Update README.md

renaissance/README.md

@@ -1,20 +1,13 @@
-This directory contains the scripts required to run and get the results of the Renaissance Benchmark taking Prometheus metrics.
+
 # About The Benchmark
 The Renaissance Benchmark Suite aggregates common modern JVM workloads, including, but not limited to, Big Data, machine-learning, and functional programming. The suite is intended to be used to optimize just-in-time compilers, interpreters, GCs, and for tools such as profilers, debuggers, or static analyzers, and even different hardware. It is intended to be an open-source, collaborative project, in which the community can propose and improve benchmark workloads.
 
 More information about this benchmark can be found on [Renaissance](https://github.com/renaissance-benchmarks/renaissance)
 # Prerequisites
 To generate the results from the Renaissance Benchmark,we need to:  
 
- -Firstly,install a kubernetes cluster(here minikube),which can be done from [Minikube](https://minikube.sigs.k8s.io/docs/start/) or for a more convinient process can be installed from the Kruize Autotune repo.  [Autotune Installation](https://github.com/kruize/autotune/blob/master/docs/autotune_install.md). 
-
--Secondly,we need to install prometheus on minikube.This can be done through the Autotune repo itself(refer the installation process) [Autotune Installation](https://github.com/kruize/autotune/blob/master/docs/autotune_install.md). 
+ - Install minikube(kubernetes cluster),which can be done from [Minikube](https://minikube.sigs.k8s.io/docs/start/) and then install prometheus on the minikube cluster. This can be done by following the steps in the [Autotune Installation](https://github.com/kruize/autotune/blob/master/docs/autotune_install.md). 
 
- -Then, we need to clone the scripts required for running the renaissance benchmark which can be done from [Benchmarks](https://github.com/Prakalp23/benchmarks/tree/renaissance) repository.This contains a number of performance scripts that we have built for the Renaissance benchmark to collect the data from the Prometheus metrics and parse the results.
- Then,clone this repo for a reference to results and for getting the Python scripts to determine the accuracy of various ML algorithms on the dataset collected
-
- `git clone https://github.com/Prakalp23/autotune-results.git`
-
  You,also need to install a driver of your choice for running  renaissance onto your local system
 
  Download a driver (docker or podman)
@@ -27,11 +20,11 @@ To generate the results from the Renaissance Benchmark,we need to:
 
  To run the benchmark on kubernetes cluster to collect performance metrics
 
- ./renaissance-run.sh --clustertype=CLUSTER_TYPE -s BENCHMARK_SERVER -e RESULTS_DIR_PATH [-w WARMUPS] [-m MEASURES] [-i TOTAL_INST] [--iter=TOTAL_ITR] [-r= set redeploy to true] [-n NAMESPACE] [-g RENAISSANCE_IMAGE] [--cpureq=CPU_REQ] [--memreq=MEM_REQ] [--cpulim=CPU_LIM] [--memlim=MEM_LIM] [-b BENCHMARKS] [-R REPETITIONS] [-d DURATION] "
+ ./scripts/perf/renaissance-run.sh --clustertype=CLUSTER_TYPE -s BENCHMARK_SERVER -e RESULTS_DIR_PATH [-w WARMUPS] [-m MEASURES] [-i TOTAL_INST] [--iter=TOTAL_ITR] [-r= set redeploy to true] [-n NAMESPACE] [-g RENAISSANCE_IMAGE] [--cpureq=CPU_REQ] [--memreq=MEM_REQ] [--cpulim=CPU_LIM] [--memlim=MEM_LIM] [-b BENCHMARKS] [-R REPETITIONS] [-d DURATION] "
 
  - **CLUSTER_TYPE**: Type of cluster. Supports openshift , minikube.
 - **BENCHMARK_SERVER**: Name of the cluster you are using
-- **RESULTS_DIR**: Directory to store results
+- **RESULTS_DIR_PATH**: Directory to store results
 - **DURATION**: Duration of each warmup and measurement run.
 - **WARMUPS**: No.of warmup runs.
 - **MEASURES**: No.of measurement runs.
@@ -41,18 +34,16 @@ To generate the results from the Renaissance Benchmark,we need to:
 - **CPU_LIM**: CPU limit
 - **MEM_LIM**: Memory limit
 - **RENAISSANCE_IMAGE**:prakalp23/renaissance1041:latest 
+- **BENCHMARKS**:Choice of a microbenchmark from Renaissance [Microbenchmarks](https://github.com/renaissance-benchmarks/renaissance)
 
 Example:./renaissance-run.sh --clustertype=minikube -s localhost -e ./results -w 1 -m 1 -i 1 --iter=1 -r -n default  --cpureq=1.5 --memreq=3152M --cpulim=1.5 --memlim=3152M -b "page-rank" -g prakalp23/renaissance1041:latest -d 60
 
  # The Experiment Results
 
- The experiment results,the generated csv file,and the scripts to test the accuracy of various ML Algorithms on the given dataset can be found on the Autotune-results page 
+ The experiment results using the above scripts generates a csv file which contains the resource usage information for the Renaissance Benchmark can be found here
 
  [Renaissance Results](https://github.com/Prakalp23/autotune-results/tree/renaissance/Renaissance)
 
- # Issues
- To run the Renaissance benchmark,you need to have it installed locally on your system,and have use the -g command everytime you trigger a run using
- `./renaissance-run.sh --clustertype=minikube -s localhost -e ./results -w 1 -m 1 -i 1 --iter=1 -r -n default  --cpureq=1.5 --memreq=3152M --cpulim=1.5 --memlim=3152M -b "page-rank" -g prakalp23/renaissance1041:latest -d 60`
 
 ## Scripts Details