Plumb through disk and memory overrides for ExtractTask [VS-1124]

.dockstore.yml

@@ -166,6 +166,7 @@ workflows:
          branches:
              - master
              - ah_var_store
+             - rsa_vs_1124
          tags:
              - /.*/
    - name: GvsImportGenomes
@@ -235,6 +236,7 @@ workflows:
          branches:
              - master
              - ah_var_store
+             - rsa_vs_1124
          tags:
              - /.*/
    - name: GvsWithdrawSamples

scripts/variantstore/wdl/GvsExtractCallset.wdl

@@ -18,6 +18,8 @@ workflow GvsExtractCallset {
     String query_project = project_id
     # This is optional now since the workflow will choose an appropriate value below if this is unspecified.
     Int? scatter_count
+    Int? memory_override
+    Int? disk_override
     Boolean zero_pad_output_vcf_filenames = true
 
     # set to "NONE" if all the reference data was loaded into GVS in GvsImportGenomes
@@ -211,6 +213,8 @@ Int effective_split_intervals_disk_size_override = select_first([split_intervals
         max_last_modified_timestamp        = GetBQTablesMaxLastModifiedTimestamp.max_last_modified_timestamp,
         extract_preemptible_override       = extract_preemptible_override,
         extract_maxretries_override        = extract_maxretries_override,
+        disk_override                      = disk_override,
+        memory_override                    = memory_override,
         emit_pls                           = emit_pls,
         emit_ads                           = emit_ads,
         write_cost_to_db                   = write_cost_to_db,
@@ -304,6 +308,8 @@ task ExtractTask {
     File? gatk_override
     Int? extract_preemptible_override
     Int? extract_maxretries_override
+    Int? disk_override
+    Int? memory_override
 
     Int? local_sort_max_records_in_ram = 10000000
 
@@ -389,8 +395,8 @@ task ExtractTask {
   >>>
   runtime {
     docker: gatk_docker
-    memory: "12 GB"
-    disks: "local-disk 150 HDD"
+    memory: select_first([memory_override, 12]) + " GB"
+    disks: "local-disk " + select_first([disk_override, 150]) + " HDD"
     bootDiskSizeGb: 15
     preemptible: select_first([extract_preemptible_override, "2"])
     maxRetries: select_first([extract_maxretries_override, "3"])

scripts/variantstore/wdl/GvsExtractCohortFromSampleNames.wdl

@@ -26,14 +26,16 @@ workflow GvsExtractCohortFromSampleNames {
     String extraction_uuid
     String filter_set_name
     String output_file_base_name
-    Int? scatter_count
 
     String? output_gcs_dir
     # set to "NONE" if all the reference data was loaded into GVS in GvsImportGenomes
     String drop_state = "NONE"
 
     Int? extract_preemptible_override
     Int? extract_maxretries_override
+    Int? extract_scatter_count_override
+    Int? extract_memory_override
+    Int? extract_disk_override
     Int? split_intervals_disk_size_override
     Int? split_intervals_mem_override
 
@@ -56,17 +58,19 @@ workflow GvsExtractCohortFromSampleNames {
   call Utils.GetBQTableLastModifiedDatetime as SamplesTableDatetimeCheck {
     input:
       project_id = query_project,
-      fq_table = "~{gvs_project}.~{gvs_dataset}.sample_info"
+      fq_table = "~{gvs_project}.~{gvs_dataset}.sample_info",
+      cloud_sdk_docker = effective_cloud_sdk_docker
   }
 
   call Utils.GetNumSamplesLoaded {
     input:
       fq_sample_table = "~{gvs_project}.~{gvs_dataset}.sample_info",
       project_id = gvs_project,
-      sample_table_timestamp = SamplesTableDatetimeCheck.last_modified_timestamp
+      sample_table_timestamp = SamplesTableDatetimeCheck.last_modified_timestamp,
+      cloud_sdk_docker = effective_cloud_sdk_docker
   }
 
-  Int effective_scatter_count = if defined(scatter_count) then select_first([scatter_count])
+  Int effective_scatter_count = if defined(extract_scatter_count_override) then select_first([extract_scatter_count_override])
                                 else if GetNumSamplesLoaded.num_samples < 100 then 50 # Quickstart
                                      else if GetNumSamplesLoaded.num_samples < 1000 then 250
                                           else if GetNumSamplesLoaded.num_samples < 5000 then 500
@@ -123,6 +127,8 @@ workflow GvsExtractCohortFromSampleNames {
       extract_maxretries_override = extract_maxretries_override,
       split_intervals_disk_size_override = split_intervals_disk_size_override,
       split_intervals_mem_override = split_intervals_mem_override,
+      memory_override = extract_memory_override,
+      disk_override = extract_disk_override,
 
       gatk_override = gatk_override,
       write_cost_to_db = write_cost_to_db