Cleanup (proddata#2)

* Remove duplicate function
* Remove other variants of the scripts
* Add eslint
* Minor formatting fixes
* Fix indentation
* Replace usages of ++/--
* Remove unused randomFloat method
* Add dangling commas
* Use dot notation for environment variables
* Allow configuring SSL
* No need to pass stats_global as a parameter
* Use const where applicable
* Use cammel case consistently for naming
* Use ===
* Declare functions before using them
* Disable some additional eslint rules for now
* Update README.md

.eslintrc.js

@@ -0,0 +1,23 @@
+module.exports = {
+  env: {
+    browser: false,
+    es2021: true,
+  },
+  extends: [
+    'airbnb-base',
+  ],
+  parserOptions: {
+    ecmaVersion: 'latest',
+    sourceType: 'module',
+  },
+  rules: {
+    "no-console": "off",
+    "quotes": "off",
+    "one-var": "off",
+    "one-var-declaration-per-line": "off",
+    "no-inner-declarations": "off",
+    "no-use-before-define": "off",
+    "no-restricted-syntax": "off",
+    "guard-for-in": "off",
+  },
+};

README.md

@@ -1,56 +1,60 @@
-# NodeIngestBench
+# Node.js CrateDB Ingest Benchmark
 
-a simple nodejs script to quickly test ingest performance on CrateDB clusters.
+A multi-process Node.js script to run high-performance ingest benchmarks on CrateDB clusters.
+The script generates random data and runs batched insert statements against a single table using CrateDB's [HTTP endpoint](https://crate.io/docs/crate/reference/en/4.7/interfaces/http.html).
 
-top measured performance with single nodejs client : 1 200 000 rows / s
+The top measured performance (single Node.js process, seven-node CrateDB cluster with 30 vCPUs each) was 1,200,000 rows/s.
 
-## new improved version (app.js)
+## Setup
 
-Ubuntu 
-
-```sh
-apt update && apt upgrade
-curl -sL https://deb.nodesource.com/setup_14.x | sudo -E bash -
-apt update
-sudo apt-get install -y nodejs
-npm install
-node app.js --batchsize 20000 --max_rows 1000000 --table "doc.benchmark" --shards 12 --concurrent_requests 20
-```
-
-
-```env
+1. Install Node.js using one of the available [installation methods](https://nodejs.org/en/download/current/)
+2. Clone this repository: `git clone https://github.com/proddata/nodeIngestBench.git`
+3. Change into the cloned repository: `cd nodeIngestBench`
+4. Install dependencies: `npm install`
+5. Configure the connection to your CrateDB cluster by creating a `.env` file:
+```bash
 CRATE_HOST=localhost
 CRATE_USER=admin
 CRATE_PASSWORD=secret-password
-CRATE_DB=doc
 CRATE_PORT=4200
 CRATE_SSL=true
 ```
 
+## Running benchmarks
+
+Start the benchmarking with `node appCluster.js`. The script takes several optional command-line parameters:
+ * `table`: The name of the table used for benchmarking. The table will automatically be created if it doesn't exist yet.
+ * `shards`: The number of shards that will be allocated for the table.
+ * `dropTable`: If `true`, the table will be dropped and re-created when running the script.
+ * `batchsize`: The number of rows that are inserted as part of a single `INSERT` statement.
+ * `processes`: The number of child processes that will be spawned. Each child process inserts data independently.
+ * `concurrent_requests`: Each child process will run this number of concurrent queries.
+ * `max_rows`: The number of rows after which a child process will terminate. Overall, the (maximum) number of rows that will be inserted is `processes * max_rows`.
 
-## example run with app.js
+## Example
 
-3 nodes (10 vCPUs) CrateDB cluster
+Below is a basic example that was run on a three-node CrateDB cluster with 10 vCPUs each:
 
 ```bash
-node app.js --batchsize 20000 --max_rows 1000000 --table "doc.benchmark" --shards 12 --concurrent_requests 20
+$ node appCluster.js --batchsize 20000 --max_rows 1000000 --shards 12 --concurrent_requests 20 --processes 1
+
 -------- Options ---------
 {
   dropTable: true,
+  processes: 1,
   batchsize: 20000,
   max_rows: 6000000,
-  table: 'doc.benchmark',
+  table: 'doc.cpu',
   shards: 6,
   concurrent_requests: 20
 }
 --------------------------
-... Buffer updated
--------- Results ---------
+[...]
+-------- Global Results ---------
 Time	 25.179 s
 Rows	 6,020,000 records
-Speed	 239,088.13 rows per sec
--------- Results ---------
+Speed	239,088.13 rows per sec
+---------------------------------
 ```
 
-=> 2,400,000 Metrics / sec
-
+As each row contains ten numeric metrics, this is equal to a throughput of 2,400,000 metrics/s.