swactor/crates/simulation/src/distribution/sim.rs

712 lines
27 KiB
Rust
Raw Normal View History

use std::collections::HashSet;
use distribution::node::{DistributedNode, DistributedNodeConfig, ResolveResult};
use distribution::swim::node::NodeAction;
use distribution::swim::probe::SwimConfig;
use distribution::types::NodeId;
use swactor::actor::ActorAddress;
use crate::trace::{Event, SimulationTrace};
use super::trace::{DistributionEventKind, DistributionSnapshot};
/// A network partition between two sets of nodes.
/// Nodes in `side_a` cannot communicate with nodes in `side_b`.
#[derive(Debug, Clone)]
pub struct Partition {
pub side_a: Vec<usize>,
pub side_b: Vec<usize>,
/// If true, A→B is blocked but B→A works (asymmetric).
pub asymmetric: bool,
}
/// An action to execute at a specific round during the simulation.
#[derive(Debug, Clone)]
pub enum SimAction {
/// Register a name on the given node, binding it to a fresh random actor.
RegisterName { node_idx: usize, name: String },
/// Register a name on the given node, binding it to a specific actor address.
RegisterNameWithActor { node_idx: usize, name: String, actor: ActorAddress },
/// Unregister a name on the given node (creates a tombstone).
UnregisterName { node_idx: usize, name: String },
/// Graceful leave — node announces its own death before being removed.
GracefulLeave { node_idx: usize },
}
/// Schedule entry for network faults.
#[derive(Debug, Clone)]
pub enum NetworkFault {
/// Introduce a partition at the given round.
Partition { round: usize, partition: Partition },
/// Heal a partition at the given round (restores full connectivity).
Heal { round: usize },
/// Set message drop rate (0.0 = no drops, 1.0 = drop all).
SetDropRate { round: usize, rate: f64 },
}
/// Configuration for a distribution simulation run.
#[derive(Debug, Clone)]
pub struct DistributionSimConfig {
pub name: String,
pub num_nodes: usize,
pub num_rounds: usize,
pub ticks_per_round: usize,
pub swim: SwimConfig,
/// Number of actors to register per node.
pub actors_per_node: usize,
/// (round, node_idx) — kill the node at the specified round.
pub kill_schedule: Vec<(usize, usize)>,
/// (round, node_idx) — revive the node at the specified round.
pub revive_schedule: Vec<(usize, usize)>,
pub cache_capacity: usize,
/// Network fault schedule.
pub network_faults: Vec<NetworkFault>,
/// Actions to execute at specific rounds (e.g. register/unregister names).
pub action_schedule: Vec<(usize, SimAction)>,
/// Custom registry config overrides.
pub registry_tombstone_ttl: Option<u64>,
pub registry_gc_interval: Option<u64>,
pub registry_dissemination_lambda: Option<usize>,
}
impl Default for DistributionSimConfig {
fn default() -> Self {
Self {
name: "distribution-sim".into(),
num_nodes: 5,
num_rounds: 50,
ticks_per_round: 3,
swim: SwimConfig {
probe_interval: 1,
probe_timeout: 3,
indirect_probes: 1,
suspicion_timeout: 5,
dead_reprobe_interval: 10,
},
actors_per_node: 2,
kill_schedule: Vec::new(),
revive_schedule: Vec::new(),
cache_capacity: 100,
network_faults: Vec::new(),
action_schedule: Vec::new(),
registry_tombstone_ttl: None,
registry_gc_interval: None,
registry_dissemination_lambda: None,
}
}
}
/// Tracks active network state during simulation.
struct NetworkState {
/// Set of (from_idx, to_idx) pairs where messages are blocked.
blocked: HashSet<(usize, usize)>,
/// Probability of dropping a message [0.0, 1.0].
drop_rate: f64,
/// Simple counter-based deterministic "random" for drop decisions.
