added cluster launch

distroduce/bash/launch_cluster.sh

@@ -1,9 +0,0 @@
-#!/bin/bash
-aws emr create-cluster --auto-scaling-role EMR_AutoScaling_DefaultRole \
---applications Name=Hadoop Name=Hive Name=Spark \
---bootstrap-actions '[{"Path":"s3://insightde/emr/bootstraps/bootstrap.sh","Name":"bootstrap"}]' \
---ebs-root-volume-size 10 \
---ec2-attributes '{"KeyName":"joshua-IAM-keypair","InstanceProfile":"EMR_EC2_DefaultRole","SubnetId":"subnet-0034e615b047fd112","EmrManagedSlaveSecurityGroup":"sg-08e546ae27d86d6a3","EmrManagedMasterSecurityGroup":"sg-08e546ae27d86d6a3"}' \
---service-role EMR_DefaultRole --release-label emr-5.26.0 --name 'test_cluster12d' \
---instance-groups '[{"InstanceCount":1,"EbsConfiguration":{"EbsBlockDeviceConfigs":[{"VolumeSpecification":{"SizeInGB":32,"VolumeType":"gp2"},"VolumesPerInstance":2}]},"InstanceGroupType":"MASTER","InstanceType":"m4.xlarge","Name":"Master - 1"},{"InstanceCount":5,"EbsConfiguration":{"EbsBlockDeviceConfigs":[{"VolumeSpecification":{"SizeInGB":32,"VolumeType":"gp2"},"VolumesPerInstance":2}]},"InstanceGroupType":"CORE","InstanceType":"m4.xlarge","Name":"Core - 2"}]' --configurations '[{"Classification":"spark-env","Properties":{},"Configurations":[{"Classification":"export","Properties":{}}]},{"Classification":"spark-defaults","Properties":{}},{"Classification":"spark","Properties":{"maximizeResourceAllocation":"true"}}]' \
---scale-down-behavior TERMINATE_AT_TASK_COMPLETION --region us-east-1

distroduce/bash/spark.json

@@ -1,26 +0,0 @@
-[
-  {
-    "Classification": "spark-env",
-    "Configurations": [
-        {
-            "Classification": "export",
-            "ConfigurationProperties": {
-                "PYSPARK_PYTHON": "/usr/bin/python3",
-                "PYSPARK_DRIVER_PYTHON": "/usr/bin/python3"
-            }
-        }
-    ]
-  },
-  {
-    "Classification": "spark-defaults",
-    "ConfigurationProperties": {
-      "spark.sql.execution.arrow.enabled": "true"
-    }
-  },
-  {
-    "Classification": "spark",
-    "Properties": {
-      "maximizeResourceAllocation": "true"
-    }
-  }
-]

distroduce/configuration/distroduce.sh

@@ -5,7 +5,6 @@ yes | sudo python3 -m pip install pathos kafka-python
 wget https://raw.githubusercontent.com/JEJodesty/cadCAD/dev/dist/cadCAD-0.0.2-py3-none-any.whl
 yes | sudo python3 -m pip install cadCAD-0.0.2-py3-none-any.whl
 
-
 # check for master node
 IS_MASTER=false
 if grep -i isMaster /mnt/var/lib/info/instance.json | grep -i true;

