How to fix RequestLimitExceeded: Throughput exceeds the current throughput limit for your account in DynamoDB

DynamoDBADVANCEDHIGHUpdated Dec 22, 2025· 124 views

DynamoDB returns RequestLimitExceeded when your application exceeds account-level throughput limits, which apply to all tables in your AWS account. This error occurs when the combined throughput across all DynamoDB tables in your account exceeds AWS-imposed limits, requiring quota management and workload distribution strategies.

Quick fix

Fixes RequestLimitExceeded: Throughput exceeds the current throughput limit for your account

# Check Service Quotas for DynamoDB
aws service-quotas list-service-quotas \
  --service-code dynamodb \
  --region us-east-1

# Check specific quota for on-demand tables
aws service-quotas get-service-quota \
  --service-code dynamodb \
  --quota-code L-4BAB22A6 \
  --region us-east-1

# Request quota increase
aws service-quotas request-service-quota-increase \
  --service-code dynamodb \
  --quota-code L-4BAB22A6 \
  --desired-value 10000 \
  --region us-east-1
  1. Check current account limits and request quota increases
  2. Implement account-level rate limiting and backoff
  3. Distribute workload across multiple AWS accounts
  4. Monitor and analyze account-level metrics
dynamodb — error
# Check Service Quotas for DynamoDB
aws service-quotas list-service-quotas \
  --service-code dynamodb \
  --region us-east-1
# Check specific quota for on-demand tables
aws service-quotas get-service-quota \
  --service-code dynamodb \
  --quota-code L-4BAB22A6 \
  --region us-east-1
# Request quota increase
aws service-quotas request-service-quota-increase \
  --service-code dynamodb \
  --quota-code L-4BAB22A6 \
  --desired-value 10000 \
  --region us-east-1
RequestLimitExceeded: Throughput exceeds the current throughput limit for your account

What this error means

The RequestLimitExceeded error in DynamoDB indicates that your AWS account has exceeded its overall throughput limits for DynamoDB operations. Unlike table-level throttling (ThrottlingException), this is an account-level limit that applies to the combined throughput across all DynamoDB tables in your account. This error typically occurs in several scenarios: 1. **Account-wide throughput limits**: AWS imposes default throughput limits per account for both provisioned and on-demand capacity modes 2. **Multi-table applications**: When multiple tables in the same account are experiencing high throughput simultaneously 3. **On-demand capacity spikes**: Sudden, unexpected traffic spikes across multiple on-demand tables 4. **Control plane operations**: Excessive administrative operations (CreateTable, UpdateTable, etc.) hitting account limits AWS implements account-level limits to ensure fair resource allocation across all customers and maintain service stability. When these limits are exceeded, DynamoDB returns RequestLimitExceeded with HTTP status code 400.

Typical symptoms

  • HTTP 400 errors with "RequestLimitExceeded: Throughput exceeds the current throughput limit for your account" message
  • Multiple tables in the same AWS account experiencing throttling simultaneously
  • Increased error rates during peak traffic periods across the entire application
  • CloudWatch metrics showing high ThrottledRequests across multiple tables
  • Batch operations failing across different tables with similar timing
  • Application performance degradation that correlates with overall account activity rather than specific table usage

Common causes

  1. Exceeding AWS account-level throughput limits for DynamoDB operations
  2. Sudden traffic spikes across multiple tables without proper capacity planning
  3. Insufficient provisioned capacity distributed across too many active tables
  4. On-demand tables collectively hitting account-level limits during peak periods
  5. Aggressive retry logic creating retry storms that worsen account-level throttling
  6. Lack of workload isolation between different applications sharing the same AWS account
  7. Exceeding the 500 concurrent control plane operations limit for the account
  8. Global tables replication consuming significant throughput across regions

How to fix "RequestLimitExceeded: Throughput exceeds the current throughput limit for your account"

1Check current account limits and request quota increases

First, identify your current account limits and request increases if needed:

bash
# Check Service Quotas for DynamoDB
aws service-quotas list-service-quotas \
  --service-code dynamodb \
  --region us-east-1

# Check specific quota for on-demand tables
aws service-quotas get-service-quota \
  --service-code dynamodb \
  --quota-code L-4BAB22A6 \
  --region us-east-1

