From 4ff461d6c17268c87f27e0ca6faab148e37b9918 Mon Sep 17 00:00:00 2001 From: llewellyn-sl Date: Fri, 10 May 2024 21:13:19 +0200 Subject: [PATCH 1/2] Update docs and release notes RE ARM64 config variable --- .../version-23.2.0/compute-envs/aws-batch.mdx | 11 +++++------ .../enterprise/configuration/overview.mdx | 6 +++--- .../version-23.3.0/compute-envs/aws-batch.mdx | 4 +++- .../configuration/configtables/features_env.yml | 5 +++++ .../version-23.4.0/enterprise/release_notes/23.2.mdx | 1 + .../enterprise/release_notes/changelog.mdx | 3 ++- .../enterprise/release_notes/enterprise_latest.mdx | 4 ++++ 7 files changed, 23 insertions(+), 11 deletions(-) diff --git a/platform_versioned_docs/version-23.2.0/compute-envs/aws-batch.mdx b/platform_versioned_docs/version-23.2.0/compute-envs/aws-batch.mdx index 0da59d86..b99a08cf 100644 --- a/platform_versioned_docs/version-23.2.0/compute-envs/aws-batch.mdx +++ b/platform_versioned_docs/version-23.2.0/compute-envs/aws-batch.mdx @@ -172,14 +172,13 @@ The bucket should be in the same region selected in the previous step. 17. Select **Enable GPUs** if you intend to run GPU-dependent workflows in the compute environment. See [GPU usage](./overview.mdx#aws-batch) for more information. -18. Select **Use Graviton CPU architecture** to execute the pipeline on Graviton-based EC2 instances (i.e., ARM64 CPU architecture). When enabled, Tower automatically selects `m6g`, `r6g` and `c6g` instance types for compute jobs. You can override this selection in the **Instance types** field under **Advanced settings**. +18. Select **Use Graviton CPU architecture** to execute the pipeline on Graviton-based EC2 instances (ARM64 CPU architecture). When enabled, Tower automatically selects `m6g`, `r6g` and `c6g` instance types for compute jobs. You can override this selection in the **Instance types** field under **Advanced settings**. -:::note -Requires Fargate, Wave containers, and Fusion v2 file system to be enabled. This feature is not compatible with GPU-based architecture. - - This is a preview feature that requires the use of Nextflow `23.06.0-edge`. Graviton support is currently only available in a [Tower Cloud](https://tower.nf/) public instance. + :::note + Requires Fargate, Wave containers, and Fusion v2 file system to be enabled. This feature is not compatible with GPU-based architecture. -::: + This is a preview feature that requires the use of Nextflow `23.06.0-edge`. In Tower Enterprise deployments, set `TOWER_ENABLE_ARM64=true` to enable Graviton instance support. + ::: 19. Enter any additional **Allowed S3 buckets** that your workflows require to read input data or write output data. The **Pipeline work directory** bucket above is added by default to the list of **Allowed S3 buckets**. diff --git a/platform_versioned_docs/version-23.2.0/enterprise/configuration/overview.mdx b/platform_versioned_docs/version-23.2.0/enterprise/configuration/overview.mdx index 5407c90e..3e60a306 100644 --- a/platform_versioned_docs/version-23.2.0/enterprise/configuration/overview.mdx +++ b/platform_versioned_docs/version-23.2.0/enterprise/configuration/overview.mdx @@ -5,8 +5,6 @@ date: "21 Apr 2023" tags: [configuration] --- -# Tower configuration - The configuration of your Tower instance can be controlled using various environment variables specified in the [tower.env](../_templates/docker/tower.env) and [tower.yml](../_templates/docker/tower.yml) files. Note that a number of core Tower configuration values must be specified using environment variables in `tower.env`. In the [tower.yml](../_templates/docker/tower.yml) file, configuration options are objects nested within the `tower` object. This is formatted as follows: @@ -44,7 +42,9 @@ Specify general Tower configuration values in your environment variables. The bo `TOWER_CRON_SERVER_PORT`: Define the HTTP port usd by the Tower cron service (default: `8080`, requires Tower 21.06.1 or later). -`TOWER_USER_WORKSPACE_ENABLED` : Enable or disable the showing of the user private workspace context. (default: `true`, requires Tower 22.1.0 or later). +`TOWER_USER_WORKSPACE_ENABLED`: Enable or disable the showing of the user private workspace context. (default: `true`, requires Tower 22.1.0 or later). + +`TOWER_ENABLE_ARM64`: Enable support for ARM64 CPU architectures. Required to enable Graviton in AWS Batch compute environments. ``` diff --git a/platform_versioned_docs/version-23.3.0/compute-envs/aws-batch.mdx