drop_counter: u64,
}
impl NetworkState {
fn new() -> Self {
Self {
blocked: HashSet::new(),
drop_rate: 0.0,
drop_counter: 0x853c49e6748fea9b, // Non-zero seed for better distribution
}
}
fn apply_fault(&mut self, fault: &NetworkFault, num_nodes: usize) {
match fault {
NetworkFault::Partition { partition, .. } => {
for &a in &partition.side_a {
for &b in &partition.side_b {
if a < num_nodes && b < num_nodes {
self.blocked.insert((a, b));
if !partition.asymmetric {
self.blocked.insert((b, a));
}
}
}
}
}
NetworkFault::Heal { .. } => {
self.blocked.clear();
}
NetworkFault::SetDropRate { rate, .. } => {
self.drop_rate = rate.clamp(0.0, 1.0);
}
}
}
/// Returns true if this message should be delivered.
fn should_deliver(&mut self, from_idx: usize, to_idx: usize) -> bool {
if self.blocked.contains(&(from_idx, to_idx)) {
return false;
}
if self.drop_rate > 0.0 {
self.drop_counter = self.drop_counter.wrapping_mul(6364136223846793005).wrapping_add(1);
let r = (self.drop_counter >> 33) as f64 / (u32::MAX as f64);
if r < self.drop_rate {
return false;
}
}
true
}
}
pub type DistTrace = SimulationTrace<DistributionEventKind, DistributionSnapshot>;
/// Run a distribution simulation, returning both the trace and the final node states.
///
/// The returned `Vec<Option<DistributedNode>>` has the same length as `config.num_nodes`.
/// Dead nodes are `None`.
pub fn run_simulation_with_nodes(config: DistributionSimConfig) -> (DistTrace, Vec<Option<DistributedNode>>) {
let (trace, nodes, _) = run_simulation_inner(config);
(trace, nodes)
}
/// Run a distribution simulation.
///
/// Creates N `DistributedNode` instances, forms a cluster via join protocol,
/// registers actors, then runs rounds of tick + deliver + resolve.
pub fn run_simulation(config: DistributionSimConfig) -> DistTrace {
let (trace, _, _) = run_simulation_inner(config);
trace
}
fn run_simulation_inner(config: DistributionSimConfig) -> (DistTrace, Vec<Option<DistributedNode>>, Vec<NodeId>) {
let mut events: Vec<Event<DistributionEventKind>> = Vec::new();
let mut snapshots_per_round: Vec<Vec<(String, DistributionSnapshot)>> = Vec::new();
let n = config.num_nodes;
let node_names: Vec<String> = (0..n).map(|i| format!("node-{i}")).collect();
// Create nodes.
let mut nodes: Vec<Option<DistributedNode>> = Vec::with_capacity(n);
let mut node_ids: Vec<NodeId> = Vec::with_capacity(n);
for _i in 0..n {
let mut node_config = DistributedNodeConfig {
swim: config.swim.clone(),
cache_capacity: config.cache_capacity,
republish_interval: 50,
..Default::default()
};
apply_registry_overrides(&mut node_config, &config);
let node = DistributedNode::new(node_config);
node_ids.push(node.node_id());
nodes.push(Some(node));
}
// Form cluster: nodes[1..] join via seed (node 0).
let seed_id = node_ids[0];
for i in 1..n {
// Seed handles join request from node i
let join_actions = nodes[0].as_mut().unwrap().handle_join_request(node_ids[i]);
events.push(Event {
tick: 0,
node_name: node_names[i].clone(),
kind: DistributionEventKind::Joined {
seed_addr: format!("node-0 ({seed_id:?})"),
},
});
// Deliver join response to node i (no network faults during setup).
let mut clean_net = NetworkState::new();
let tagged_responses = deliver_actions_tagged_with_net(
&join_actions,
0,
seed_id,
&mut nodes,
&node_ids,
&mut clean_net,
);
for (responder_idx, response_actions) in tagged_responses {
deliver_actions_tagged_with_net(
&response_actions,
responder_idx,
node_ids[responder_idx],
&mut nodes,
&node_ids,
&mut clean_net,
);
}
}
// Tick-settle: several rounds to let SWIM converge initial membership.
let mut clean_net = NetworkState::new();
for _ in 0..10 {
tick_all_and_deliver(&mut nodes, &node_ids, &mut events, &node_names, 0, &mut clean_net);
}
// Register actors on each node, then propagate entries.
