[dlc] add resource prepare support (#5065)

examples/trials/mnist-pytorch/config_dlc_resource.yml

+# working directory on DSW, please provie FULL path
+searchSpaceFile: search_space.json
+# the command on trial runner(or, DLC container), be aware of data_dir
+trialCommand: python mnist.py --data_dir /root/data/{your_data_dir}
+trialConcurrency: 1  # NOTE: please provide number <= 3 due to DLC system limit.
+maxTrialNumber: 10
+tuner:
+  name: TPE
+  classArgs:
+    optimize_mode: maximize
+# ref: https://help.aliyun.com/document_detail/203290.html?spm=a2c4g.11186623.6.727.6f9b5db6bzJh4x
+trainingService:
+  platform: dlc
+  type: Worker
+  image: registry-vpc.cn-beijing.aliyuncs.com/pai-dlc/pytorch-training:1.6.0-gpu-py37-cu101-ubuntu18.04
+  jobType: PyTorchJob                             # choices: [TFJob, PyTorchJob]
+  podCount: 1
+  ecsSpec: "{'resource_id':'rg19d2oleg252kke','cpu':2,'memory':8,'gpu':0,'gputype':'','shared_memory':''}" # resource ID，e.g., rg19d2oleg252kke
+  region: cn-hangzhou
+  workspaceId: ${your_workspace_id}
+  accessKeyId: ${your_ak_id}
+  accessKeySecret: ${your_ak_key}
+  nasDataSourceId: ${your_nas_data_source_id}     # NAS datasource ID，e.g., datat56by9n1xt0a
+  ossDataSourceId: ${your_oss_data_source_id}     # optional, OSS data source id.
+  localStorageMountPoint: /home/admin/workspace/  # default NAS path on DSW, MUST provide full path.
+  containerStorageMountPoint: /root/data/         # default NAS path on DLC container, change it according your setting

ts/nni_manager/config/dlc/dlcUtil.py

@@ -8,6 +8,7 @@ import pathlib
 import sys
 import traceback
 import time
+import ast
 from argparse import ArgumentParser
 # ref: https://help.aliyun.com/document_detail/203290.html?spm=a2c4g.11186623.6.727.6f9b5db6bzJh4x
 from alibabacloud_pai_dlc20201203.client import Client
@@ -71,15 +72,6 @@ if __name__ == "__main__":
                 data_source_id=args.oss_data_source_id,
             )
 
-        # job spec
-        spec = JobSpec(
-            type=args.type,
-            image=args.image,
-            pod_count=args.pod_count,
-            ecs_spec=args.ecs_spec,
-        )
-
-        
         if args.workspace_id == 'None':
             args.workspace_id = None
             logging.info("args.workspace_id %s %s",args.workspace_id,type(args.workspace_id))
@@ -87,14 +79,56 @@ if __name__ == "__main__":
         data_sources = [nas_1]
         if oss:
             data_sources = [nas_1, oss]
-        req = CreateJobRequest(
-            display_name=args.experiment_name,
-            job_type=args.job_type,
-            job_specs=[spec],
-            data_sources=data_sources,
-            user_command=args.user_command,
-            workspace_id=args.workspace_id,
-        )
+
+        if args.ecs_spec[0] == '{' and args.ecs_spec[-1] == '}':
+            config = ast.literal_eval(args.ecs_spec)
+            resource_id = config['resource_id']
+            cpu = config.get('cpu',1)
+            memory = config.get('memory',2)
+            gpu = config.get('gpu',0)
+            gputype = config.get('gputype',"")
+            shared_memory = config.get('shared_memory',"")
+
+            spec = JobSpec(
+                type=args.type,
+                image=args.image,
+                pod_count=args.pod_count,
+                resource_config=ResourceConfig(
+                    cpu=str(cpu),
+                    memory=str(memory) + "Gi",
+                    gpu=str(gpu),
+                    gputype=str(gputype),
+                    shared_memory=str(shared_memory)
+                ),
+            )
+
+            # 声明任务的执行内容。
+            req = CreateJobRequest(
+                display_name=args.experiment_name,
+                job_type=args.job_type,
+                job_specs=[spec],
+                data_sources=data_sources,
+                user_command=args.user_command,
+                workspace_id=args.workspace_id,
+                resource_id=str(resource_id),
+            )
+        else:
+            # job spec
+            spec = JobSpec(
+                type=args.type,
+                image=args.image,
+                pod_count=args.pod_count,
+                ecs_spec=args.ecs_spec,
+            )
+
+            req = CreateJobRequest(
+                display_name=args.experiment_name,
+                job_type=args.job_type,
+                job_specs=[spec],
+                data_sources=data_sources,
+                user_command=args.user_command,
+                workspace_id=args.workspace_id,
+            )
 
         response = client.create_job(req)
         job_id = response.body.job_id
@@ -135,5 +169,4 @@ if __name__ == "__main__":
 
                         
     except Exception as e:
-        logging.exception('DLC submit Exception: \n')
-
+        logging.exception('DLC submit Exception: \n')
\ No newline at end of file