Polkadot v1.0 : Sharding thiab Economic Security (1)

This article is taken from an article on Polkadot sharing and financial security published by Polkadot co-founder Robert. This sentence is afirst division! Please check!

This article is about technology after Polkadot. Polkadot is a hard blockchain with heterogeneous shards. In this context, sharding means the distribution of work across multiple sub-blockchains called parachains. Heterogeneity means that each blockchain has its own potential for exchanging state for a specific application. Many industries have different products, so the channel in one situation can help its users as much as possible. Polkadot provides security and messaging to all Parachains.

This sentence is usually written for readers who have some understanding of blockchain consensus. We hope it will be useful for non-developers and provide insight into the issues and problems facing blockchain and Polkadot processes. I won't go into detail about every nuance of the solution, but I will go into detail about the type of thinking and how to ensure your goals are met. .

Polkadot Front Chains

A simple understanding of the blockchain only includes a chain that extends genetically to the end. In many cases, it is only the main idea of ​​the chain that occurs in the network and represents each consensus as a change of state. The ultimate goal of any blockchain consensus is to provide researchers with the most accurate information. However, this chain, which we can call the end or the chain, is only the end of life of the many chains that will exist. The role of the blockchain approval algorithm is to start with as many chains as possible and ultimately only one of them decides.

The Creator who blocks can create the boxes below, but may have several victories at a time. When the work is evidence, new food creates a new technique that has a difficult species of the blocks of blocks for a block. In environmental competition, multiple attempts are required to find the block that meets the condition. Remember when writing these sentences, the Bitcoin rate is 162 exasashes one twice to help the next step, which allows the next block of blocks Cost is lower than the target value every 10 minutes. Many miners can see a lot of solving, which can show small legs in blocking the chain. The future miners should choose the block that the legs can grow. Rules as the longest way. In the case of the interval that started by the past block, he was stressful and difficult to overwhelm more than the chain. Thus, the main lighter of the longest chain can be considered as the ultimate interaction of the result.

The Ouroboros proof-of-concept family is simulated using a cryptographic technique called a Verifiable Random Function (VRF) by dividing the time into individual openings and giving each user access to the registration process and the set of validators. The duration of the proof of function to establish the true random value of a hole is less than the initial value with the certificate which allows the user to create new blocks. Similar to proof of work, many users can create VRF values ​​under the startup and building blocks at the same time, making it a fork. Experts are deliberately encouraged to teach forks and are fined for cutting their stakes on strings. This process also provides the correct value according to the longest string selection rule.

Unlike Proof of Action miners, Ouroboros-style network validators only need to run a calculation to have a chance of creating a blockchain, unlike miners who need to do more hashing. This allows users to spend most of their time building blocks in a business and providing better accounting in the blockchain.

The Polkadot Relay Chain uses a process called BABE, a modification of the Ouroboros Praos. A special improvement of BABE over Praos is that BABE relies on a mid-range NTP server to alert users at the current time.

Probabilistic determinism is fine, but waiting for a block to reach the deepest chain length is pointless because the network is designed to adapt to the worst case of exposure to certain levels of stress and aggression. Most networks have probably already agreed on which blocks are part of the on-chain process. In fact, almost any network can reach from pre-block level down to the minimum depth of the longest chain. To this end, we introduce the concept of the ultimate gadget and the ultimate. The Finality Gadget is a two-step approval protocol that runs to the peak of the performance blockchain, proving a faster process for blocks that the network ultimately decides. This approval process describes additional financial security tools. A decision room cannot carry out two concurrent tests without reducing all user equipment by at least 1/3.

Polkadot's relay chain uses a decision support tool called GRANDPA. It works roughly by having the end of the substrings in length and asking voters to reselect the block they think is the head of the longest string. GRANDPA is currently running on the Polkadot and Kusama networks and can complete the last of the new blocks in less than 3 seconds on Kusama with 900 users at the time of writing.

The combination of BABE and GRANDPA allows Polkadot to efficiently grow the chain using only the inputs from a single validator. This work was signed off by most validators and completed soon after. The combination of these devices means that in the network connection Polkadot achieves high latency and low latency, while under unfavorable conditions the wired connection can be achieved with too high and (too) high latency.