b/platform_versioned_docs/version-23.3.0/compute-envs/aws-batch.mdx index a684c9cd..9c2c3164 100644 --- a/platform_versioned_docs/version-23.3.0/compute-envs/aws-batch.mdx +++ b/platform_versioned_docs/version-23.3.0/compute-envs/aws-batch.mdx @@ -161,7 +161,9 @@ Fargate requires the Fusion v2 file system and a **spot** provisioning model. Fa 20. Select **Use Graviton CPU architecture** to execute on Graviton-based EC2 instances (ARM64 CPU architecture). When enabled, `c6g` instance types are used by default for compute jobs. You can specify your own **Instance types** under [**Advanced options**](#advanced-options). Seqera Platform supports all AWS Batch-compatible Graviton2 and Graviton3 instance types. :::note - Graviton requires Fargate, Wave containers, and Fusion v2 file system to be enabled. This feature is not compatible with GPU-based architecture. + Graviton requires Fargate, Wave containers, and Fusion v2 file system to be enabled. This feature is not compatible with GPU-based architecture. + + In Tower Enterprise deployments, set `TOWER_ENABLE_ARM64=true` to enable Graviton instance support. ::: 21. Enter any additional **Allowed S3 buckets** that your workflows require to read input data or write output data. The **Pipeline work directory** bucket above is added by default to the list of **Allowed S3 buckets**. diff --git a/platform_versioned_docs/version-23.3.0/enterprise/configuration/configtables/features_env.yml b/platform_versioned_docs/version-23.3.0/enterprise/configuration/configtables/features_env.yml index ac14c43b..c4d6a956 100644 --- a/platform_versioned_docs/version-23.3.0/enterprise/configuration/configtables/features_env.yml +++ b/platform_versioned_docs/version-23.3.0/enterprise/configuration/configtables/features_env.yml @@ -9,6 +9,11 @@ Description: > Define the Wave containers service endpoint URL. Value: 'Example: `https://wave.seqera.io`' +- + Environment variable: '`TOWER_ENABLE_ARM64`' + Description: > + Enable support for ARM64 CPU architecture. Required to enable Graviton instance support in AWS Batch compute environments. + Value: 'Default: `false`' - Environment variable: '`TOWER_ENABLE_AWS_SSM`' Description: > diff --git a/platform_versioned_docs/version-23.4.0/enterprise/release_notes/23.2.mdx b/platform_versioned_docs/version-23.4.0/enterprise/release_notes/23.2.mdx index 83b98672..7be0ae5c 100644 --- a/platform_versioned_docs/version-23.4.0/enterprise/release_notes/23.2.mdx +++ b/platform_versioned_docs/version-23.4.0/enterprise/release_notes/23.2.mdx @@ -8,6 +8,7 @@ Leveraging the Fusion file system, you can now run the Nextflow head job of your ### Other improvements +- Add support for Graviton architecture in AWS Batch compute environments - Allow Launcher users to create, edit, and upload datasets - Harmonize list sorting in Compute environments and Credentials list pages - Update Enable GPU label, sublabel and add warning when activated diff --git a/platform_versioned_docs/version-23.4.0/enterprise/release_notes/changelog.mdx b/platform_versioned_docs/version-23.4.0/enterprise/release_notes/changelog.mdx index ba44fb25..12523774 100644 --- a/platform_versioned_docs/version-23.4.0/enterprise/release_notes/changelog.mdx +++ b/platform_versioned_docs/version-23.4.0/enterprise/release_notes/changelog.mdx @@ -15,6 +15,7 @@ tags: [changelog] - Allow previewing of Nextflow output files in Data Explorer - New base image nginx 1.25.3 for tower-frontend unprivileged - Seqera Platform Enterprise license model change — requires new licenses for existing Enterprise customers +- Remove tower.enable.arm64 config option - Feature: Changed default AzBatch image to ubuntu-server - Feature: Set private address for head job configuration in Google Batch - Feature: VM instance template support for Google Batch @@ -99,7 +100,7 @@ tags: [changelog] ### 23.2.0 - 31 Jul 2023 - Added: Support for Fargate for head job -- Added: Support for Graviton architecture to AWS Batch compute environment (Cloud preview) +- Added: Support for Graviton architecture in AWS Batch compute environments - Added: Ability to rename Actions, CEs, Pipelines, and Workspaces - Added: support for AWS SES (simple email service) as alternative to SMTP service for sending emails - Added: Ability to edit the names of Tower entities: diff --git a/platform_versioned_docs/version-23.4.0/enterprise/release_notes/enterprise_latest.mdx b/platform_versioned_docs/version-23.4.0/enterprise/release_notes/enterprise_latest.mdx index 993ec413..b9bc32c5 