> For the complete documentation index, see [llms.txt](https://labelfi-1.gitbook.io/labelfi/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://labelfi-1.gitbook.io/labelfi/basics/technical-architecture/5.1-distributed-data-storage.md).

# 5.1 Distributed Data Storage

Labelfi utilizes blockchain-based decentralized storage systems, such as IPFS (InterPlanetary File System) and Filecoin, to handle the large volumes of data needed for annotation projects. These systems offer several advantages:

* Distributed Storage: Rather than storing data on a centralized server, IPFS enables data to be distributed across multiple nodes in a peer-to-peer network. This enhances data resilience, as files are replicated across nodes, reducing the risk of data loss or downtime \[12].
* Scalability: The decentralized nature of the storage allows Labelfi to easily scale as the amount of annotated data grows. Filecoin’s incentivized model ensures that storage providers are compensated for offering space, supporting unlimited growth \[13].
* Cost Efficiency: By leveraging decentralized storage, Labelfi avoids the high costs associated with traditional cloud services, offering a more cost-efficient solution for large-scale data labeling projects \[14].