Use Streaming for Binary Data Transfers

When moving large binary files between services, avoid loading the whole file into n8n's memory. Set N8N_DEFAULT_BINARY_DATA_MODE to filesystem and use the HTTP Request node with Response Format set to File: the download is written straight to disk, and the next node reads it from disk to upload. The file never fully loads into Node.js memory.

How do you transfer large binary files in n8n?

When moving binary data between services (e.g., downloading from S3 and uploading to Google Drive), avoid loading the entire file into n8n's memory. Use the HTTP Request node's binary streaming capability to pipe data directly.

Real-world example: A workflow downloads daily backup archives (500 MB+) from a remote server and uploads them to cloud storage.

Workflow structure:

[Schedule Trigger] -> [HTTP Request (Download)] -> [S3 Upload]

How do you configure the HTTP Request node for streaming?

HTTP Request node configuration for download:

Key settings that enable efficient binary handling:

# Ensure these are set in your environment

N8N_DEFAULT_BINARY_DATA_MODE=filesystem
N8N_BINARY_DATA_STORAGE_PATH=/data/n8n-binary

What should you watch out for with disk space?

When filesystem mode is active, the HTTP Request node writes the downloaded file directly to disk. The S3 node then reads from disk to upload. The file never fully loads into Node.js memory.

Warning: Temp Disk Space Ensure the binary data storage path has sufficient disk space. A 500 MB file download creates a 500 MB temp file. Clean up old binary data by configuring execution pruning (see Tip 4).

Related: Flatten Deeply Nested API Responses · Use Docker Compose with Health Checks for n8n and PostgreSQL