X-Git-Url: https://git.arvados.org/arvados.git/blobdiff_plain/4aee7d57faff02fc6b7b6f750dc22a29e58bb963..HEAD:/sdk/cwl/arvados_cwl/arv-cwl-schema-v1.0.yml diff --git a/sdk/cwl/arvados_cwl/arv-cwl-schema-v1.0.yml b/sdk/cwl/arvados_cwl/arv-cwl-schema-v1.0.yml index dce1bd4d02..aeb41db568 100644 --- a/sdk/cwl/arvados_cwl/arv-cwl-schema-v1.0.yml +++ b/sdk/cwl/arvados_cwl/arv-cwl-schema-v1.0.yml @@ -240,6 +240,12 @@ $graph: MiB. Default 256 MiB. Will be added on to the RAM request when determining node size to request. jsonldPredicate: "http://arvados.org/cwl#RuntimeConstraints/keep_cache" + acrContainerImage: + type: string? + doc: | + The container image containing the correct version of + arvados-cwl-runner to use when invoking the workflow on + Arvados. - name: ClusterTarget type: record @@ -260,3 +266,225 @@ $graph: project_uuid: type: string? doc: The project that will own the container requests and intermediate collections + + +- name: OutputStorageClass + type: record + extends: cwl:ProcessRequirement + inVocab: false + doc: | + Specify the storage class to be used for intermediate and final output + fields: + class: + type: string + doc: "Always 'arv:StorageClassHint" + jsonldPredicate: + _id: "@type" + _type: "@vocab" + intermediateStorageClass: + type: + - "null" + - string + - type: array + items: string + doc: One or more storages classes + finalStorageClass: + type: + - "null" + - string + - type: array + items: string + doc: One or more storages classes + +- type: record + name: PropertyDef + doc: | + Define an arvados metadata property that will be set on a + container request or output collection. + fields: + - name: propertyName + type: string + doc: The property key + - name: propertyValue + type: [Any] + doc: The property value + + +- name: ProcessProperties + type: record + extends: cwl:ProcessRequirement + inVocab: false + doc: | + Specify metadata properties that will be set on the submitted + container request associated with this workflow or step. + fields: + class: + type: string + doc: "Always 'arv:ProcessProperties" + jsonldPredicate: + _id: "@type" + _type: "@vocab" + processProperties: + type: PropertyDef[] + jsonldPredicate: + mapSubject: propertyName + mapPredicate: propertyValue + + +- name: cwltool:CUDARequirement + type: record + extends: cwl:ProcessRequirement + inVocab: false + doc: | + Require support for NVIDA CUDA (GPU hardware acceleration). + fields: + class: + type: string + doc: 'cwltool:CUDARequirement' + jsonldPredicate: + _id: "@type" + _type: "@vocab" + cudaVersionMin: + type: string + doc: | + Minimum CUDA version to run the software, in X.Y format. This + corresponds to a CUDA SDK release. When running directly on + the host (not in a container) the host must have a compatible + CUDA SDK (matching the exact version, or, starting with CUDA + 11.3, matching major version). When run in a container, the + container image should provide the CUDA runtime, and the host + driver is injected into the container. In this case, because + CUDA drivers are backwards compatible, it is possible to + use an older SDK with a newer driver across major versions. + + See https://docs.nvidia.com/deploy/cuda-compatibility/ for + details. + cudaComputeCapability: + type: + - 'string' + - 'string[]' + doc: | + CUDA hardware capability required to run the software, in X.Y + format. + + * If this is a single value, it defines only the minimum + compute capability. GPUs with higher capability are also + accepted. + + * If it is an array value, then only select GPUs with compute + capabilities that explicitly appear in the array. + cudaDeviceCountMin: + type: ['null', int, cwl:Expression] + default: 1 + doc: | + Minimum number of GPU devices to request. If not specified, + same as `cudaDeviceCountMax`. If neither are specified, + default 1. + cudaDeviceCountMax: + type: ['null', int, cwl:Expression] + doc: | + Maximum number of GPU devices to request. If not specified, + same as `cudaDeviceCountMin`. + +- name: UsePreemptible + type: record + extends: cwl:ProcessRequirement + inVocab: false + doc: | + Specify a workflow step should opt-in or opt-out of using preemptible (spot) instances. + fields: + class: + type: string + doc: "Always 'arv:UsePreemptible" + jsonldPredicate: + _id: "@type" + _type: "@vocab" + usePreemptible: boolean + +- name: OutputCollectionProperties + type: record + extends: cwl:ProcessRequirement + inVocab: false + doc: | + Specify metadata properties that will be set on the output + collection associated with this workflow or step. + fields: + class: + type: string + doc: "Always 'arv:OutputCollectionProperties" + jsonldPredicate: + _id: "@type" + _type: "@vocab" + outputProperties: + type: PropertyDef[] + jsonldPredicate: + mapSubject: propertyName + mapPredicate: propertyValue + + +- name: KeepCacheType + type: enum + symbols: + - ram_cache + - disk_cache + doc: + - | + ram_cache: Keep blocks will be cached in RAM only. + - | + disk_cache: Keep blocks will be cached to disk and + memory-mapped. The disk cache leverages the kernel's virtual + memory system so "hot" data will generally still be kept in + RAM. + +- name: KeepCacheTypeRequirement + type: record + extends: cwl:ProcessRequirement + inVocab: false + doc: | + Choose keep cache strategy. + fields: + - name: class + type: string + doc: "'arv:KeepCacheTypeRequirement'" + jsonldPredicate: + _id: "@type" + _type: "@vocab" + - name: keepCacheType + type: KeepCacheType? + doc: | + Whether Keep blocks loaded by arv-mount should be kept in RAM + only or written to disk and memory-mapped. The disk cache + leverages the kernel's virtual memory system so "hot" data will + generally still be kept in RAM. + +- name: OutOfMemoryRetry + type: record + extends: cwl:ProcessRequirement + inVocab: false + doc: | + Detect when a failed tool run may have run out of memory, and + re-submit the container with more RAM. + fields: + - name: class + type: string + doc: "'arv:OutOfMemoryRetry" + jsonldPredicate: + _id: "@type" + _type: "@vocab" + - name: memoryErrorRegex + type: string? + doc: | + A regular expression that will be used on the text of stdout + and stderr produced by the tool to determine if a failed job + should be retried with more RAM. By default, searches for the + substrings 'bad_alloc' and 'OutOfMemory'. + - name: memoryRetryMultiplier + type: float? + doc: | + If the container failed on its first run, re-submit the + container with the RAM request multiplied by this factor. + - name: memoryRetryMultipler + type: float? + doc: | + Deprecated misspelling of "memoryRetryMultiplier". Kept only + for backwards compatability, don't use this.