mirror of
https://github.com/pezkuwichain/pezkuwi-subxt.git
synced 2026-05-07 01:28:07 +00:00
80616f6d03
[litep2p](https://github.com/altonen/litep2p) is a libp2p-compatible P2P networking library. It supports all of the features of `rust-libp2p` that are currently being utilized by Polkadot SDK. Compared to `rust-libp2p`, `litep2p` has a quite different architecture which is why the new `litep2p` network backend is only able to use a little of the existing code in `sc-network`. The design has been mainly influenced by how we'd wish to structure our networking-related code in Polkadot SDK: independent higher-levels protocols directly communicating with the network over links that support bidirectional backpressure. A good example would be `NotificationHandle`/`RequestResponseHandle` abstractions which allow, e.g., `SyncingEngine` to directly communicate with peers to announce/request blocks. I've tried running `polkadot --network-backend litep2p` with a few different peer configurations and there is a noticeable reduction in networking CPU usage. For high load (`--out-peers 200`), networking CPU usage goes down from ~110% to ~30% (80 pp) and for normal load (`--out-peers 40`), the usage goes down from ~55% to ~18% (37 pp). These should not be taken as final numbers because: a) there are still some low-hanging optimization fruits, such as enabling [receive window auto-tuning](https://github.com/libp2p/rust-yamux/pull/176), integrating `Peerset` more closely with `litep2p` or improving memory usage of the WebSocket transport b) fixing bugs/instabilities that incorrectly cause `litep2p` to do less work will increase the networking CPU usage c) verification in a more diverse set of tests/conditions is needed Nevertheless, these numbers should give an early estimate for CPU usage of the new networking backend. This PR consists of three separate changes: * introduce a generic `PeerId` (wrapper around `Multihash`) so that we don't have use `NetworkService::PeerId` in every part of the code that uses a `PeerId` * introduce `NetworkBackend` trait, implement it for the libp2p network stack and make Polkadot SDK generic over `NetworkBackend` * implement `NetworkBackend` for litep2p The new library should be considered experimental which is why `rust-libp2p` will remain as the default option for the time being. This PR currently depends on the master branch of `litep2p` but I'll cut a new release for the library once all review comments have been addresses. --------- Signed-off-by: Alexandru Vasile <alexandru.vasile@parity.io> Co-authored-by: Dmitry Markin <dmitry@markin.tech> Co-authored-by: Alexandru Vasile <60601340+lexnv@users.noreply.github.com> Co-authored-by: Alexandru Vasile <alexandru.vasile@parity.io>
104 lines
4.8 KiB
Rust
104 lines
4.8 KiB
Rust
// This file is part of Substrate.
|
|
|
|
// Copyright (C) Parity Technologies (UK) Ltd.
|
|
// SPDX-License-Identifier: GPL-3.0-or-later WITH Classpath-exception-2.0
|
|
|
|
// This program is free software: you can redistribute it and/or modify
|
|
// it under the terms of the GNU General Public License as published by
|
|
// the Free Software Foundation, either version 3 of the License, or
|
|
// (at your option) any later version.
|
|
|
|
// This program is distributed in the hope that it will be useful,
|
|
// but WITHOUT ANY WARRANTY; without even the implied warranty of
|
|
// MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
|
|
// GNU General Public License for more details.
|
|
|
|
// You should have received a copy of the GNU General Public License
|
|
// along with this program. If not, see <https://www.gnu.org/licenses/>.
|
|
|
|
//! Polite gossiping.
|
|
//!
|
|
//! This crate provides gossiping capabilities on top of a network.
|
|
//!
|
|
//! Gossip messages are separated by two categories: "topics" and consensus engine ID.
|
|
//! The consensus engine ID is sent over the wire with the message, while the topic is not,
|
|
//! with the expectation that the topic can be derived implicitly from the content of the
|
|
//! message, assuming it is valid.
|
|
//!
|
|
//! Topics are a single 32-byte tag associated with a message, used to group those messages
|
|
//! in an opaque way. Consensus code can invoke [`ValidatorContext::broadcast_topic`] to
|
|
//! attempt to send all messages under a single topic to all peers who don't have them yet, and
|
|
//! [`ValidatorContext::send_topic`] to send all messages under a single topic to a specific peer.