let mut actor_registry: Vec<(ActorAddress, usize)> = Vec::new(); // (actor, owning_node_idx)
let mut pending_entries = Vec::new(); // (entry, owning_node_idx)
for node_idx in 0..n {
if let Some(ref mut node) = nodes[node_idx] {
for a in 0..config.actors_per_node {
let actor = ActorAddress::new_random();
let entry = node.register_actor(actor, a as u64 + 1);
events.push(Event {
tick: 0,
node_name: node_names[node_idx].clone(),
kind: DistributionEventKind::ActorRegistered {
actor_id: format!("{:?}", &actor.0[..4]),
},
});
pending_entries.push((entry, actor, node_idx));
actor_registry.push((actor, node_idx));
}
}
}
// Propagate directory entries to all other nodes.
for (entry, actor, owner_idx) in &pending_entries {
let actor_id = format!("{:?}", &actor.0[..4]);
for other_idx in 0..n {
if other_idx != *owner_idx {
if let Some(ref mut other_node) = nodes[other_idx] {
other_node.store_directory_entry(entry.clone());
events.push(Event {
tick: 0,
node_name: node_names[other_idx].clone(),
kind: DistributionEventKind::ActorStored {
actor_id: actor_id.clone(),
on_node: node_names[*owner_idx].clone(),
},
});
}
}
}
}
// Run simulation rounds.
let mut rng_buf = [0u8; 8];
let mut net = NetworkState::new();
for round in 1..=config.num_rounds {
// Apply network faults for this round.
for fault in &config.network_faults {
let fault_round = match fault {
NetworkFault::Partition { round, .. } => *round,
NetworkFault::Heal { round } => *round,
NetworkFault::SetDropRate { round, .. } => *round,
};
if fault_round == round {
net.apply_fault(fault, n);
}
}
// Apply kill schedule.
for &(kill_round, kill_idx) in &config.kill_schedule {
if kill_round == round && kill_idx < n {
nodes[kill_idx] = None;
events.push(Event {
tick: round as u64,
node_name: node_names[kill_idx].clone(),
kind: DistributionEventKind::NodeKilled,
});
}
}
// Apply revive schedule.
for &(revive_round, revive_idx) in &config.revive_schedule {
if revive_round == round && revive_idx < n {
let mut node_config = DistributedNodeConfig {
swim: config.swim.clone(),
cache_capacity: config.cache_capacity,
republish_interval: 50,
..Default::default()
};
apply_registry_overrides(&mut node_config, &config);
let revived = DistributedNode::new(node_config);
node_ids[revive_idx] = revived.node_id();
nodes[revive_idx] = Some(revived);
// Rejoin the cluster via seed.
let join_actions = nodes[0].as_mut().unwrap().handle_join_request(node_ids[revive_idx]);
let tagged_responses = deliver_actions_tagged_with_net(
&join_actions,
0,
node_ids[0],
&mut nodes,
&node_ids,
&mut net,
);
for (responder_idx, response_actions) in tagged_responses {
deliver_actions_tagged_with_net(
&response_actions,
responder_idx,
node_ids[responder_idx],
&mut nodes,
&node_ids,
&mut net,
);
}
events.push(Event {
tick: round as u64,
node_name: node_names[revive_idx].clone(),
kind: DistributionEventKind::NodeRevived,
});
}
}
// Execute scheduled actions for this round.
