refactor code

Signed-off-by: kusumachalasani <kchalasa@redhat.com>

Author: kusumachalasani

monitoring/local_monitoring/ReadMe.md

@@ -28,9 +28,9 @@ By default, it runs on the `Kind` cluster.
 ```
 
 ```
-Usage: ./local_monitoring_demo.sh [-s|-t] [-c cluster-type] [-e recommendation_experiment] [-l] [-p] [-r] [-i kruize-image] [-u kruize-ui-image] [-b] [-n namespace] [-d load-duration] [-m benchmark-manifests] [-g num of unpartitioned gpu resources]
+Usage: ./local_monitoring_demo.sh [-s|-t] [-c cluster-type] [-e recommendation_experiment] [-l] [-p] [-r] [-i kruize-image] [-u kruize-ui-image] [-b] [-n namespace] [-d load-duration] [-m benchmark-manifests]
 c = supports minikube, kind and openshift cluster-type
-e = supports container, namespace and gpu
+e = supports container, namespace and gpu. Default - none.
 i = kruize image. Default - quay.io/kruize/autotune_operator:<version as in pom.xml>
 l = Run a load against the benchmark
 p = expose prometheus port
@@ -41,18 +41,27 @@ b = deploy the benchmark.
 n = namespace where benchmark is deployed. Default - default
 d = duration to run the benchmark load
 m = manifests of the benchmark
-g = number of unpartitioned gpu resources available
 ```
 
 ## Understanding the Demo
 
-This demo focuses on using the TFB (TechEmpower Framework Benchmarks) benchmark to simulate different load conditions and observe how Kruize-Autotune reacts with its recommendations. Here’s a breakdown of what happens during the demo:
-
-- TFB deployment in default Namespace
-    - The TFB benchmark is initially deployed in the default namespace, comprising two key deployments
-      - tfb-qrh: Serving as the application server.
-      - tfb-database: Database to the server.
-    - Load is applied to the server for 20 mins within this namespace to simulate real-world usage scenarios
+This demo focuses on installing kruize and also install the benchmarks if asked for through `-e` parameter.
+- By default, it installs kruize and provides the URL to access the kruize UI service where the user can create experiments and generate recommendations.
+- To use demo benchmarks to create and generate recommendations through a script, pass -e for container, namespace and gpu benchmarks.
+    - For container and namespace type, benchmark 'TFB' is deployed in a namespace.
+    - For gpu type, benchmark 'human-eval' is deployed.
+
+Here’s a breakdown of what happens during the demo:
+
+- Deploys benchmarks in a namespace (if -e is passed)
+    - If -e is container/namespace
+        - The TFB benchmark is initially deployed in the namespace, comprising two key deployments
+          - tfb-qrh: Serving as the application server.
+          - tfb-database: Database to the server.
+        - Load is applied to the server for 20 mins within this namespace to simulate real-world usage scenarios
+    - If -e is gpu
+        - The human-eval benchmark is deployed as job in the namespace.
+        - The job is set to run for atleast 20 mins to generate the recommendations.
 - Install Kruize
   - Installs kruize under openshift-tuning name.
 - Metadata Collection and Experiment Creation
@@ -62,6 +71,9 @@ This demo focuses on using the TFB (TechEmpower Framework Benchmarks) benchmark
   - Generates Recommendations for all the experiments created.
 
 ## Recommendations for different load Simulations observed on Openshift
+
+TFB (TechEmpower Framework Benchmarks) benchmark is simulated in different load conditions and below are the different recommendations observed from Kruize-Autotune.
+
 ### IDLE 
 - Experiment: `monitor_tfb-db_benchmark`
   - Shows an IDLE scenario where CPU recommendations are not generated due to minimal CPU usage (less than a millicore).

monitoring/local_monitoring/local_monitoring_demo.sh

@@ -68,7 +68,6 @@ export start_demo=1
 export APP_NAMESPACE="default"
 export LOAD_DURATION="1200"
 export BENCHMARK_MANIFESTS="resource_provisioning_manifests"
-export GPUS="0"
 export EXPERIMENT_TYPE=""
 # Iterate through the commandline options
 while getopts c:i:e:n:d:m:g:lbprstu: gopts
@@ -115,9 +114,6 @@ do
 		m)
 			BENCHMARK_MANIFESTS="${OPTARG}"
 			;;
-		g)
-			GPUS="${OPTARG}"
-			;;
 		*)
 			usage
 	esac
@@ -153,4 +149,5 @@ elif [ ${start_demo} -eq 2 ]; then
 	kruize_local_demo_update ${BENCHMARK} >> "${LOG_FILE}" 2>&1
 else
 	kruize_local_demo_terminate
+g = number of unpartitioned gpu resources available
 fi