feat(quotas): Add an opportunistic cache for Quotas

CHANGELOG.md

@@ -16,6 +16,7 @@
 - Add `http.request.header.cookie` to fields scrubbed by cookie rules. ([#5408](https://github.com/getsentry/relay/pull/5408))
 - Map `request_id` to `trace_id` in vercel log drain transform. ([#5333](https://github.com/getsentry/relay/pull/5333))
 - Populate `sentry.platform` during OTLP ingestion. ([#5411](https://github.com/getsentry/relay/pull/5411))
+- Implement optional opportunistic caching for quotas. ([#5416](https://github.com/getsentry/relay/pull/5416))
 - Add db attribute normalization to V2 spans. ([#5399](https://github.com/getsentry/relay/pull/5399))
 
 **Internal**:

relay-config/src/config.rs

@@ -1193,6 +1193,13 @@ pub struct Processing {
     /// Maximum rate limit to report to clients.
     #[serde(default = "default_max_rate_limit")]
     pub max_rate_limit: Option<u32>,
+    /// Configures the quota cache.
+    ///
+    /// The quota cache, caches the specified percentage of remaining quota in memory to reduce the
+    /// amount of synchronizations required with Redis.
+    ///
+    /// By default quota caching is disabled.
+    pub quota_cache_percentage: Option<f32>,
     /// Configuration for attachment uploads.
     #[serde(default)]
     pub upload: UploadServiceConfig,
@@ -1214,6 +1221,7 @@ impl Default for Processing {
             attachment_chunk_size: default_chunk_size(),
             projectconfig_cache_prefix: default_projectconfig_cache_prefix(),
             max_rate_limit: default_max_rate_limit(),
+            quota_cache_percentage: None,
             upload: UploadServiceConfig::default(),
         }
     }
@@ -2601,6 +2609,11 @@ impl Config {
         self.values.processing.max_rate_limit.map(u32::into)
     }
 
+    /// Amount of remaining quota which is cached in memory.
+    pub fn quota_cache_percentage(&self) -> Option<f32> {
+        self.values.processing.quota_cache_percentage
+    }
+
     /// Cache vacuum interval for the cardinality limiter in memory cache.
     ///
     /// The cache will scan for expired values based on this interval.

relay-quotas/Cargo.toml

@@ -19,11 +19,13 @@ workspace = true
 [dependencies]
 async-trait = { workspace = true }
 hashbrown = { workspace = true }
+papaya = { workspace = true }
 itertools = { workspace = true }
 relay-base-schema = { workspace = true }
 relay-common = { workspace = true }
 relay-log = { workspace = true, optional = true }
 relay-redis = { workspace = true, optional = true }
+relay-statsd = { workspace = true }
 serde = { workspace = true }
 smallvec = { workspace = true }
 thiserror = { workspace = true, optional = true }