for (action_round, action) in &config.action_schedule {
if *action_round == round {
match action {
SimAction::RegisterName { node_idx, name } => {
if *node_idx < n {
if let Some(ref mut node) = nodes[*node_idx] {
let actor = ActorAddress::new_random();
node.register_name(name.clone(), actor);
events.push(Event {
tick: round as u64,
node_name: node_names[*node_idx].clone(),
kind: DistributionEventKind::NameRegistered {
name: name.clone(),
node_idx: *node_idx,
},
});
}
}
}
SimAction::RegisterNameWithActor { node_idx, name, actor } => {
if *node_idx < n {
if let Some(ref mut node) = nodes[*node_idx] {
node.register_name(name.clone(), *actor);
events.push(Event {
tick: round as u64,
node_name: node_names[*node_idx].clone(),
kind: DistributionEventKind::NameRegistered {
name: name.clone(),
node_idx: *node_idx,
},
});
}
}
}
SimAction::UnregisterName { node_idx, name } => {
if *node_idx < n {
if let Some(ref mut node) = nodes[*node_idx] {
node.unregister_name(name);
events.push(Event {
tick: round as u64,
node_name: node_names[*node_idx].clone(),
kind: DistributionEventKind::NameUnregistered {
name: name.clone(),
node_idx: *node_idx,
},
});
}
}
}
SimAction::GracefulLeave { node_idx } => {
if *node_idx < n {
if let Some(ref mut node) = nodes[*node_idx] {
let leave_actions = node.leave();
// Deliver the leave actions (disseminate death announcement)
let tagged_responses = deliver_actions_tagged_with_net(
&leave_actions,
*node_idx,
node_ids[*node_idx],
&mut nodes,
&node_ids,
&mut net,
);
for (responder_idx, response_actions) in tagged_responses {
deliver_actions_tagged_with_net(
&response_actions,
responder_idx,
node_ids[responder_idx],
&mut nodes,
&node_ids,
&mut net,
);
}
}
// Remove the node after leave
nodes[*node_idx] = None;
events.push(Event {
tick: round as u64,
node_name: node_names[*node_idx].clone(),
kind: DistributionEventKind::NodeKilled,
});
}
}
}
}
}
// Tick all live nodes and deliver actions.
for _ in 0..config.ticks_per_round {
tick_all_and_deliver(
&mut nodes,
&node_ids,
&mut events,
&node_names,
round as u64,
&mut net,
);
}
// Resolve actors from random nodes.
getrandom::getrandom(&mut rng_buf).unwrap();
let resolver_idx = usize::from_ne_bytes(rng_buf) % n;
for &(actor, _owner_idx) in &actor_registry {
if let Some(ref mut resolver) = nodes[resolver_idx] {
let result = resolver.resolve_actor(&actor);
let actor_id = format!("{:?}", &actor.0[..4]);
match result {
ResolveResult::Cached(found_on) => {
events.push(Event {
tick: round as u64,
node_name: node_names[resolver_idx].clone(),
kind: DistributionEventKind::ActorResolved {
actor_id,
found_on: format!("{found_on:?}"),
},
});
}
ResolveResult::NeedsLookup { .. } => {
events.push(Event {
tick: round as u64,
node_name: node_names[resolver_idx].clone(),
kind: DistributionEventKind::ActorResolveFailed {
actor_id,
reason: "needs_lookup".into(),
},
});
}
ResolveResult::NotFound => {
events.push(Event {
tick: round as u64,
node_name: node_names[resolver_idx].clone(),
kind: DistributionEventKind::ActorResolveFailed {
actor_id,
reason: "not_found".into(),
},
});
}
}
}
}
// Snapshot all nodes.
let mut round_snapshots = Vec::new();
for (idx, maybe_node) in nodes.iter_mut().enumerate() {
let snap = match maybe_node {
Some(node) => DistributionSnapshot {
member_count: node.members().len(),
routing_table_size: node.routing_table().len(),
directory_entry_count: node.directory().entry_count(),
cache_size: node.cache().len(),
repair_queue_size: node.repair_queue().len(),
registry_size: node.registry().len(),
registry_tombstone_count: node.registry().tombstone_count(),
is_alive: true,
},
None => DistributionSnapshot {
member_count: 0,
routing_table_size: 0,
directory_entry_count: 0,
cache_size: 0,
repair_queue_size: 0,
registry_size: 0,
registry_tombstone_count: 0,
is_alive: false,
},
};
round_snapshots.push((node_names[idx].clone(), snap));
}
snapshots_per_round.push(round_snapshots);
}
// Build topology edges (all-to-seed for the join topology).
let topology_edges: Vec<(String, String)> = (1..n)
.map(|i| (node_names[i].clone(), node_names[0].clone()))
.collect();
let trace = SimulationTrace {
name: config.name,
trace_type: "distribution".into(),
node_names,
topology_edges,
events,
snapshots_per_round,
num_rounds: config.num_rounds,
};
(trace, nodes, node_ids)
}
fn apply_registry_overrides(node_config: &mut DistributedNodeConfig, config: &DistributionSimConfig) {
if let Some(ttl) = config.registry_tombstone_ttl {
node_config.registry.tombstone_ttl = ttl;
}
if let Some(interval) = config.registry_gc_interval {
node_config.registry.gc_interval = interval;
}
if let Some(lambda) = config.registry_dissemination_lambda {
node_config.registry.dissemination_lambda = lambda;
}
}
/// Tick all live nodes and deliver their actions to other nodes.
