Solving the Problem of Azure Application Latency - Part Three: Dealing with Networking Issues

In this series, we’re exploring the foundational aspects of optimizing and monitoring applications within the Azure ecosystem using SolarWinds® Observability. In part one, we introduced application performance monitoring (APM) in SolarWinds Observability to identify latency issues in an Azure service. Then, in part two, we explored how to leverage Azure resource metrics in SolarWinds Observability infrastructure monitoring to address resource usage issues. 

The complexity of cloud environments requires a comprehensive approach to performance tuning so you can ensure your applications are efficient and responsive. To gain a comprehensive view of the infrastructure and application layers, you’ll need a mixture of networking metrics from Azure Monitor along with metrics related to APM. By taking this approach, you’ll be able to quickly pinpoint the root causes of latency and other performance issues. 

In this last part of our series, we’ll explore network troubleshooting within Azure. We'll guide you through a step-by-step approach, using real-world scenarios to illustrate how to leverage SolarWinds Observability and Azure Monitor together to diagnose and resolve networking issues. 

Integrating Azure Monitor with SolarWinds Observability 

Integrating Azure Monitor metrics with SolarWinds Observability provides a consolidated view of important metrics—including network insights—alongside APM metrics. To integrate Azure Monitor Metrics with SolarWinds Observability, follow these steps: 

First, log into SolarWinds Observability to access your SolarWinds Observability dashboard. From there, you can add an Azure cloud account. 

Add Azure Cloud Account

Choose the specific Azure Monitor metrics you want to integrate. In this demo, we’ll focus on network-related metrics for a targeted approach. 

Metrics explorer

Now your network metrics are being ingested, you can customize the SolarWinds dashboards to display Azure Monitor metrics alongside your APM metrics. These metrics could include network performance, latency, throughput, and other relevant data. 

Add the widget to the dashboard

Benefits of consolidated observability 

With Azure Monitor metrics integrated into SolarWinds Observability, organizations gain several advantages: 

• Holistic view: Provides a more comprehensive view of your cloud infrastructure's health and performance. 
• Enhanced troubleshooting: Correlates network performance issues with application performance metrics to accelerate troubleshooting. 
• Streamlined monitoring: Reduces context-switching between tools by combining all critical metrics into one dashboard. 
• Proactive management: Integrates the infrastructure view, helping teams to anticipate and address issues before they escalate. 

Identifying Networking Issues with SolarWinds Observability and Azure Monitor Metrics 

When end-users encounter high latency with an Azure-backed application, and this latency is related to network issues, identifying, and resolving the issue requires you to look at metrics across various segments of the infrastructure stack. 

Cross-reference SolarWinds Observability, Azure Monitor, and Application Performance Monitoring (APM) Metrics 

Combining SolarWinds Observability and Azure Monitor is essential for obtaining a comprehensive view of network health. As you focus on microservice interactions and dependencies in SolarWinds Observability, you can complement this with Azure Monitor's broader capabilities, such as tracking cloud resources and applications. Used together with APM insights into application performance, metrics from Azure Monitor will enhance your network analysis. Engineers can cross-reference data from these sources to gain a multi-dimensional understanding of network performance. 

 
Imagine encountering a sudden spike in latency when monitoring your application's performance using APM metrics. By examining and cross-referencing SolarWinds Observability data, you can determine if the latency spike corresponds to a specific service interaction within your application. 

Next, consulting Azure Monitor reveals any related alerts or resource constraints triggered simultaneously. This holistic approach allows you to pinpoint the root cause of network issues and also aids in preemptive monitoring.  

Benefits of unified performance views 

A unified performance view fosters a more proactive approach to IT management. Comprehensive visibility across the network, applications, and infrastructure helps engineers better anticipate potential issues before they escalate into critical problems. Within a single dashboard, you can identify anomalies, diagnose root causes, and implement steps for resolution.  
 
Consider a scenario where a service running in Azure experiences slowdowns (latency) during peak traffic periods. Leveraging SolarWinds Observability and Azure Monitor, engineers can correlate the surge in traffic with resource constraints in the underlying infrastructure, such as database bottlenecks, insufficient server capacity, or network constraints. 

Azure Monitoring

Equipped with this insight, your response team can scale resources or optimize network performance to mitigate the issue, ensuring a seamless end-user experience for your customers. 

Method for tracking down networking issues 

To effectively identify and isolate a network issue, you must take a systematic approach to troubleshooting in SolarWinds Observability and Azure Monitor.  

1. Begin with a broad overview of network health within Azure Monitor: Assess the network’s overall health, looking for any alerts or anomalies in resource utilization. 
2. Drill down into specifics with SolarWinds Observability: Dig deeper into the specifics of service interactions for potential issues, aiming to isolate the problem to a particular microservice or communication pathway. 
3. Cross-reference with APM metrics: Align SolarWinds Observability and Azure Monitor findings with metrics in the APM module built into SolarWinds Observability. This ensures that issues correlate with actual performance impacts observed at the application level. 
4. Utilize diagnostic tools from Azure Monitor: For deeper analysis and clarity, employ diagnostic tools from Azure Monitor, including log analytics and application insights. 
5. Resolution and follow-up: Maintain continuous monitoring to prevent future occurrences. 