distroduce/configuration/launch_cluster.py

@@ -0,0 +1,125 @@
+import os, json, boto3
+from typing import List
+
+ec2_attributes = {
+    "KeyName":"joshua-IAM-keypair",
+    "InstanceProfile":"EMR_EC2_DefaultRole",
+    "SubnetId":"subnet-0034e615b047fd112",
+    "EmrManagedSlaveSecurityGroup":"sg-08e546ae27d86d6a3",
+    "EmrManagedMasterSecurityGroup":"sg-08e546ae27d86d6a3"
+}
+
+bootstrap_actions = [
+    {
+        "Path":"s3://insightde/emr/bootstraps/distroduce.sh",
+        "Name":"bootstrap"
+    }
+]
+
+instance_groups = [
+    {
+        "InstanceCount":5,
+        "EbsConfiguration":
+            {"EbsBlockDeviceConfigs":
+                [
+                    {
+                            "VolumeSpecification":
+                                {"SizeInGB":32,"VolumeType":"gp2"},
+                            "VolumesPerInstance":2
+                    }
+                ]
+            },
+        "InstanceGroupType":"CORE",
+        "InstanceType":"m4.xlarge",
+        "Name":"Core - 2"
+    },
+    {
+        "InstanceCount":1,
+        "EbsConfiguration":
+            {
+                "EbsBlockDeviceConfigs":
+                    [
+                        {
+                            "VolumeSpecification":
+                                {"SizeInGB":32,"VolumeType":"gp2"},
+                            "VolumesPerInstance":2
+                        }
+                    ]
+            },
+        "InstanceGroupType":"MASTER",
+        "InstanceType":"m4.xlarge",
+        "Name":"Master - 1"
+    }
+]
+
+configurations = [
+    {
+        "Classification":"spark-env",
+        "Properties":{},
+        "Configurations":
+            [
+                {
+                    "Classification":"export",
+                    "Properties":{
+                        "PYSPARK_PYTHON": "/usr/bin/python3",
+                        "PYSPARK_DRIVER_PYTHON": "/usr/bin/python3"
+                    }
+                }
+            ]
+    },
+    {
+        "Classification":"spark-defaults",
+        "Properties":{
+            "spark.sql.execution.arrow.enabled": "true"
+        }
+    },
+    {
+        "Classification":"spark",
+        "Properties":{
+            "maximizeResourceAllocation":"true"
+        }
+    }
+]
+
+
+
+def create_distroduce_cluster(name, region, ec2_attributes, bootstrap_actions, instance_groups, configurations):
+    def log_uri(name, region):
+        return f's3n://{name}-{region}/elasticmapreduce/'
+
+    os.system(f"""
+        aws emr create-cluster \
+        --applications Name=Hadoop Name=Hive Name=Spark \
+        --ec2-attributes '{json.dumps(ec2_attributes)}' \
+        --release-label emr-5.26.0 \
+        --log-uri '{str(log_uri(name, region))}' \
+        --instance-groups '{json.dumps(instance_groups)}' \
+        --configurations '{json.dumps(configurations)}' \
+        --auto-scaling-role EMR_AutoScaling_DefaultRole \
+        --bootstrap-actions '{json.dumps(bootstrap_actions)}' \
+        --ebs-root-volume-size 10 \
+        --service-role EMR_DefaultRole \
+        --enable-debugging \
+        --name '{name}' \
+        --scale-down-behavior TERMINATE_AT_TASK_COMPLETION \
+        --region {region}
+    """)
+
+
+def benchmark(names: List[str], region, ec2_attributes, bootstrap_actions, instance_groups, configurations):
+    current_dir = os.path.dirname(__file__)
+    s3 = boto3.client('s3')
+    bucket = 'insightde'
+
+    file = 'distroduce.sh'
+    abs_path = os.path.join(current_dir, file)
+    key = f'emr/bootstraps/{file}'
+
+    s3.upload_file(abs_path, bucket, key)
+    for name in names:
+        create_distroduce_cluster(name, region, ec2_attributes, bootstrap_actions, instance_groups, configurations)
+
+
+name = 'distibuted_produce'
+region = 'us-east-1'
+benchmark([name], region, ec2_attributes, bootstrap_actions, instance_groups, configurations)

distroduce/configuration/launch_cluster.sh

@@ -0,0 +1,16 @@
+#!/bin/bash
+aws emr create-cluster \
+--applications Name=Hadoop Name=Hive Name=Spark \
+--ec2-attributes '{"KeyName":"joshua-IAM-keypair","InstanceProfile":"EMR_EC2_DefaultRole","SubnetId":"subnet-0034e615b047fd112","EmrManagedSlaveSecurityGroup":"sg-08e546ae27d86d6a3","EmrManagedMasterSecurityGroup":"sg-08e546ae27d86d6a3"}' \
+--release-label emr-5.26.0 \
+--log-uri 's3n://aws-logs-251682129355-us-east-1/elasticmapreduce/' \
+--instance-groups '[{"InstanceCount":5,"EbsConfiguration":{"EbsBlockDeviceConfigs":[{"VolumeSpecification":{"SizeInGB":32,"VolumeType":"gp2"},"VolumesPerInstance":2}]},"InstanceGroupType":"CORE","InstanceType":"m4.xlarge","Name":"Core - 2"},{"InstanceCount":1,"EbsConfiguration":{"EbsBlockDeviceConfigs":[{"VolumeSpecification":{"SizeInGB":32,"VolumeType":"gp2"},"VolumesPerInstance":2}]},"InstanceGroupType":"MASTER","InstanceType":"m4.xlarge","Name":"Master - 1"}]' \
+--configurations '[{"Classification":"spark-env","Properties":{},"Configurations":[{"Classification":"export","Properties":{}}]},{"Classification":"spark-defaults","Properties":{}},{"Classification":"spark","Properties":{"maximizeResourceAllocation":"true"}}]' \
+--auto-scaling-role EMR_AutoScaling_DefaultRole \
+--bootstrap-actions '[{"Path":"s3://insightde/emr/bootstraps/distroduce.sh","Name":"bootstrap"}]' \
+--ebs-root-volume-size 10 \
+--service-role EMR_DefaultRole \
+--enable-debugging \
+--name 'distibuted_produce' \
+--scale-down-behavior TERMINATE_AT_TASK_COMPLETION \
+--region us-east-1

distroduce/configuration/spark_submit.sh

@@ -1,4 +1,3 @@
 #!/bin/bash
 PRIVATE_IP=`hostname -I | xargs`
-sudo sed -i -e '$a\export PYSPARK_PYTHON=/usr/bin/python3' /etc/spark/conf/spark-env.sh
 spark-submit --master yarn --py-files distroduce.zip messaging_sim.py $PRIVATE_IP

distroduce/configuration/toEC2.sh

@@ -1,5 +1,4 @@
 #!/bin/bash
-
 scp -i ~/.ssh/joshua-IAM-keypair.pem ~/Projects/event-bench/event_bench.py \
 hadoop@ec2-3-230-158-62.compute-1.amazonaws.com:/home/hadoop/