Shulou(Shulou.com)05/31 Report--

In this issue, the editor will bring you about how to choose monitoring tools for DevOps and SRE. The article is rich in content and analyzes and narrates it from a professional point of view. I hope you can get something after reading this article.

How to select monitoring tools for DevOps and SRE

Summary: when developing reliability or implementing flexible DevOps practices, the core of decision-making is data. If you do not carefully monitor key metrics such as uptime, network load, and resource usage, you will ignore where to spend energy on development or improve operational practices. Fortunately, a variety of monitoring tools can be used to help you collect and view this data.

When developing reliability or implementing resilient DevOps practices, data is at the heart of the decision. If you do not carefully monitor key metrics such as uptime, network load, and resource usage, you will ignore where to spend energy on development or improve operational practices. Fortunately, a variety of monitoring tools can be used to help you collect and view this data.

While it may be tempting to try to fully monitor everything in the system, more centralized monitoring will be easier to implement and provide you with more actionable data. SRE practices such as SLO are most useful when based on metrics of customer impact. Determining what and how to monitor is an important decision. In this blog post, we will show you the basics. We will also suggest some popular monitoring tools for your consideration.

Where is the monitoring implemented?

It is important to determine where monitoring is implemented in the system architecture. This will enable you to develop the architecture around monitoring tools without having to modify existing code. Depending on the location of the implementation, monitoring tools will be able to observe different types of data. The following is a classification of the most common types of monitoring implementations and examples of tools that provide this type of monitoring:

Resource monitoring: also known as server monitoring or infrastructure monitoring, it operates by collecting data about how the server operates. The resource monitoring tool reports RAM usage, CPU load, and remaining disk space. In an architecture with physical servers, information about the health of the hardware, such as CPU temperature and component uptime, also helps to avoid server failures. In a cloud-based environment, aggregation of virtual server systems is more useful.

Network monitoring: this will view data coming in and out of the computer network. Your monitoring tool captures all incoming requests and outgoing responses in all components (such as switches, firewalls, servers, etc.). The data collected from network monitoring can be as simple as the total amount of data back and forth, or as subtle as the frequency of specific requests.

Application performance monitoring: the APM solution collects data about overall service performance. These tools send their own requests to the service and track metrics such as the speed and integrity of the response. The goal is to facilitate the detection and diagnosis of application performance problems to ensure that the service runs at the expected level.

Third-party component monitoring: this involves monitoring the health and availability of third-party components in the architecture. In this era of microservices, your service may depend on the proper operation of external services, from cloud hosting to advertising servers. Like application performance monitoring, tools can check the status of these services based on their own requests.

You may need to include some monitoring in each type of monitoring in the overall solution. Priority is given to the use of robust redundant monitoring tools to ensure that potential problems are not missed. At the same time, metrics and alerts should be bound to services to ensure that they are relevant to business impact.

What do you need from the data?

Having actionable data is not just about the data itself. In order to respond correctly to what the monitoring tool reports, you need to display the data in the most useful way. Monitoring tools can do something for you:

Trigger an alarm to create an event log when the indicator exceeds a certain threshold, create a dashboard that provides key service health components at a glance according to the parameter highlighting, create a log database that can be queried when making development decisions or responding to events, get into the habit of asking yourself, "what do I need to think about now in order to make the best choice?" What data and important metrics will be included in the visualization.

Another important point to consider for open source and purchase is where you will find monitoring tools and who will maintain them. Open source and purchasable tools have their own advantages and disadvantages.

Open source monitoring tool

These tools are free, which is an advantage for companies with limited tool budgets. They are also fully customizable, allowing you to integrate them into your own architecture. However, this customization will require specialized development time and perhaps specialized knowledge. In addition, there is no SLA guarantee for availability, security, update frequency, etc. Your team will take on these responsibilities.

Purchased monitoring tools

These tools are expensive, but they have powerful features that open source tools cannot provide. The service provider will be responsible for keeping the tool functional and up-to-date. The provider may provide customer service, training, documentation, and other resources to help you integrate tools with the stack. In the age of reliability, it is worth considering investing to ensure that the eyes of surveillance are always open.

Comparison of monitoring tools

Here are the 10 most popular SRE and DevOps monitoring tools considered for your system.

AppDynamics is a monitoring platform focused on APM. Other features they provide include AI-based insights, end-user monitoring for simulating customer journeys, and business monitoring with integrated revenue analysis. You can register for a free trial.

DataDog is a monitoring platform for cloud-scale services. It has powerful functions in visualization, alarm, data merging and analysis. They associate performance metrics with business impact. DataDog offers a free trial.

Prometheus is a popular open source monitoring tool that provides alerts, queries, visualization, and many other useful features. The dedicated development community provides a large number of documentation and instructions to help you get started.

New Relic is a monitoring platform that provides several components that can also be used independently: New Relic APM (Application performance Monitoring), New Relic Browser and New Relic Infrastructure. They provide applications for iOS and Android, giving you more monitoring options.

Nagios offers open source (Nagios Core) and purchasable options (Nagios XI). They provide a highly customizable interface and can monitor the entire IT network. They also highlight their ease of use through the configuration wizard to guide users in setting up new monitoring services.

Dynatrace allows cross-team collaboration with its monitoring platform to provide a shared single repository of monitoring data. They also include autonomous cloud capabilities and the ability to introduce monitoring capabilities into the deployed Internet of things layer. They also offer free trials.

Solarwinds offers several products, each dedicated to different areas of monitoring: network management, system management, database management, IT security, IT service management, application management and hosting service providers. Each can be tried for free.

Site24x7 specializes in website monitoring and provides tools such as status pages and health diagnostics for Web services such as AWS and Azure. They also provide comprehensive Web transaction monitoring, allowing you to simulate usage and collect metrics. They offer several pricing plans based on the services they need.

SignalFx provides a wide range of microservice integration so that you can see a complete picture of the health of the service. This is important if your service contains many third-party components. Their focus is to help you build your architecture from a single model to a micro-service model.

PRTG Network Monitor is a complete monitoring service that can be integrated into many stages and locations of the architecture. They provide monitoring on the network, on a single server, on specific applications, and on everything in between. The provider also provides a free version.

This is how to choose monitoring tools for DevOps and SRE. If you happen to have similar doubts, you might as well refer to the above analysis to understand. If you want to know more about it, you are welcome to follow the industry information channel.

Welcome to subscribe "Shulou Technology Information " to get latest news, interesting things and hot topics in the IT industry, and controls the hottest and latest Internet news, technology news and IT industry trends.