This systematic and integrated approach simplifies network management in the cloud era, where complexity is met with sophisticated, easy-to-use monitoring solutions. 

Azure netflows over time in SolarWinds Observability 

Strategies to Tune Azure and Address Networking Issues 

As cloud infrastructures evolve into more complex systems, effective performance tuning becomes increasingly important, especially for networking. To help address these complexities and optimize network performance, it’s important to explore various strategies and approaches. In this section, we’ll look at specific techniques to help you tackle the challenges and complexities of modern cloud networks. 

Resource scaling 

Resource scaling is one of the most direct ways to address networking issues in Azure. This involves adjusting networking resources—such as bandwidth and virtual network gateways—based on demand. For example, during peak usage times, scaling up bandwidth can alleviate network congestion, while scaling down during off-peak hours can reduce costs. 

Right-sizing  

This strategy is about aligning resources with networking workload requirements. It involves assessing and allocating the appropriate type and size of networking resources. For example, choosing the right size of Azure Virtual Network (VNet) and selecting appropriate network interface types can significantly influence performance. 

Network optimization 

Network optimization encompasses several techniques designed to enhance network performance. These include: 

• Optimizing routing paths. 
• Employing Azure Traffic Manager for load balancing. 
• Addressing bottlenecks arising from suboptimal network configurations.  

 Regularly reviewing network topology and adjusting based on traffic patterns is key to this strategy. 

Network topology visualization

Demonstration and Examples 

Leveraging SolarWinds Observability, you can analyze network faults, availability, and performance. This will help you quickly identify and address a bottleneck causing network latency, thus reducing network downtime​​. Imagine a scenario where you’re experiencing frequent connection timeouts in your Azure cloud environment. How would you troubleshoot and respond? 

1. Integrate SolarWinds Observability with Azure

To set up an integration between SolarWinds Observability and Azure, begin by installing the SolarWinds Observability Agent on your cloud platform host(s). 

2. Monitor network performance with SolarWinds tools

Next, set up met network monitoring. Begin by configuring the SolarWinds Network Performance Monitor and NetFlow Traffic Analyzer. For detailed guidance on this, reference the official documentation. 

3. Analyze Azure performance over time 

For ongoing monitoring and analysis of your Azure environment, prioritize the use of SolarWinds Observability as it automatically discovers newly created virtual machines and containers. As your Azure infrastructure map changes, SolarWinds Observability will instantly update you on the current state of your Azure setup.  

Example Dashboard 

Customize the SolarWinds Observability dashboards to display the most relevant metrics specific to your Azure environment. Make sure to focus on the key performance indicators that matter most to your business or application. 

 As SolarWinds Observability continuously monitors your application and the underlying network, set up alerts on the key metrics to notify you when specific thresholds are breached. This will help you address performance issues promptly, preventing potential disruptions and downtime. 

Create alert condition

Alert trends over time 

4. Scale your services based on SolarWinds Observability network findings 

Finally, scale your services based on your findings. SolarWinds Observability collects a wide array of telemetry, providing real-time analysis of network performance to identify trends and potential bottlenecks. Armed with this critical information, you can: 

• Proactively scale services up or down, ensuring that applications remain responsive without incurring unnecessary costs. 
• Improve your operational efficiency.  
• Automate scaling based on precise network findings, reducing manual intervention and allowing for more agile responses to changing demands. 

5. Implement monitoring best practices for effective networking 

Effective networking in Azure is not a one-time setup but an ongoing process that requires constant monitoring and adjustments. Key best practices include: 

• Continuous monitoring: Lean on the observability platform to track key metrics, such as latency, packet loss, and throughput. This helps you promptly identify issues and trends that may require attention. 
• Establish traffic shaping policies: Implement policies that control network traffic, ensuring optimal performance and preventing congestion. 
• Perform iterative tuning: Review networking configurations, iteratively tuning to adapt to the cloud environment’s dynamic workloads and user demands. 
• Utilize Azure's native tools: Leverage Azure's built-in monitoring and diagnostic tools—such as Network Watcher and Azure Network Performance Monitor—for in-depth insights into networking performance. 

 As you implement these strategies, you’ll significantly enhance your Azure network performance, leading to more reliable and efficient cloud operations. The key is to continually monitor, analyze, and adjust while ensuring that the networking infrastructure evolves in tandem with the changing demands of your cloud environment. 

Conclusion 

This last installment of our series explored how to diagnose and resolve network issues within the Azure ecosystem using SolarWinds Observability and Azure Monitor. We walked through the integration of these tools with your Azure environment, also using Azure Resource Manager for metric collection along with the robust suite of monitoring tools from SolarWinds Observability to analyze and improve network performance. 

Throughout this series, we’ve worked to bring you a cohesive understanding of optimizing and maintaining high-performing applications on Azure. Explore the entire series to gain a holistic view of how to leverage SolarWinds Observability for addressing latency issues and enhancing application performance on Azure. 

Sign up for a free SolarWinds Observability trial and get started with monitoring applications on Azure.