↓ Skip to main content
Cloud / Azure / Products / Durable Task Scheduler - Managed Backend for Durable Functions

Durable Task Scheduler - Managed Backend for Durable Functions

Azure Durable Task Scheduler is the recommended storage provider for Durable Functions and the Durable Task SDKs, with Dedicated and Consumption SKUs.

compute
Pricing Model Dedicated SKU: fixed monthly cost per Capacity Unit (CU); Consumption SKU: pay per action dispatched. Current pricing on the Azure Functions pricing page (includes Durable Task Scheduler pricing)
Availability Available in most Azure regions; current region list retrievable via Azure CLI
Data Sovereignty EU regions available (part of most Azure regions per the vendor)
Reliability SLA per provider (see official SLA page) SLA

What is Durable Task Scheduler?

Per Microsoft, the Durable Task Scheduler provides durable execution in Azure: a fault-tolerant approach to running code that handles failures and interruptions through automatic retries and state persistence. It is the recommended storage provider for Durable Functions and the Durable Task SDKs, and it addresses scenarios such as distributed transactions, multi-agent orchestration, data processing, and infrastructure management.

For Durable Functions, the Durable Task Scheduler can be used with any Azure Functions SKU; for the Durable Task SDKs, with any compute. The scheduler runs as a standalone Azure resource, separate from the application, with dedicated compute and memory for dispatching orchestrator, activity, and entity work items and for storing and querying history.

Core Features

  • Two billing SKUs: The Dedicated SKU provides predictable performance through preallocated Capacity Units (up to 3 CUs, up to 2,000 actions/sec per CU, up to 90 days of data retention, high availability from 3 CUs). The Consumption SKU offers pay-as-you-use billing per action dispatched, supports up to 500 actions per second, and retains data for up to 30 days.
  • Operational separation: The scheduler runs as a separate resource, reducing CPU and memory consumption in the app compared with self-managed storage providers, isolating faults, and scaling independently of the app.
  • Task hubs: Logical containers for orchestration and entity instances; multiple task hubs per scheduler allow isolating environments (dev/test/prod) or teams, each with its own monitoring dashboard.
  • Dashboard and emulator: A built-in dashboard allows filtering, drill-down, and management operations such as pausing, terminating, or restarting orchestrations. A local emulator running in Docker enables development and testing without deploying to Azure.
  • Autopurge retention: Configurable automatic cleanup of stale orchestration data to optimize storage usage.

Typical Use Cases

Distributed transactions: Multi-step business processes with automatic retries on failure and persistent state across system boundaries.

Multi-agent orchestration: Coordinating multiple AI agents or steps in agentic workflows with reliable state management.

Data processing: Long-running processing pipelines that survive interruptions and reliably resume progress.

Infrastructure management: Multi-step deployment or change workflows with error handling and traceability through the dashboard.

Benefits

  • A managed backend service, separate from the app, reduces resource consumption and isolates faults
  • Independent scaling of the scheduler relative to the application, including sharing one scheduler across multiple teams or projects
  • Built-in monitoring dashboard for orchestrations and entities without separate tooling integration
  • Local emulator enables development and testing without Azure resources
  • Two SKUs for different requirements: predictable cost with Dedicated, variable workloads with Consumption

Integration with innFactory

As a Microsoft Solutions Partner, innFactory helps you choose between the Dedicated and Consumption SKU for your Durable Functions or Durable Task SDK workloads, plan capacity based on your action volume, and connect via private endpoints while distinguishing this service from self-managed storage providers.

Contact us for a no-obligation consultation on Durable Task Scheduler.

Typical Use Cases

Storage provider for Durable Functions and the Durable Task SDKs for reliable, fault-tolerant orchestrations
Distributed transactions with automatic retries and state persistence
Multi-agent orchestration for AI and agentic workloads
Data-processing pipelines with long-running, interruptible flows
Infrastructure-management workflows with multiple sequential steps

Technical Specifications

0th Durable Task Scheduler is the recommended storage provider for Durable Functions and the Durable Task SDKs; usable with any Azure Functions SKU for Durable Functions, and with any compute for the Durable Task SDKs
1st Two billing SKUs: Dedicated SKU (fixed monthly cost per Capacity Unit, up to 2,000 actions/sec per CU, up to 3 CUs, up to 90 days of data retention, high availability from 3 CUs) and Consumption SKU (pay per action, up to 500 actions/sec, up to 30 days of data retention, no high availability)
2nd Quota limits: Dedicated SKU up to 25 schedulers and 25 task hubs per region per subscription; Consumption SKU up to 10 schedulers and 5 task hubs per region per subscription
3rd Each scheduler runs as a standalone Azure resource (Microsoft.DurableTask/scheduler) with dedicated compute and memory, separate from the app, connected via a TLS-secured gRPC connection
4th Task hubs are logical containers for orchestration and entity instances; one scheduler can host multiple task hubs for different environments or teams
5th Includes a dashboard for monitoring and managing orchestrations (filtering, pausing, terminating, restarting) and a local emulator for development in Docker
6th Maximum payload size of 1 MB each for orchestrator/activity inputs and outputs, external event data, custom status, and entity state; orchestration instance IDs limited to 100 characters

Frequently Asked Questions

What is the difference between the Dedicated SKU and the Consumption SKU?