Sharding: interesting with the Subset option

Back to sharding. The purpose of sharding is to achieve segregation by segregation operating as a trade through multiple chains called shards. The shards have been mentioned and protected by the parent blockchain. In Polkadot, the parent blockchain is called a chain link and the shards are called parachains. Since most of the information appearing in the relay chain is a transaction that references a new Parachain block, it is cheaper to fork the relay chain itself. Here I know the difference between the fork between the relay chain and the final chain. Most of our work here is to ensure that the end of the chain relay (which the user thinks of as standard chains) only has data access to Parachain blocks.

This image or a similar image has been shared on the internet. It shows how to divide users into groups, assign them to Parachain, and get Parachain block comments from authors. In this article, I give many explanations of subtle concepts.

If each validator only needs to identify some but not all parachain cable blocks, sharding only improves scalability. If you have 10 parachains and each client must identify each block in all 10 chains, you can place all transactions on a blockchain and call it a day. The trick is to find a way that allows each user to do as much insurance as possible while managing financial security. Experts advise that bad parachain blocks are not financially viable. In particular, it should not be possible for the attackers' validation process to stall and include the Parachain malfunction in the final relay chain until all staking charges are satisfied. In fact, the validators, a small group of validators, may have conspired to obtain invalid parachain blocks mentioned by the unfinished fork of the relay chain, but they were ignored until the fork was completed and the offender be condemned.

A parachain block is complete when the relay chain block referred to it is complete.

A Parachain block is completed when the Relay Chain block is about to be completed.

Let us make some specific reflections on the enemy that we are defending.

The attacker can control up to a third of all validators and can control all validators to work as needed.

Protesters can see all messages on the network from honest employees and users they control.

The attacker can deny up to X% of the user's access at any time, blocking messages sent or received.

Time must be allowed before the protester rejects the consumer product.

Although we won't be creating security proofs for the process in this post, these restrictions should give us some insight into the type of attack we're trying to prevent.

In fact, the main unit of our parachain concept is not a parachain, but what we call a core or short core. This is similar to the processor core. It works in parallel and organizes work into individual slots. Each Parachain has its own special core. That is to say, it is always reserved for something important. However, it is also possible to make multiple strings of a key. The only difference is the scheduling algorithm.

A core service is a good description of an extension chain. The core is directly related to the amount of work the analyst has done. Each core can be completed up to a parachain blocking a relay chain block at the top.

In a shared blockchain system, each Parachain block is identified by a small number of users, so the data available is an important factor in ensuring that the data needed to identify the Parachain block can become misleading again. Core availability is controlled by a chain of relays and acts like parachain blocks waiting for data to become available. The primary purpose of the available devices is to track the initial settling time and provide resistance when the data is slower than usual.

The kernel logic available is as follows: A key is empty when it is ready to receive new Parachain blocks, and was empty at that time. Then the data will be available or the process will have time to complete. At this point, the kernel is empty again.

It is useful to divide the Parachain Consensus into 5 different protocols entangled in the Relay Chain Consensus.

link

Support

performance

Accept

conflict

unityThis is the process of creating a parachain block. The constructor creates the parachain blocks and sends them to the consumer.

SupportThe parachain block is first identified by a small group of relay chain validators and registered in the relay chain. One of the main benefits of support is that users can be affected if subsequent processes fail.

performanceThe process by which users can share the information needed to identify a parachain block and make it available for further analysis.

AcceptThe process by which random validators fetch data and kill parachain blocks. They agree or initiate a dispute depending on whether the parachain block is considered valid.

conflictThis is the process of resolving consumer disputes between Parachain blocks, ignoring Parachain blocks incorrectly and punishing violators. Conflicts are only preventable and not expected to happen often.

The user can participate in all these processes simultaneously and often has several instantiations. For example, end users can participate in a vote to approve a parachain blocking pipeline while supporting new blocks and replacing old ones.

This internal comparison also affects the use of the architecture. Each of these rules is used as an independent subsystem and all subsystems operate in parallel. Everyone is always doing something.