It’s ALIVE!!! ArcGIS Server: The beating heart of ArcGIS Enterprise

It’s ALIVE!!! ArcGIS Server is not dead. On the contrary, it’s far from it! So how should it be cared for and fed?

Greetings, denizens of the dark world of ArcGIS Enterprise administration.

When we think about ArcGIS Enterprise, we often think of a collection of (body) parts that we stitch together and animate to replicate ArcGIS Online (plus a few added advantages) on our own infrastructure. Over the years, significant attention has been paid to system functionality and ease of use. When it comes to the care and feeding of an Enterprise deployment, however, we know that we must manage the multiple parts to make the whole function smoothly and not be a shambling hulk.

Buried below the surface of ArcGIS Enterprise is ArcGIS Server. We often forget that Server is a vital part—the heart, as I like to refer to it—of our deployments. Keeping that heart healthy and beating and far from the realm of the undead is paramount.

Let’s look at some of the things that are crucial to keeping that ticker ticking.

Deployment architecture: Stitching the parts together

To build the body of the deployment, we connect the constituent parts: ArcGIS Server, Portal for ArcGIS, ArcGIS Web Adaptor and ArcGIS Data Store. Piecing them together to ensure they work properly in conjunction with each other can seem like complex surgery. Despite their seeming disparity, however, they will rise and work together, with proper care and understanding.

Try this tutorial on ArcGIS Enterprise base deployments to get started. 

Ensuring that proper resources are available to feed the beast is essential. Although the minimum requirements will result in a working deployment, they will likely not yield exactly the level of performance that you need. It’s recommended that you get a better understanding of ArcGIS Enterprise system requirements before getting started. 

Connecting the parts with each other properly ensures that they will work in unison. This diagram of essential ArcGIS Enterprise ports will give you more context for stitching the parts of your deployment into a working whole. 

Service tuning: Keeping the heart beating

Once the deployment takes its first breath, the heart must be kept beating. Any content that has a functional capacity is based on a service which, in turn, is published on an ArcGIS Server instance of a specific role. Just like the muscles of the body, these services have their strengths and weaknesses, and must be kept in tune to provide not only the function but also the desired capacity. Understanding the nature of the different service types, their capabilities and characteristics—minimum and maximum instances, timeouts, shared vs. dedicated instances, etc.—will allow the administrator to keep the heart from experiencing an arrest.

Read more about ArcGIS Server licensing roles, how to configure service instance settings, and understanding services, layers and maps.

Server capacity: Exhuming the Capacity Planning Tool

Once upon a time, there was a mad scientist by the name of Dave Peters. Dave is the genius who literally wrote the book on ArcGIS Server (Building a GIS: System Architecture Design Strategies for Managers, unfortunately now out of print. If you can get your hands on a copy, do it! This author doesn’t advocate stealing from colleagues, but now that it’s out there…). 

Peters’ genius is so much so, that he created something known as the Capacity Planning Tool (CPT). The CPT is a free utility that can help anticipate the capacity and responsiveness of a Server site using current and future hardware and infrastructure metrics as well as service load to determine the service time from the users’ perspective. While the CPT hasn’t been updated since the Mad Scientist went into retirement, it can be updated with current hardware metrics to be relevant today.

Server monitoring: Monitoring the vital signs

Finally! The deployment is walking the village streets. To keep it on its feet, the Doctor must monitor its vital signs. Many instruments and tools exist for this purpose. One of the best, which has been designed for the ArcGIS Enterprise ecosystem, is ArcGIS Monitor. Taking a page from Dave Peters and continuing his work, Monitor allows administrators to track the overall performance of their deployment and discover if any component or service is experiencing problems. The administrator can then diagnose the issue. Monitor’s capabilities include optimizing the deployment and monitoring its health, troubleshooting system and performance issues, and generating metrics on the deployment. 

An alternative option is to use service information available via the ArcGIS Server Administrator REST API and the ArcGIS API for Python, which can be extracted and presented in an ArcGIS Dashboards application.

ArcGIS Enterprise and ArcGIS Server are reasonably complex systems that take effort to create and manage. It doesn’t take a mad scientist to keep the heart beating, however. Through knowledge, understanding and diligence, you too can successfully construct and launch a deployment. The system does not need to be perceived as a monster.

Want to learn more? Consider the following training courses to build and develop the skills necessary to develop and maintain an ArcGIS Enterprise deployment. There’s even a course to teach you how to use the Capacity Planning Tool called System Architecture Design Strategies.

For the deepest dive into the inner workings of ArcGIS Server (and ArcGIS Enterprise), however, I recommend signing up for the ArcGIS Enterprise: Administration Workflows course. You’ll discover best practices for managing servers, data and services while ensuring system performance over time.

Want to stay informed about all the latest training opportunities at Esri Canada? Visit Esri Canada’s Communication Preference Centre and select the “Training” checkbox to get a monthly roundup straight to your inbox.