Files
pezkuwi-subxt/cumulus/client/relay-chain-minimal-node/src/network.rs
T
Aaro Altonen 80616f6d03 Integrate litep2p into Polkadot SDK (#2944)
[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>
2024-04-08 16:44:13 +00:00

180 lines
6.2 KiB
Rust

// Copyright (C) Parity Technologies (UK) Ltd.
// This file is part of Cumulus.
// Cumulus 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.
// Cumulus 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 Cumulus. If not, see <http://www.gnu.org/licenses/>.
use polkadot_core_primitives::{Block, Hash, Header};
use sp_runtime::traits::NumberFor;
use sc_network::{
config::{
NetworkConfiguration, NonReservedPeerMode, NotificationHandshake, PeerStore, ProtocolId,
SetConfig,
},
peer_store::PeerStoreProvider,
service::traits::NetworkService,
NotificationMetrics,
};
use sc_network::{config::FullNetworkConfiguration, NetworkBackend, NotificationService};
use sc_network_common::{role::Roles, sync::message::BlockAnnouncesHandshake};
use sc_service::{error::Error, Configuration, NetworkStarter, SpawnTaskHandle};
use std::{iter, sync::Arc};
/// Build the network service, the network status sinks and an RPC sender.
pub(crate) fn build_collator_network<Network: NetworkBackend<Block, Hash>>(
config: &Configuration,
mut network_config: FullNetworkConfiguration<Block, Hash, Network>,
spawn_handle: SpawnTaskHandle,
genesis_hash: Hash,
best_header: Header,
notification_metrics: NotificationMetrics,
) -> Result<
(Arc<dyn NetworkService>, NetworkStarter, Arc<dyn sp_consensus::SyncOracle + Send + Sync>),
Error,
> {
let protocol_id = config.protocol_id();
let (block_announce_config, _notification_service) = get_block_announce_proto_config::<Network>(
protocol_id.clone(),
&None,
Roles::from(&config.role),
best_header.number,
best_header.hash(),
genesis_hash,
notification_metrics.clone(),
network_config.peer_store_handle(),
);
// Since this node has no syncing, we do not want light-clients to connect to it.
// Here we set any potential light-client slots to 0.
adjust_network_config_light_in_peers(&mut network_config.network_config);
let peer_store = network_config.take_peer_store();
spawn_handle.spawn("peer-store", Some("networking"), peer_store.run());
let network_params = sc_network::config::Params::<Block, Hash, Network> {
role: config.role.clone(),
executor: {
let spawn_handle = Clone::clone(&spawn_handle);
Box::new(move |fut| {
spawn_handle.spawn("libp2p-node", Some("networking"), fut);
})
},
fork_id: None,
network_config,
genesis_hash,
protocol_id,
metrics_registry: config.prometheus_config.as_ref().map(|config| config.registry.clone()),
block_announce_config,
bitswap_config: None,
notification_metrics,
};
let network_worker = Network::new(network_params)?;
let network_service = network_worker.network_service();
let (network_start_tx, network_start_rx) = futures::channel::oneshot::channel();
// The network worker is responsible for gathering all network messages and processing
// them. This is quite a heavy task, and at the time of the writing of this comment it
// frequently happens that this future takes several seconds or in some situations
// even more than a minute until it has processed its entire queue. This is clearly an
// issue, and ideally we would like to fix the network future to take as little time as
// possible, but we also take the extra harm-prevention measure to execute the networking
// future using `spawn_blocking`.
spawn_handle.spawn_blocking("network-worker", Some("networking"), async move {
if network_start_rx.await.is_err() {
tracing::warn!(
"The NetworkStart returned as part of `build_network` has been silently dropped"
);
// This `return` might seem unnecessary, but we don't want to make it look like
// everything is working as normal even though the user is clearly misusing the API.
return
}
network_worker.run().await;
});
let network_starter = NetworkStarter::new(network_start_tx);
Ok((network_service, network_starter, Arc::new(SyncOracle {})))
}
fn adjust_network_config_light_in_peers(config: &mut NetworkConfiguration) {
let light_client_in_peers = (config.default_peers_set.in_peers +
config.default_peers_set.out_peers)
.saturating_sub(config.default_peers_set_num_full);
if light_client_in_peers > 0 {
tracing::debug!(target: crate::LOG_TARGET, "Detected {light_client_in_peers} peer slots for light clients. Since this minimal node does support\
neither syncing nor light-client request/response, we are setting them to 0.");
}
config.default_peers_set.in_peers =
config.default_peers_set.in_peers.saturating_sub(light_client_in_peers);
}
struct SyncOracle;
impl sp_consensus::SyncOracle for SyncOracle {
fn is_major_syncing(&self) -> bool {
false
}
fn is_offline(&self) -> bool {
true
}
}
fn get_block_announce_proto_config<Network: NetworkBackend<Block, Hash>>(
protocol_id: ProtocolId,
fork_id: &Option<String>,
roles: Roles,
best_number: NumberFor<Block>,
best_hash: Hash,
genesis_hash: Hash,
metrics: NotificationMetrics,
peer_store_handle: Arc<dyn PeerStoreProvider>,
) -> (Network::NotificationProtocolConfig, Box<dyn NotificationService>) {
let block_announces_protocol = {
let genesis_hash = genesis_hash.as_ref();
if let Some(ref fork_id) = fork_id {
format!("/{}/{}/block-announces/1", array_bytes::bytes2hex("", genesis_hash), fork_id)
} else {
format!("/{}/block-announces/1", array_bytes::bytes2hex("", genesis_hash))
}
};
Network::notification_config(
block_announces_protocol.into(),
iter::once(format!("/{}/block-announces/1", protocol_id.as_ref()).into()).collect(),
1024 * 1024,
Some(NotificationHandshake::new(BlockAnnouncesHandshake::<Block>::build(
roles,
best_number,
best_hash,
genesis_hash,
))),
// NOTE: `set_config` will be ignored by `protocol.rs` as the block announcement
// protocol is still hardcoded into the peerset.
SetConfig {
in_peers: 0,
out_peers: 0,
reserved_nodes: Vec::new(),
non_reserved_mode: NonReservedPeerMode::Deny,
},
metrics,
peer_store_handle,
)
}