> For the complete documentation index, see [llms.txt](https://docs.lumia.org/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.lumia.org/ecosystem/lumia/architecture/data-availability/why-the-data-availability-choice-matters.md).

# Why the data availability choice matters

**Who this is for:** developers and issuers weighing whether Lumia's data-availability choice is sound, not just how it works mechanically. **You should already know:** [Where Lumia's transaction data lives](https://app.gitbook.com/s/chmhPiVNAafGw48c4oU2/ecosystem/lumia/architecture/data-availability/where-lumia-transaction-data-lives) and [How data-availability sampling scales](https://app.gitbook.com/s/chmhPiVNAafGw48c4oU2/ecosystem/lumia/architecture/data-availability/how-data-availability-sampling-scales).

A dedicated data-availability network is, in general, a reasonable choice compared to the two realistic alternatives for a validium: posting data directly on the base chain, or trusting a small fixed committee to attest availability.

### Cost, compared to posting data on the base chain

The whole reason a validium exists instead of a pure rollup is that posting full transaction data to a base chain like Ethereum is expensive at scale (see [Validium](https://app.gitbook.com/s/chmhPiVNAafGw48c4oU2/ecosystem/lumia/architecture/data-availability/validium)). A dedicated DA network is built specifically to make that data cheap to publish and verify, since it isn't also carrying the cost of general-purpose execution the way a base chain is. That's the core economic case for using one at all rather than falling back to full on-chain posting.

### Security, compared to a fixed signer committee

The other realistic alternative for a validium is a data-availability committee (DAC): a small, fixed set of parties who sign attestations that data is available. That model's security is only as strong as that committee: if enough members collude or go offline, the guarantee fails, and the committee's size is typically small enough that this isn't a purely theoretical concern. A sampling-based network like Avail would spread that guarantee across many independent light clients performing random sampling instead of a small fixed set of signers. That's a meaningfully different, and broader, trust base than a handful of committee members. See [Where Lumia's transaction data lives](https://app.gitbook.com/s/chmhPiVNAafGw48c4oU2/ecosystem/lumia/architecture/data-availability/where-lumia-transaction-data-lives) for which mechanism Lumia runs.