Real-Time Monitoring 🖥️

PyRun offers real-time insights into your workload performance, helping you monitor CPU, disk, network, and memory metrics as your tasks execute. This guide will walk you through each section of the real-time monitoring dashboard, enabling you to understand and optimize your workloads effectively.

Real-Time Overview Tab

The Real-Time Overview tab provides a summary of key metrics across your workload. Here’s what each chart represents:

1. Average CPU Usage

This gauge-style chart shows the real-time CPU usage across your workload. The color-coded scale helps you quickly assess CPU intensity:

• Green indicates light usage, suitable for low-demand workloads.
• Yellow shows moderate usage, suggesting some load.
• Red indicates high CPU usage, signaling intensive processing that may require attention.

2. Average Disk & Network Usage (MB)

This chart gives you a snapshot of disk and network data usage in megabytes (MB):

• Disk Read and Disk Write indicate storage access levels.
• Network Read and Network Write show data transfer rates over the network. This chart helps you identify potential bottlenecks in data access or network performance.

3. Average Data Transfer Speed

This chart tracks the speed of data transfer, covering both disk and network read/write rates:

• Disk Read Rate and Disk Write Rate
• Network Read Rate and Network Write Rate These metrics are useful for evaluating how quickly data is accessed, supporting optimal workload performance.

4. Average CPU Utilization Breakdown

The CPU utilization breakdown shows where CPU time is being spent, divided into:

• System Time, User Time, IO Wait Time, Children System Time, and Children User Time This donut chart helps you see which processes are consuming CPU resources, enabling you to adjust workloads as needed.

Gantt Chart Tab

The Gantt Chart tab gives a detailed view of task durations and concurrent calls in real time. This chart helps you visualize:

• How long each task is taking (User Time, System Time, Upload Time, Other Time)
• The number of Active Calls at any point, showing peaks in task concurrency Use this view to understand task timing and spot areas where optimization could reduce wait times or processing delays.

CPU Usage Tab

In the CPU Metrics tab, you can track detailed CPU usage over time with these key metrics:

• avgUsage provides the average CPU usage.
• minUsage and maxUsage show the usage range.
• quantile80 and quantile20 represent higher and lower percentiles for finer-grained insights. This helps you analyze CPU load variability and spot peaks or lulls that might affect workload performance.

Disk Read/Write Usage Tab

The Disk Metrics tab offers insights into the volume of data being read and written on disk. Key metrics include:

• totalRead and totalWrite for total data accessed.
• minRead and maxRead for observing data access peaks and lows.
• quantile80Read and quantile20Read for percentile-based analysis. Monitoring disk usage is essential for data-intensive workloads, helping you avoid I/O bottlenecks.

Disk Read/Write Rate Tab

The Disk Read/Write Rate chart shows how fast data is read from and written to disk in MB/s:

• totalRead and totalWrite show aggregate data transfer.
• minRead and maxRead indicate the speed range.
• quantile80Read and quantile20Read help you understand typical vs. peak rates. Use this to check if data access speeds meet your workload’s needs or if storage limits are being reached.

Memory Usage Tab

The Memory Metrics tab displays your workload’s memory usage in real time:

• Min Usage and Max Usage provide the range of memory used.
• 80th Percentile and 20th Percentile show memory demand patterns. Tracking memory usage helps you ensure workloads are not exceeding memory capacity, which could lead to slowdowns or failures.

Network Read/Write Data Usage Tab

In Network Metrics, the Network Read/Write Data Usage chart tracks the amount of data moving across the network:

• sumRead and sumWrite represent total network data read and written.
• minRead and maxRead show minimum and maximum data usage.
• quantile80Read and quantile20Read offer percentile-based insights. This chart helps you identify if network constraints are affecting workload efficiency, especially during high-demand periods.

Network Rate Tab

Finally, the Network Rate chart monitors the read and write speed of network data in MB/s:

• sumRead and sumWrite show total data transfer rates.
• minRead and maxRead provide the speed range.
• quantile80Read and quantile20Read allow you to observe typical rates versus peak usage. This information is essential for workloads with high data transfer needs, ensuring your network can handle the demand.

PyRun's real-time monitoring dashboard provides the insights you need to optimize your workload performance, helping you quickly spot bottlenecks and adjust resources to maintain efficiency.