OPTIONS]

Options

Option	Description	Required	Default
-w, --workspace	Workspace numeric identifier or reference in OrganizationName/WorkspaceName format (defaults to TOWER_WORKSPACE_ID environment variable)	No
-i, --id	Compute environment unique identifier.	Yes
-n, --name	Compute environment name.	Yes

Select a primary compute environment to use by default in a workspace. You can override the workspace primary compute environment when you create or launch a pipeline.

Command:

tw compute-envs primary set --name=my_aws_ce

Example output:

Primary compute environment for workspace 'user' was set to 'my_aws_ce (1sxCxvxfx8xnxdxGxQxqxH)'

tw compute-envs validate

Validate a compute environment by re-running its pre-flight checks (credential and work-directory).

tw compute-envs validate [OPTIONS]

Options

Option	Description	Required	Default
-i, --id	Compute environment unique identifier.	Yes
-n, --name	Compute environment name.	Yes
-w, --workspace	Workspace numeric identifier or reference in OrganizationName/WorkspaceName format (defaults to TOWER_WORKSPACE_ID environment variable)	No
--force	Skip the pre-flight checks and force an INVALID compute environment (with an AVAILABLE credential) to AVAILABLE. Rejected otherwise.	No