# Request quota increase
aws service-quotas request-service-quota-increase \
  --service-code dynamodb \
  --quota-code L-4BAB22A6 \
  --desired-value 10000 \
  --region us-east-1

Key quotas to check:
- L-4BAB22A6: Maximum on-demand throughput per account
- L-4BAB22A7: Maximum provisioned throughput per account
- L-4BAB22A8: Maximum tables per account
- L-4BAB22A9: Maximum control plane operations per second

Best practices:
- Monitor quota usage in AWS Service Quotas console
- Request increases before reaching 80% of current limits
- Provide business justification for quota increase requests

2Implement account-level rate limiting and backoff

Add account-aware rate limiting to prevent exceeding account limits:

javascript
// Account-level rate limiter
class AccountRateLimiter {
  constructor(maxRequestsPerSecond = 1000) {
    this.maxRequestsPerSecond = maxRequestsPerSecond;
    this.requestCount = 0;
    this.windowStart = Date.now();
    this.windowSize = 1000; // 1 second in milliseconds
  }

  async acquire() {
    const now = Date.now();

    // Reset window if more than 1 second has passed
    if (now - this.windowStart > this.windowSize) {
      this.requestCount = 0;
      this.windowStart = now;
    }

    // Check if we can proceed
    if (this.requestCount >= this.maxRequestsPerSecond) {
      const waitTime = this.windowStart + this.windowSize - now;
      await new Promise(resolve => setTimeout(resolve, waitTime));
      return this.acquire(); // Recursive call after waiting
    }

    this.requestCount++;
    return true;
  }
}

// Usage with DynamoDB client
const accountLimiter = new AccountRateLimiter(800); // 800 requests/second

async function makeDynamoDBRequest(operation, params) {
  await accountLimiter.acquire();
  return await dynamodb[operation](params).promise();
}

// Implement exponential backoff with jitter for account-level errors
const delay = (retryCount) => {
  const baseDelay = 100; // 100ms
  const maxDelay = 10000; // 10 seconds
  const delay = Math.min(maxDelay, baseDelay * Math.pow(2, retryCount));
  const jitter = delay * 0.2 * Math.random(); // Add 20% jitter
  return delay + jitter;
};

Implementation notes:
- Set conservative limits below your actual account quotas
- Use distributed rate limiting for multi-instance applications
- Implement circuit breakers to fail fast during sustained account-level throttling

3Distribute workload across multiple AWS accounts

For high-throughput applications, consider using multiple AWS accounts:

bash
# Create new AWS account using AWS Organizations
aws organizations create-account \
  --email "dynamodb-app-2@example.com" \
  --account-name "DynamoDB-App-2" \
  --iam-user-access-to-billing ALLOW

# Configure cross-account access
aws sts assume-role \
  --role-arn "arn:aws:iam::NEW_ACCOUNT_ID:role/DynamoDBAppRole" \
  --role-session-name "CrossAccountSession"

# Set up AWS Resource Access Manager (RAM) for shared resources
aws ram create-resource-share \
  --name "DynamoDB-Shared" \
  --resource-arns "arn:aws:dynamodb:us-east-1:MAIN_ACCOUNT_ID:table/SharedTable" \
  --principals "NEW_ACCOUNT_ID"

Workload distribution strategies:
1. Service-based isolation: Different services in different accounts
2. Tenant-based isolation: Different customer segments in different accounts
3. Region-based isolation: Different regions in different accounts for global applications
4. Environment-based isolation: Production, staging, development in separate accounts

AWS Organizations best practices:
- Use Service Control Policies (SCPs) to enforce security boundaries
- Implement AWS Control Tower for multi-account governance
- Set up AWS Budgets for cost monitoring per account