The Dedicated SKU provides a fixed monthly cost per Capacity Unit (CU) with up to 2,000 actions per second per CU, up to 3 CUs, up to 90 days of data retention, and high availability from 3 CUs. The Consumption SKU bills per action dispatched, supports up to 500 actions per second, retains data for up to 30 days, and offers no high availability. Per Microsoft, Dedicated suits production workloads with predictable volume, while Consumption suits dev/test and variable workloads.

Does Durable Task Scheduler replace the previous Azure Storage provider for Durable Functions?

Per Microsoft, Durable Task Scheduler is the recommended storage provider for Durable Functions and the Durable Task SDKs. It runs as a separate, managed resource decoupled from the app, which reduces resource consumption in the app and enables independent scaling compared with self-managed (BYO) storage providers.

How much does Durable Task Scheduler cost?

Cost has two components: the scheduler fee based on the chosen SKU (Dedicated: fixed price per CU; Consumption: pay per action dispatched) and the compute resources of the hosting platform, billed separately, such as Azure Functions for Durable Functions or a container/VM platform for the Durable Task SDKs. Per Microsoft, current pricing is on the Azure Functions pricing page, which also includes Durable Task Scheduler pricing.

How do applications connect to the Durable Task Scheduler?

Applications connect via a TLS-secured gRPC connection, authenticated by the app's identity. For private connectivity, private endpoints route traffic through a virtual network instead of the public internet. Work items are streamed from the scheduler to the app using a push model, removing the need for polling.

Note: All product information on this page has been compiled with care, but is provided without guarantee and may be outdated or incomplete. Cloud services evolve rapidly — features, pricing, SLAs, and availability change frequently. Authoritative and up-to-date information can only be found on the official product page of Azure (official documentation). This page does not represent an offer by Azure.

Microsoft Solutions Partner

innFactory is a Microsoft Solutions Partner. We provide expert consulting, implementation, and managed services for Azure.

Microsoft Solutions Partner Microsoft Data & AI

Similar Products from Other Clouds

Other cloud providers offer comparable services in this category. As a multi-cloud partner, we help you choose the right solution.

Google Cloud

Cluster Toolkit - Open Source Blueprints for HPC and AI Clusters

Cluster Toolkit simplifies deploying HPC, AI, and ML clusters on Google Cloud with customizable, open-source blueprints …

Pricing Cluster Toolkit itself is a free …
SLA SLA per provider for each provisioned Google Cloud service (see official SLA page)
Compare →
AWS

Research and Engineering Studio on AWS: Open Source Virtual Desktop Portal

Research and Engineering Studio on AWS (RES) is an open source web portal for Windows and Linux virtual desktops for …

Pricing Per the documentation, RES itself is …
SLA SLA per provider (see official SLA page)
Compare →
STACKIT

STACKIT Compute Engine - Virtual Machines

STACKIT Compute Engine: scalable VMs from German and Austrian data centers. GDPR-compliant, OpenStack-based.

Pricing Pay-as-you-go (hourly billing)
SLA SLA as published by the provider
Compare →
STACKIT

STACKIT Compute Engine GPU - GPU Instances

STACKIT Compute Engine GPU: NVIDIA H100, A100, and L40S from German data centers for AI training and inference, …

Pricing Pay-per-use (hourly billing)
SLA SLA as published by the provider
Compare →
STACKIT

STACKIT Linux Server - Managed Linux VMs

STACKIT Linux Server: VMs with Ubuntu, Debian, RHEL, Rocky Linux, AlmaLinux and openSUSE, operated in German and …

Pricing Compute Engine hourly price per vCPU/RAM …
SLA SLA as published by the provider
Compare →
STACKIT

STACKIT Red Hat Enterprise Linux - RHEL VMs

STACKIT RHEL: Red Hat Enterprise Linux images from German data centers, SAP-certified. GDPR compliant.

Pricing Pay-per-use for VM and RHEL license …
SLA SLA as published by the provider
Compare →

50 comparable products found across other clouds.

Ready to start with Durable Task Scheduler - Managed Backend for Durable Functions?

Our certified Azure experts help you with architecture, integration, and optimization.

Schedule Consultation