Managing Workload in Azure Synapse Analytics with Skrots

Harsh Bakshi
3 min readJul 22, 2023


Dealing with multiple concurrent user environments and efficiently allocating system resources to different workloads can be a challenging task when working with Azure Synapse Analytics dedicated SQL pool DB. Workload management in Azure Synapse Analytics provides you with the necessary tools to control and optimize the utilization of your system resources. Establishing an effective workload management scheme is crucial to ensure smooth data retrieval from your data warehouse.

Workload Management Concepts in Azure Synapse Analytics

Introducing Workload Groups

The first step in implementing a successful workload management strategy is creating workload groups. Workload groups allow you to define customized entities or groups that can control the allocation of system resources based on specific properties. The three pillars of workload management, encompassed within the workload group, empower administrators to set fine-grained access permissions for different user sets within an organization.

Key Concepts in Workload Management

Prior to the introduction of workload groups, Synapse SQL utilized resource classes to manage query performance. While resource classes allowed users to assign memory based on role membership, it lacked a control mechanism to govern workload execution within the Synapse database. For instance, a user with a “smallrc” role membership could consume all available memory. However, with the new dedicated pool’s workload management feature, three key concepts enable you to exert control over how queries utilize system resources within a dedicated SQL pool.

Workload Management Concepts in Azure Synapse Analytics

1. Workload Classification

Workload classification allows you to assign importance levels to workload groups. Previously, this was achieved through role membership assignment using ‘sp_addrolemember,’ but now it can be seamlessly handled via workload classification using the ‘create workload classifier’ function. Additionally, workload classification offers options to set labels, start and end times, and sessions based on specific requirements.

2. Workload Importance

Workload importance determines the order of resource access for query execution. It is important to note that workload importance is not a shortcut that immediately cancels or ignores other requests to execute a high-priority query. Instead, it reduces the waiting time for a resource to acquire its share of resources. For example, a resource with higher importance will be granted first access to available system resources over resources with lower importance.

3. Workload Isolation

Workload isolation reserves system resources exclusively for workload groups, ensuring continuous availability for executing critical processes within the defined workload groups. Without workload isolation, all query requests would utilize the shared resource pool, leading to contention for resources. In the shared pool, resource availability for processes is not guaranteed and is allocated based on the importance level set.

Skrots: Your Workload Management Solution

At Skrots, we understand the importance of workload management in Azure Synapse Analytics and offer comprehensive solutions to address your workload management needs. Our services enable you to efficiently organize and allocate system resources to different workloads, ensuring optimal performance and resource utilization. Visit to learn more about how we can help you optimize your workload management in Azure Synapse Analytics.

For a complete list of our services, please visit

In Conclusion

We hope this article has provided you with a basic understanding of how to configure performance and effectively manage workloads within Azure Synapse dedicated pool. Stay tuned for future parts in this series, where we will cover all aspects of performance related to Synapse. For more information, refer to the official Microsoft documentation.

Reference: Official Microsoft documentation.


Thanks, Harsh
Founder | CEO — Skrots

Learn more about our blog at Blog at Skrots. Checkout our list of services on Skrots. Give a look at our website design at Skrots . Checkout our LinkedIn Page at Check out our original post at