Search code examples
webblockchainipfsdecentralized-applications

Storing data persistently on IPFS

Recently I developed an alternative to Google Drive using IPFS (the decentralized storage technology). The app serverd it's purpose but suffered from 2 major problems:

  1. App was super cool for small files, but on large files, the download was very slow and eventually stopped.
  2. Data was not persistent, means I lost few files after few hours of upload.

My questions:

Is IPFS a persistent storage system? If no what measures can be used to make it persistent?

Solution

  • Understood your question.So coming to the points. Is IPFS a persistent storage system ?

    IPFS is a distributed system that can (among other things) resolve a content hash to the content it represents. This content can never truly be guaranteed to be available (maybe you're offline, maybe all of the peers with it are offline, maybe you're behind a powerful NAT, maybe the network split and the peers with the content are on the other partition). In IPFS the simple or decentralized system an object is online only when the nodes that are holding the object spend energy.

    And your second part,IPFS is mainly for the permanence and permanence!=persistent.IPFS itself currently handles this by means of "pinning", which excludes an object and its children from garbage collection within one IPFS node.

    Work is going on to make it more persistent.One of them is Filecoin (paper), and there a couple of concrete ideas for an ipfs-cluster tool.