Miners collectively complain about the high gas fee. How to solve the problem of Filecoin's high gas fee from the root?
On January 14, Beijing time, the Protocol Laboratory submitted the FIP-0010 proposal. The WindowPoSt verification proposal under the FIP-0010 chain, WindowPoSt can be uploaded to the chain without accepting verification. Later, the verifier under the chain can call DisputeWindowedPoSt for verification. If the verifier finds the wrong sector, he can get a certain percentage of rewards. The FIP-0010 proposal aims to reduce gas consumption while reducing pressure on the Filecoin chain.
Gas consumption fluctuation
This will affect the reduction of Gas consumption. I believe you should be able to get the answer after reading this article. The Protocol Lab continues to act on gas fees, and it can also illustrate the important impact of gas fees on the cost of miners from the side. The main cost of miners is the pre-collateral cost, the cost of storing the computing power to burn Gas, and the package transaction fee gifted to the miner. According to Renren Mine’s APP, as of 11 am on January 14th, 8.5952 FIL/T of pledged was sealed (the pledged coins will be returned to the miners after the end of the sector’s life cycle), and the newly added GAS consumes 8.3899 FIL/T (including storage of computing power) The GAS and the tip of the miner). This means that the newly added 1T computing power has reached 16.9851FIL. If a FIL is calculated at 21 US dollars, the 1T cost is 356.685 US dollars, which is equivalent to 2282RMB.
Why is Gas consumption so huge?
Gas consumption sources are mainly divided into two categories:
(1) GAS fee for increasing computing power The cost
of packaging the message submitted to the chain is called the basic burning fee. The storage of computing power (message on the chain) first requires proof of replication. It can be seen from the figure below that PreCommitSector (P2) and ProveCommitSector (C2) are the most costly gas, and P2 and C2 are both part of the copy proof SDR. The specific concept can be understood in detail in the article "Hardware Science | Filecoin Mining Hardware Configuration" by Glacier Lab. This part of the Gas fee will be directly transferred to the F009 account, which is a destruction account set up by the agreement laboratory and can only be entered but not exited. The basic burning fee is fixed.
Gas consumption proportion
②Packing transaction fees for miners are called tips. For this part of transaction fees, miners can set their own limits. Here is the introduction of a concept GasPremium. In order to get information and morning link, miners can increase GasPremium to get news on the chain as soon as possible. Due to the performance limitations of the Filecoin mainnet, the Filecoin mainnet can only generate 2880 blocks a day. According to the Filecoin browser calculation, one block can hold approximately 300-400 pieces of information. In order for miners to complete the packaging as soon as possible, they can choose to increase GasPremium. When the package transaction fee is higher, the message is more likely to be packaged. This has caused the gas in the entire network to rise and fall into a "death cycle". The performance of the Filecoin mainnet is not enough to support the daily increase in computing power, resulting in information congestion, which is the root cause of the skyrocketing gas fee.
③Excess burning fee.
The agreement laboratory will make appropriate penalties for the excess, which is the excess burning fee.
(2) GAS cost to maintain computing power
The SubmitWindowedPoSt in the space-time proof phase will be submitted every half an hour, and the average gas consumption is the largest. However, it is submitted every half an hour and the consumption ratio is not large.
Protocol Lab’s Proposals for Gas Fees
In addition to FIP-0010, Protocol Labs also put forward more proposals to solve Filecoin’s gas fees.
FIP—0005 is a technical proposal submitted by the Protocol Lab on October 22, 2020. The proposal removes expensive miner reward attribution calculations from PreCommitSector and Confirm Sector Proofs Valid, releases chain bandwidth, reduces gas fee consumption, and reduces Miner costs.
FIP-0007 is a proposal submitted on November 27, 2020. It aims to improve Filecoin HAMT and AMT in terms of performance and security through three smaller independent proposals. The proposal will also reduce the size of IPLD operations and serialization, and reduce gas consumption to a certain extent.
FIP-0008 is a technical proposal submitted by Alex et al. on November 4, 2020. In Filecoin's gas fee structure, ProveCommitSector accounts for the largest proportion of nearly 63%. The previous miner PreCommit Sector method only supports submitting a single sector at a time. Such operations are performed frequently and require a lot of gas costs. This proposal adds a scheme for miners to submit sector pre-commitment in batches, which can greatly reduce Gas costs.
On December 16, the Filecoin mainnet released a new version of Lotus 1.3.0. This upgrade includes the protocol laboratory’s latest proposal to improve the gas fee, FIP-0009. The FIP-0009 proposal will return the gas fee burned by the successful implementation of the WindowPoSt Central Plan to the miners, thereby reducing the gas fee consumption of the network.
In addition to the proposal, Filecoin reduced the consumption of Gas fees by introducing the Ethereum improvement proposal EIP-1559. For details, please refer to "Filecoin Official Information|Filecoin's EIP-1559".
Filecoin's high gas fee reasons and solutions
As mentioned earlier, the performance of the Filecoin mainnet is not enough to support the daily increase in computing power, resulting in information jams. This is the root cause of the skyrocketing gas fee. Most blockchain projects have faced such TPS performance bottlenecks, including mainstream projects such as BTC and ETH. Here is a common way of expansion, ETH2.0 is to expand the capacity by using the shard chain.
Take Ethereum as an example: The sharding technology divides the nodes in the network into different shards. Each shard handles different transaction data. Each shard handles the same transaction event together. Transactions can be processed in parallel in the network to greatly increase the amount of network concurrency. In order to realize the random distribution of nodes and the distribution of area validators, a separate blockchain (called the beacon chain in Ethereum 2.0) needs to be established to perform the operations required to maintain the entire network. This method is quadratic sharding, which increases transaction volume through a two-tier network design.
Specifically, the nodes on the Ethereum network are divided into 100 pieces. The sharding system is maintained by the validator management contract (VMC) issued on the main chain. Each area exists independently. When there are transactions in Ethereum , Select only one of the shards for processing, if there are M transactions to be processed in the network, now each node only needs to process M/100, and then the data of these packaged sub-blocks form a block on the main chain , Which is equivalent to 100 times the capacity of the main chain block.
In addition to sharding, common expansion methods include state channel, segregated witness, lightning network, side chain, etc. It is worth mentioning that Juan Bent mentioned in the meet up that Filecoin may be expanded in the form of shards.
Concluding remarks
As a star project of distributed storage and a representative of the application of blockchain technology, Filecoin can be said to be the most legendary blockchain project. After the Filecoin mainnet went live in October 2020, Filecoin's computing power has exceeded 1.875 EiB, and the development speed of data storage is beyond the reach of the traditional data storage field. The autonomy and reliable storage of data brought by it gives Internet privacy and security. It is private data. Great choice for storage and cold data storage.
The innovation of technological innovation requires participants to look at it from a higher dimension, a longer-term perspective, and a more inclusive perspective. I believe that with the improvement of blockchain expansion technologies such as sharding, more companies will adopt IPFS&Filecoin distributed storage data, and the price of Filecoin will continue to rise.