100644 --- a/platform_versioned_docs/version-23.4.0/enterprise/release_notes/enterprise_latest.mdx +++ b/platform_versioned_docs/version-23.4.0/enterprise/release_notes/enterprise_latest.mdx @@ -59,6 +59,10 @@ If you specify a DNS name as your `TOWER_SERVER_URL`, but access your Seqera ins A previous iteration of the rebranded Seqera documentation noted `seqera` as the default and example value for certain variables (such as default database names). The rebranding from Nextflow Tower to Seqera Platform is an ongoing, incremental process and as such, legacy `tower` values and naming conventions used by the Seqera backend will remain in place until a future release. Updates to configuration variables and values will be communicated well in advance to prepare users for any breaking changes. +**ARM64 CPU architecture support enabled by default** + +The `TOWER_ENABLE_ARM64` configuration environment variable is no longer needed to enable Graviton instance support in AWS Batch compute environments. ARM64 CPU architecture support is enabled by default. + ## Upgrade steps This version requires a database schema update. Follow these steps to update your DB instance and the Seqera installation. From c99a03c6e453ffe73464aa6a97a5e3c02abaedc0 Mon Sep 17 00:00:00 2001 From: llewellyn-sl Date: Tue, 14 May 2024 10:21:48 -0400 Subject: [PATCH 2/2] Code review fixes --- .../version-23.3.0/compute-envs/aws-batch.mdx | 2 +- .../enterprise/configuration/configtables/features_env.yml | 2 +- .../enterprise/release_notes/enterprise_latest.mdx | 2 +- 3 files changed, 3 insertions(+), 3 deletions(-) diff --git a/platform_versioned_docs/version-23.3.0/compute-envs/aws-batch.mdx b/platform_versioned_docs/version-23.3.0/compute-envs/aws-batch.mdx index 9c2c3164..756ac8a3 100644 --- a/platform_versioned_docs/version-23.3.0/compute-envs/aws-batch.mdx +++ b/platform_versioned_docs/version-23.3.0/compute-envs/aws-batch.mdx @@ -163,7 +163,7 @@ Fargate requires the Fusion v2 file system and a **spot** provisioning model. Fa :::note Graviton requires Fargate, Wave containers, and Fusion v2 file system to be enabled. This feature is not compatible with GPU-based architecture. - In Tower Enterprise deployments, set `TOWER_ENABLE_ARM64=true` to enable Graviton instance support. + In Seqera Enterprise deployments, set `TOWER_ENABLE_ARM64=true` to enable Graviton instance support. ::: 21. Enter any additional **Allowed S3 buckets** that your workflows require to read input data or write output data. The **Pipeline work directory** bucket above is added by default to the list of **Allowed S3 buckets**. diff --git a/platform_versioned_docs/version-23.3.0/enterprise/configuration/configtables/features_env.yml b/platform_versioned_docs/version-23.3.0/enterprise/configuration/configtables/features_env.yml index c4d6a956..4cd31870 100644 --- a/platform_versioned_docs/version-23.3.0/enterprise/configuration/configtables/features_env.yml +++ b/platform_versioned_docs/version-23.3.0/enterprise/configuration/configtables/features_env.yml @@ -12,7 +12,7 @@ - Environment variable: '`TOWER_ENABLE_ARM64`' Description: > - Enable support for ARM64 CPU architecture. Required to enable Graviton instance support in AWS Batch compute environments. + Enable support for ARM64 CPU architecture (AWS Batch compute environments only). Required to enable Graviton instance support. Value: 'Default: `false`' - Environment variable: '`TOWER_ENABLE_AWS_SSM`' diff --git a/platform_versioned_docs/version-23.4.0/enterprise/release_notes/enterprise_latest.mdx b/platform_versioned_docs/version-23.4.0/enterprise/release_notes/enterprise_latest.mdx index b9bc32c5..dc62ba75 100644 --- a/platform_versioned_docs/version-23.4.0/enterprise/release_notes/enterprise_latest.mdx +++ b/platform_versioned_docs/version-23.4.0/enterprise/release_notes/enterprise_latest.mdx @@ -61,7 +61,7 @@ A previous iteration of the rebranded Seqera documentation noted `seqera` as the **ARM64 CPU architecture support enabled by default** -The `TOWER_ENABLE_ARM64` configuration environment variable is no longer needed to enable Graviton instance support in AWS Batch compute environments. ARM64 CPU architecture support is enabled by default. +The **Use Graviton CPU architecture** option is now available by default during AWS Batch compute environment creation. The `TOWER_ENABLE_ARM64` configuration environment variable is no longer needed to enable ARM64 CPU architecture support. ## Upgrade steps