fn tick_all_and_deliver(
nodes: &mut [Option<DistributedNode>],
node_ids: &[NodeId],
events: &mut Vec<Event<DistributionEventKind>>,
node_names: &[String],
tick: u64,
net: &mut NetworkState,
) {
let n = nodes.len();
// Collect tick actions from all live nodes.
let mut all_actions: Vec<(usize, Vec<NodeAction>)> = Vec::new();
for idx in 0..n {
if let Some(ref mut node) = nodes[idx] {
let actions = node.tick();
// Record membership changes as events.
for action in &actions {
if let NodeAction::MembershipChanged { node_id, state, .. } = action {
events.push(Event {
tick,
node_name: node_names[idx].clone(),
kind: DistributionEventKind::MembershipChanged {
target: format!("{node_id:?}"),
new_state: format!("{state:?}"),
},
});
}
}
if !actions.is_empty() {
all_actions.push((idx, actions));
}
}
}
// Deliver all actions and collect responses.
for (sender_idx, actions) in all_actions {
let tagged_responses = deliver_actions_tagged_with_net(
&actions,
sender_idx,
node_ids[sender_idx],
nodes,
node_ids,
net,
);
// Deliver responses back, using the actual responder's identity.
for (responder_idx, response_actions) in tagged_responses {
deliver_actions_tagged_with_net(
&response_actions,
responder_idx,
node_ids[responder_idx],
nodes,
node_ids,
net,
);
}
}
}
/// Deliver actions to the appropriate target nodes, respecting network conditions.
/// Returns responses tagged with the index of the responding node.
/// `None` nodes (killed) silently drop actions — simulates network loss.
fn deliver_actions_tagged_with_net(
actions: &[NodeAction],
sender_idx: usize,
sender_id: NodeId,
nodes: &mut [Option<DistributedNode>],
node_ids: &[NodeId],
net: &mut NetworkState,
) -> Vec<(usize, Vec<NodeAction>)> {
let mut tagged_responses: Vec<(usize, Vec<NodeAction>)> = Vec::new();
for action in actions {
match action {
NodeAction::SendPing {
to,
sequence,
piggyback,
..
} => {
if let Some(idx) = node_ids.iter().position(|id| id == to) {
if net.should_deliver(sender_idx, idx) {
if let Some(ref mut node) = nodes[idx] {
let resp =
node.handle_ping(sender_id, *sequence, piggyback);
if !resp.is_empty() {
tagged_responses.push((idx, resp));
}
}
}
}
}
NodeAction::SendAck {
to,
sequence,
piggyback,
..
} => {
if let Some(idx) = node_ids.iter().position(|id| id == to) {
if net.should_deliver(sender_idx, idx) {
if let Some(ref mut node) = nodes[idx] {
let resp = node.handle_ack(sender_id, *sequence, piggyback);
if !resp.is_empty() {
tagged_responses.push((idx, resp));
}
}
}
}
}
NodeAction::SendJoinResponse { to, members, .. } => {
if let Some(idx) = node_ids.iter().position(|id| id == to) {
if net.should_deliver(sender_idx, idx) {
if let Some(ref mut node) = nodes[idx] {
let resp = node.handle_join_response(members.clone());
if !resp.is_empty() {
tagged_responses.push((idx, resp));
}
}
}
}
}
NodeAction::SendPingReq {
relay,
target,
sequence,
piggyback,
..
} => {
if let Some(idx) = node_ids.iter().position(|id| id == relay) {
if net.should_deliver(sender_idx, idx) {
if let Some(ref mut node) = nodes[idx] {
let resp = node.handle_ping_req(
sender_id,
*target,
*sequence,
piggyback,
);
if !resp.is_empty() {
tagged_responses.push((idx, resp));
}
}
}
}
}
NodeAction::MembershipChanged { .. } => {
// Notifications — no delivery needed
}
}
}
tagged_responses
}