Network performance

The Filecoin Network is a decentralized storage market and network that provides data persistence via a decentralized protocol and publicly verifiable storage proofs on a blockchain. Current Filecoin network performance is primarily determined by security parameters and Filecoin's proof constructions (opens new window).

It is difficult to provide highly reliable benchmarks for Filecoin network performance today due to the network being extremely early stage and therefore, somewhat volatile, and because Filecoin is a distributed system. However, as you begin interacting with Filecoin, you can use these heuristics to understand general Filecoin network performance and how it fits with your use case.

Financial transfers

A message that requires transferring FIL is often extremely fast, and will take on average ~1 blocktime (or around 30 seconds) to be reflected on chain. We consider 120 blocks (1 hour) a conservative number of confirmations for high-value transfers.

Data storage

The Filecoin data storage protocol has a few key components once a deal is proposed and accepted:

1. Funding the storage market actor: This process takes ~1-2 minutes and ensures that both the client and the miner have funds and collateral to pay for the deal.

2. Data transfer: This portion of the deal flow involves the client's node sending the relevant pieces of data to the mining node. The data transfer rate varies widely, depending on the network and disk bandwidths of both the client and the miner. Usually, the network speed between client and miner will be the key determining factor.

3. Deal shows up on-chain: once the data is received by the miner, it will be verified to make sure it matches the deal parameters and the deal will be published on the chain.

4. Sector sealing: Once the deal shows up on-chain, the miner must still complete generating a Proof-of-Replication and sealing the sector (opens new window). This process is currently estimated to take ~1.5hours for a 32GB sector on a machine that meets these minimum hardware requirements for mining.

For most storage clients, the storage performance that matters is the time it takes from deal acceptance to deal appearance on-chain. This is the sum of steps (1) to (3) above. Based on current high-level benchmarks, these steps are estimated to take around ~5-10 minutes for a 1MiB file.

Data retrieval

There are two methods by which one can directly retrieve data from the Filecoin network:

• Fast retrieval: By default, some Filecoin clients, like lotus, enable storage miners to store an unsealed copy of the stored data in addition to a sealed copy. The sealed copy is necessary for the ongoing storage proofs that a miner must submit, while the unsealed copy can be used for quicker retrievals of the data from the storage miner. While this is a useful feature, there is no guarantee that all miners are storing extra unsealed copies of the stored data, as this is not a verifiable part of the protocol. In lotus, this feature is called fast-retrieval.

• Retrieval after unsealing: Because of the Filecoin protocol’s design, storage miners are essentially cryptographically guaranteed to store client data in its sealed format. Therefore, if the storage miner doesn’t have an unsealed copy of the data stored, they will have to unseal the sealed data first (i.e. decoding the encoded data) and then serve it back up to the requester, i.e. the retrieval client.

In both methods, the data retrieval process after a retrieval deal is accepted includes:

1. Funding the payment channel for retrieval: Similar to the storage deal payment channel funding above, except for the purpose of data retrieval. The timing estimates for payment channel creation and funding are roughly the same as noted above.

2. Unsealing (if needed): The miner unseals (decodes) the data so it can be read by the data requester. Sealing and unsealing are symmetric processes, which means they take roughly the same amount of time in either direction. Thus, the unsealing step is estimated to take around as long as the sealing step listed above, or around ~3hours for a 32GiB sector on a machine running minimum hardware requirements.

3. Data transfer: The miner begins transferring the data back to the data requester. This is also transferred back at a rate similar to the original data transfer rate, which is highly variable depending on a number of different factors.

Because of the various steps involved in the data retrieval process, Filecoin storage currently meets similar performance bars as traditional warm or cold storage. In order to get performance that is similar to other hot storage solutions, most users utilize Filecoin in conjunction with a caching layer such as IPFS. These hybrid, multi-tiered storage solutions use IPFS for hot storage and Filecoin for affordable, frequent, and versioned backups. Some example hybrid storage products include Powergate and Textile Buckets.