Public data
What is public network data?
One of the main value propositions of a public blockchain network is that it be, well, public. This means that the information about what’s going on the network needs to be readily available. Networks like Ethereum have this more or less built in and available as part of the software: each node has an API that will return any information you ask for; if you have a lot of requests, well, you just need more nodes.
To be clear, that’s how Linea works too—it’s just that Ethereum's incentive mechanisms mean that anyone can stake ETH and run a validator node. Linea doesn't rely on staking to ensure correctness. As an outcome, its nodes are run by teams at Consensys. Therefore, in order to provide for the massive interest that the network has generated, a lot of nodes have to be run. This allows dapps to ask for information about the state of the network without impeding the actual execution of transactions by overloading the client.
What do the public RPCs do?
Receive and respond to requests from users and dapps according to the Ethereum JSON-RPC API standard. This includes providing information to MetaMask users about their accounts on Linea.
How do they do it?
By leveraging the expertise and resources of Infura, Consensys, and the Ethereum ecosystem as a whole, Infura is running a number of nodes to provide this service, in two main capacities:
Client-facing RPC-API nodes
- RPC nodes do the “traditional” work of EVM nodes:
- They receive updated network state information from the sequencer and state manager, and providing information about that state to users and dapps when they request it.
- They receive incoming transactions.
To handle the traffic, Infura runs the nodes behind a load balancer. Those transactions submitted by users are, therefore, balanced across the nodes, and are thereby routed into the memory pool, for subsequent ingestion and processing by the coordinator and sequencer.
Archive nodes
Archive nodes support those dapps and those research activities that rely on historical data.
This type of data request can be resource-intensive, and could put the live network nodes at risk of having their performance impacted, and thus threaten the overall health of the network. For this reason, Infura deploys archive nodes in a scalable architecture; these kinds of transactions aren’t being run all the time–-but when they are, more nodes are spun up as needed to keep up with demand.