4Monitor and analyze account-level metrics

Set up comprehensive monitoring for account-level throughput:

bash
# Create CloudWatch dashboard for account-level metrics
aws cloudwatch put-dashboard \
  --dashboard-name "DynamoDB-Account-Metrics" \
  --dashboard-body '{
    "widgets": [
      {
        "type": "metric",
        "properties": {
          "metrics": [
            ["AWS/DynamoDB", "ThrottledRequests", { "stat": "Sum" }],
            ["AWS/DynamoDB", "ConsumedReadCapacityUnits", { "stat": "Sum" }],
            ["AWS/DynamoDB", "ConsumedWriteCapacityUnits", { "stat": "Sum" }]
          ],
          "period": 300,
          "stat": "Average",
          "region": "us-east-1",
          "title": "Account-Level Throughput"
        }
      }
    ]
  }'

# Set up CloudWatch alarms for account limits
aws cloudwatch put-metric-alarm \
  --alarm-name "DynamoDB-Account-Throttling-High" \
  --alarm-description "Account-level throttling exceeds threshold" \
  --metric-name "ThrottledRequests" \
  --namespace "AWS/DynamoDB" \
  --statistic "Sum" \
  --period 300 \
  --threshold 100 \
  --comparison-operator "GreaterThanThreshold" \
  --evaluation-periods 2 \
  --alarm-actions "arn:aws:sns:us-east-1:ACCOUNT_ID:DevOps-Alerts"

Key metrics to monitor:
- Account-level: ThrottledRequests, ConsumedReadCapacityUnits, ConsumedWriteCapacityUnits
- Table-level: Provisioned/Consumed capacity for each active table
- Operation-level: SuccessfulRequestLatency, UserErrors, SystemErrors
- Cost: Estimated charges from AWS Cost Explorer

Alerting strategy:
- Set alarms at 70% of account limits
- Implement progressive alerting (Warning → Critical → Emergency)
- Integrate with incident management systems (PagerDuty, OpsGenie)

Advanced notes

## Account-Level Throughput Architecture

### AWS Account Limits Structure:
DynamoDB enforces limits at multiple levels:
1. Account-level limits: Apply to all tables in the account
2. Table-level limits: Apply to individual tables
3. Partition-level limits: Apply to individual partitions within tables

### Default Account Limits (varies by region):
- On-demand tables: Typically 40,000 read units and 40,000 write units per account
- Provisioned tables: Varies based on account age and usage history
- Control plane: 500 operations per second per account
- Concurrent operations: Limits on BatchGetItem, BatchWriteItem, Query, and Scan operations

### When to Contact AWS Support:
1. Urgent quota increases for production emergencies
2. Suspected service issues affecting multiple accounts
3. Architecture reviews for high-throughput applications
4. Cost optimization consultations for multi-account setups
5. Performance troubleshooting that crosses account boundaries

### Best Practices Summary:
1. Monitor proactively: Set alarms at 70% of account limits
2. Design for scale: Use multiple accounts for high-throughput applications
3. Implement backpressure: Use rate limiting and queuing
4. Optimize queries: Reduce overall throughput requirements
5. Test at scale: Regularly test account-level limits
6. Document architecture: Maintain clear documentation of multi-account setup
7. Automate management: Use Infrastructure as Code for consistency
8. Review regularly: Quarterly reviews of account limits and usage patterns

Frequently asked questions

What does "RequestLimitExceeded: Throughput exceeds the current throughput limit" mean?
The RequestLimitExceeded error in DynamoDB indicates that your AWS account has exceeded its overall throughput limits for DynamoDB operations. Unlike table-level throttling (ThrottlingException), this is an account-level limit that applies to the combined throughput across all DynamoDB tables in your account. This error typically occurs in several scenarios: 1.
How do I fix "RequestLimitExceeded: Throughput exceeds the current throughput limit"?
DynamoDB returns RequestLimitExceeded when your application exceeds account-level throughput limits, which apply to all tables in your AWS account. This error occurs when the combined throughput across all DynamoDB tables in your account exceeds AWS-imposed limits, requiring quota management and workload distribution strategies.
What causes "RequestLimitExceeded: Throughput exceeds the current throughput limit"?
Exceeding AWS account-level throughput limits for DynamoDB operations Sudden traffic spikes across multiple tables without proper capacity planning Insufficient provisioned capacity distributed across too many active tables
Is the "RequestLimitExceeded: Throughput exceeds the current throughput limit" error serious?
Yes. This is a high-severity error that usually blocks work until it is fixed, but the steps above resolve it.

Official resources & further reading