|
|
//!
|
|
//! # Usage
|
|
//!
|
|
//! - Implement the [`Network`] trait, representing the low-level networking primitives. It is
|
|
//! already implemented on `sc_network::NetworkService`.
|
|
//! - Implement the [`Validator`] trait. See the section below.
|
|
//! - Decide on a protocol name. Each gossiping protocol should have a different one.
|
|
//! - Build a [`GossipEngine`] using these three elements.
|
|
//! - Use the methods of the [`GossipEngine`] in order to send out messages and receive incoming
|
|
//! messages.
|
|
//!
|
|
//! The [`GossipEngine`] will automatically use [`Network::add_set_reserved`] and
|
|
//! [`NetworkPeers::remove_peers_from_reserved_set`] to maintain a set of peers equal to the set of
|
|
//! peers the node is syncing from. See the documentation of `sc-network` for more explanations
|
|
//! about the concepts of peer sets.
|
|
//!
|
|
//! # What is a validator?
|
|
//!
|
|
//! The primary role of a [`Validator`] is to process incoming messages from peers, and decide
|
|
//! whether to discard them or process them. It also decides whether to re-broadcast the message.
|
|
//!
|
|
//! The secondary role of the [`Validator`] is to check if a message is allowed to be sent to a
|
|
//! given peer. All messages, before being sent, will be checked against this filter.
|
|
//! This enables the validator to use information it's aware of about connected peers to decide
|
|
//! whether to send messages to them at any given moment in time - In particular, to wait until
|
|
//! peers can accept and process the message before sending it.
|
|
//!
|
|
//! Lastly, the fact that gossip validators can decide not to rebroadcast messages
|
|
//! opens the door for neighbor status packets to be baked into the gossip protocol.
|
|
//! These status packets will typically contain light pieces of information
|
|
//! used to inform peers of a current view of protocol state.
|
|
|
|
pub use self::{
|
|
bridge::GossipEngine,
|
|
state_machine::TopicNotification,
|
|
validator::{DiscardAll, MessageIntent, ValidationResult, Validator, ValidatorContext},
|
|
};
|
|
|
|
use sc_network::{multiaddr, types::ProtocolName, NetworkBlock, NetworkEventStream, NetworkPeers};
|
|
use sc_network_sync::SyncEventStream;
|
|
use sc_network_types::PeerId;
|
|
use sp_runtime::traits::{Block as BlockT, NumberFor};
|
|
use std::iter;
|
|
|
|
mod bridge;
|
|
mod state_machine;
|
|
mod validator;
|
|
|
|
/// Abstraction over a network.
|
|
pub trait Network<B: BlockT>: NetworkPeers + NetworkEventStream {
|
|
fn add_set_reserved(&self, who: PeerId, protocol: ProtocolName) {
|
|
let addr =
|
|
iter::once(multiaddr::Protocol::P2p(who.into())).collect::<multiaddr::Multiaddr>();
|
|
let result = self.add_peers_to_reserved_set(protocol, iter::once(addr).collect());
|
|
if let Err(err) = result {
|
|
log::error!(target: "gossip", "add_set_reserved failed: {}", err);
|
|
}
|
|
}
|
|
fn remove_set_reserved(&self, who: PeerId, protocol: ProtocolName) {
|
|
let result = self.remove_peers_from_reserved_set(protocol, iter::once(who).collect());
|
|
if let Err(err) = result {
|
|
log::error!(target: "gossip", "remove_set_reserved failed: {}", err);
|
|
}
|
|
}
|
|
}
|
|
|
|
impl<T, B: BlockT> Network<B> for T where T: NetworkPeers + NetworkEventStream {}
|
|
|
|
/// Abstraction over the syncing subsystem.
|
|
pub trait Syncing<B: BlockT>: SyncEventStream + NetworkBlock<B::Hash, NumberFor<B>> {}
|
|
|
|
impl<T, B: BlockT> Syncing<B> for T where T: SyncEventStream + NetworkBlock<B::Hash, NumberFor<B>> {}
|