PDP Archival Usage

Akave integrates Proof-of-Data-Possession (PDP) for an archival storage tier backed by PDP-serving providers (PDP-SPs).

This section explains:

• How data flows into PDP archival storage
• How to check the archival status of a file
• How to download a file from archival storage using PDP

About Filecoin’s PDP Implementation

Akave’s archival storage tier is powered by Filecoin’s Proof-of-Data-Possession (PDP) protocol. PDP is a cryptographic proof system that allows storage providers to demonstrate they are storing client data without requiring the client to download the entire dataset for verification. Filecoin’s PDP implementation enables:

• Efficient verification: Storage providers can prove data possession with minimal bandwidth
• Decentralized archival: Data is stored across multiple independent storage providers
• Cost-effective long-term storage: Leverages Filecoin’s competitive storage marketplace

For new users looking to learn more about Filecoin’s PDP implementation, visit the Filecoin PDP documentation and explore datasets on the Filecoin PDP Explorer.

1. Uploading Data

Data upload with PDP does not require any special flags; archival processing happens in the background as part of the network’s pipeline.

Standard upload:

akavecli file upload <bucket> <file> \
  --account <wallet-name> \
  --node-address connect.akave.ai:9500

Once uploaded, the file is first available from the primary storage tier. After the network has aggregated the data into a multiple of 2 GB and uploaded it to PDP-SPs, it becomes available in archival PDP storage.

2. PDP Archival Cadence (2 GB Batches)

Archival uploads to PDP-SPs occur in batches/ pieces of approximately 2 GB across the network.

This implies:

• Data is not immediately PDP-available after upload.
• The network aggregates data until it reaches a multiple of 2 GB:
  • If the network has uploaded 2 * n GB, your data becomes PDP-available when the network reaches 2 * (n+1) GB.
• PDP archival availability is therefore eventually consistent.
• Data once it is uploaded to the PDP archival layer is then deleted from Akave Storage Nodes.

If your file is not yet PDP-available, you can still download it using standard storage (without the --archival flag).

3. Checking PDP Archival Status

Use the archival-metadata command:

akavecli archival-metadata <bucket> <file> \
  --account <wallet-name> \
  --node-address connect.akave.ai:9500

Example outputs:

• Fully available:

  Bucket: test1, File: test100mb.txt
  Status: Available for download from archival storage (all blocks have PDP data)

• Not fully available:

  Bucket: test1, File: test-5gb.txt
  Status: Not fully available in archival storage (some blocks are missing PDP data)

Verbose Mode

Add -v or --verbose to include detailed chunk and block information:

akavecli archival-metadata <bucket> <file> \
  --account <wallet-name> \
  --node-address connect.akave.ai:5500 \
  --verbose

Verbose output includes:

• Total chunks
• Chunk CIDs
• Block CIDs
• PDP-SP URLs per block
• Offsets and sizes

This is useful for debugging archival gaps or looking up status of Dataset on Filecoin’s PDP Explorer.

4. Downloading Data via PDP

To explicitly download from PDP archival storage, use the --archival flag:

akavecli file download <bucket> <file> <output-dir> \
  --archival \
  --account <wallet-name> \
  --node-address connect.akave.ai:9500