Administering Fabric

In this chapter, we will cover how to enable Fabric for a tenant, the options for overwriting the tenant-level settings at a capacity level, and how to associate a capacity with a workspace. We will also cover the different types of capacities and how tenants, capacities, and workspaces are all tied together. This information is vital to being able to administer Fabric effectively across teams and capacities.

We will go through the following topics in this chapter:

• Enabling Microsoft Fabric in your tenant
• What are capacities?
• Managing Spark job configurations

Before diving into capacities and capacity management, it is important to understand the tenant-level settings that are available for tenant administrators of Microsoft Fabric.

Enabling Microsoft Fabric in your tenant

Fabric is governed at two levels: tenant and capacity. By default, Fabric is enabled on all capacities and for all users in a tenant. Figure 8.1 shows the granular controls that allow admins to select specific security groups that will have access to Fabric rather than every individual in the organization. This can be useful if you are not ready to make Fabric widely available but instead would like to have smaller groups or departments evaluate the functionality and perform reviews before rolling out to the larger enterprise. To modify the tenant-level setting, navigate to Admin portal | Tenant settings | Microsoft Fabric, where you will be presented with the following options:

Figure 8.1 – Tenant-level settings for enabling Microsoft Fabric

Some organizations will want to manage Fabric more granularly at the capacity level. Capacity-level settings allow you to disable Fabric at the tenant level while selectively...

What are capacities?

Every operation needs compute power. For each workload in Fabric, the optimal type of compute may look slightly different. Where SQL runs best on CPUs, Spark will be better on GPUs. Some workloads, such as Power BI, require large amounts of memory, while others such as Data Factory may just need small amounts of CPU for orchestrating activities. Fabric capacities remove the complexity of choosing the right type and size compute for each operation that is run by providing a single, blended compute metric called a capacity unit that is delivered through Fabric capacities, often referred to as a Fabric SKU or simply an F SKU.

Therefore, Fabric capacity is simply a pool of available resources that is shared by all the workspaces assigned to a capacity and is shared by all the workloads running in those workspaces. Based on the type of operation running, the appropriate blend of compute and memory will be allocated and torn down after the operation is complete because...

Managing Spark job configurations

Microsoft Fabric is based on the Software-as-a-Service (SaaS) model, which provides features such as auto-integration, auto-optimization, and auto-tuning for all the workloads, including the fully managed Spark runtime, which powers data engineering and data science experiences. While the default configuration will work fine in most cases, empowering you to do more with less, there are times when you’ll need more control. For these types of scenarios, you can control several aspects of the Fabric Spark runtime and Spark jobs to suit your business needs.

Open any Fabric workspace, click on the Workspace settings icon at the top, and then, under Data Engineering/Science, select Spark compute. You will find various options to make changes, as shown in Figure 8.4:

Figure 8.4 – Spark compute settings

In the next few sections, we will look at Spark configurations, including starter and custom Spark pools, configuring...

Summary

In this chapter, you learned how to enable Fabric in your tenant, what capacities are, how to manage capacities, how to override tenant settings at the capacity level, and how to manage Spark settings for workspaces. The flexibility provided by the Fabric capacity model allows organizations to decide how much capacity is needed, when it is needed, and which workspaces need it. The added benefit of the Fabric SaaS model is that it removes the need to complicate capacity planning by assigning the right kind of compute at runtime through a serverless infrastructure that delivers the right CPU/GPU/memory configuration based on the workload requesting the compute resources.

In the next chapter, we will discuss the importance of security and governance in Fabric and how to apply best practices across the entire tenant.