Making Bulk Updates in Azure Cosmos DB

azure cosmosdb how-to
20 Sep 2025

Here is quick guide on how to perform bulk updates in Azure Cosmos DB using the .NET SDK.

The Cosmos DB .NET SDK provides built-in support for bulk operations, which can significantly improve the performance of your application when you need to perform multiple operations in a single request. Below is an example of how to perform bulk updates using the SDK.

using Microsoft.Azure.Cosmos;
using System;
using System.Collections.Generic;
using System.Threading.Tasks;
using System.Linq;

var cosmosClient = new CosmosClient("your-cosmos-db-connection-string", new CosmosClientOptions
{
    AllowBulkExecution = true
});
var database = cosmosClient.GetDatabase("your-database-id");
var container = database.GetContainer("your-container-id");

var tasks = new List<Task>(1000); // Adjust the size based on your needs, here we are preparing for 1000 updates in one go
var cost = 0D;
foreach (var item in itemsToUpdate) // itemsToUpdate is a collection of items you want to update
{
    var task = container.CreateItemAsync(item, new PartitionKey(item.PartitionKeyValue));
    // Optionally, you can add continuation tasks to handle success or failure
    tasks.Add(task.ContinueWith(t =>
    {
        if (t.Status == TaskStatus.RanToCompletion)
        {
            cost += t.Result.RequestCharge;
            Console.WriteLine($"Item with id {item.Id} updated successfully.");
        }
        else
        {
            Console.WriteLine($"Failed to update item with id {item.Id}: {t.Exception?.Message}");
        }
    }));
}
await Task.WhenAll(tasks);
Console.WriteLine($"Total Request Charge for bulk update: {cost:0.##} RUs");

In this example, we build a list of tasks to perform asynchronous updates on items in a Cosmos DB container. We iterate over the collection of items to update, create a task for each operation, and add it to the list. Finally, we use Task.WhenAll to run all tasks concurrently.

Each task is configured with a continuation callback that checks whether the operation succeeded or failed and logs the result accordingly.

Benefits of this approach:

  • Efficiency: By batching multiple operations into a single request, you reduce the number of network round-trips, which can significantly improve performance.
  • Concurrency: The SDK handles the concurrency of operations, allowing you to perform many updates in parallel.
  • Backend Requests: The SDK optimizes the number of backend requests, which can help reduce costs associated with request units (RUs).

Important Considerations

  • Throughput: Ensure that your Cosmos DB container has sufficient throughput (RU/s) to handle the bulk operations. You may need to increase the provisioned throughput temporarily during bulk operations. Here are a few more important considerations when performing bulk updates in Azure Cosmos DB:

  • Error handling and retries: Bulk operations may encounter transient failures or rate limiting (429 Too Many Requests). Implement robust retry logic with exponential backoff to handle these gracefully and ensure data consistency.

  • Idempotency: Ensure your update operations are idempotent or handle duplicates carefully, since retries or partial failures might cause an operation to be executed multiple times.

  • Request size and batch limits: Keep each batch within Cosmos DB’s maximum request size (2 MB) and operation count limits to avoid request rejections.

  • Throughput scaling: If bulk updates consume a high number of RUs, consider scaling throughput with autoscale or manually increasing provisioned RU/s temporarily during the operation.

  • Data modelling impacts: Bulk updates might affect indexing policy performance; review indexing options to optimize update costs.

Paired with effective throughput planning, these measures ensure that bulk updates in Cosmos DB remain efficient, reliable and cost-effective.