The woes of siloed monitoring and how to leave them behind in minutes via Azure Marketplace
Published Oct 04 2021 09:04 AM 1,674 Views

In this guest blog post, Karl Kalash, Product Marketing Manager at Datadog, examines the problems that can arise from traditional siloed monitoring, how Datadog’s approach unifies information, and how the Microsoft Azure Marketplace simplifies purchasing and invoicing for customers of the software-as-a-service (SaaS) solution:

Modern applications are built with a multi-tier, service-oriented architecture. Problems that affect end users can occur at any level in the tech stack, and they need to be caught before issues cascade and become critical. Traditional siloed monitoring, in which each team uses its own monitoring tool and users are constantly chasing the red or tracking the yellow, is no longer sufficient. All teams do not have a single source of truth and are often blindsided at critical moments. Context switching and finger-pointing are often the end result. This impacts every reason that led you to the cloud in the first place: It negatively impacts your scalability, reduces agility, increases costs, affects cloud migration and platform modernization projects, and limits application optimization.

The problems are magnified by thousands of instances distributed over the cloud and on-premises environments. Siloed monitoring tools limit your team’s ability to respond to challenges at the cloud scale or adopt new technologies. Such tool and agent sprawl is expensive, both in terms of costs and computing resource consumption; it's also highly inefficient because of agent and data redundancy. Metrics indicate something is wrong, but they lack details. Traces can point to the general area, but aren’t specific. Logs have high volume and varying value, with cost and visibility tradeoffs. Everything must be manually queried and correlated.

This leads to reactive teams, excessive alerts, and constant firefighting. Poor quality of life for your operations teams could impact not just performance, but also hiring and growth. And most importantly, you lack the business context of what matters to your end users or customers.


By contrast, Datadog collects data from all servers, containers, applications, and cloud-based services with a single unified agent. The same agent that collects metrics also collects traces and logs. All that data is sent to Datadog’s SaaS platform, which enables numerous levels of monitoring:

● Infrastructure monitoring for back-end servers and cloud-based instances
● Serverless monitoring for functions and serverless applications
● Application performance monitoring on the application server side, which gives you insights aggregated across all requests
● Log management for all front-end and back-end logs, allowing you to analyze the details of issues
● Synthetics for proactive testing of application programming interface (API) endpoints and websites to detect issues before they impact customers
● Real user monitoring for your browsers and mobile devices
● Network performance to manage and optimize traffic flow across the entire stack


With everything enabled in one platform, teams gain a single source of truth with zero context switching. Datadog's machine learning and automated correlations enable insights and actionable information, from tracking service level objectives to real-time root-cause analysis and integrations with collaboration and workflow tools.

Customers can go from shopping for a Datadog plan in the Azure Marketplace to visualizing real-time Azure metrics in their Datadog account in minutes. They can streamline the purchase, configuration, and management of Datadog as a first-class service in the Azure portal.


Image 1.jpg


Datadog’s listing in the Azure Marketplace simplifies the purchasing and invoicing for new customers by having their usage appear directly on their Azure invoice, consolidating their Datadog and Azure costs. Once customers have purchased a Datadog plan through the Azure Marketplace, they’ll immediately start receiving standard Azure Monitor metrics in their new Datadog account. To further simplify the setup process, customers can enable single sign-on with Azure Active Directory during Datadog account creation.

Customers can see all their Azure data in the same place as data from across the rest of their stack in other clouds or on premises, thereby streamlining their migration process to Azure.


Image 2.jpg


Datadog’s Azure integration pulls in every metric users see in Azure and allows them to monitor new Azure services in real time automatically. The log-forwarding process from Azure to Datadog has also been completely automated: Rather than building out a log-forwarding pipeline with Diagnostic Settings, Event Hubs, and Functions, customers can configure everything with just a few clicks.

This makes it easier than ever to leverage Datadog’s powerful observability, compliance, and security capabilities for a successful cloud journey in Azure. The Datadog App Service Extension can also be added, managed, and deployed to our customer’s Azure Web Apps through the Datadog resource blade, simplifying the APM setup process.


image 3.jpg



Version history
Last update:
‎Oct 04 2021 11:51 AM
Updated by: