feat(search): pin Foyer and add split-range disk cache config

LICENSE-3rdparty.csv

@@ -161,6 +161,7 @@ clap_builder,https://github.com/clap-rs/clap,MIT OR Apache-2.0,The clap_builder
 clap_lex,https://github.com/clap-rs/clap,MIT OR Apache-2.0,The clap_lex Authors
 cmac,https://github.com/RustCrypto/MACs,MIT OR Apache-2.0,RustCrypto Developers
 cmov,https://github.com/RustCrypto/utils,Apache-2.0 OR MIT,RustCrypto Developers
+cmsketch,https://github.com/mrcroxx/cmsketch-rs,Apache-2.0,MrCroxx <mrcroxx@outlook.com>
 coarsetime,https://github.com/jedisct1/rust-coarsetime,BSD-2-Clause,Frank Denis <github@pureftpd.org>
 cobs,https://github.com/jamesmunns/cobs.rs,MIT OR Apache-2.0,"Allen Welkie <>, James Munns <james@onevariable.com>"
 codespan-reporting,https://github.com/brendanzab/codespan,Apache-2.0,Brendan Zabarauskas <bjzaba@yahoo.com.au>
@@ -186,6 +187,7 @@ constant_time_eq,https://github.com/cesarb/constant_time_eq,CC0-1.0 OR MIT-0 OR
 convert_case,https://github.com/rutrum/convert-case,MIT,rutrum <dave@rutrum.net>
 core-foundation,https://github.com/servo/core-foundation-rs,MIT OR Apache-2.0,The Servo Project Developers
 core-foundation-sys,https://github.com/servo/core-foundation-rs,MIT OR Apache-2.0,The Servo Project Developers
+core_affinity,https://github.com/Elzair/core_affinity_rs,MIT OR Apache-2.0,Philip Woods <elzairthesorcerer@gmail.com>
 cpp_demangle,https://github.com/gimli-rs/cpp_demangle,MIT OR Apache-2.0,"Nick Fitzgerald <fitzgen@gmail.com>, Jim Blandy <jimb@red-bean.com>, Kyle Huey <khuey@kylehuey.com>"
 cpufeatures,https://github.com/RustCrypto/utils,MIT OR Apache-2.0,RustCrypto Developers
 crc,https://github.com/mrhooray/crc-rs,MIT OR Apache-2.0,"Rui Hu <code@mrhooray.com>, Akhil Velagapudi <4@4khil.com>"
@@ -306,6 +308,7 @@ event-listener-strategy,https://github.com/smol-rs/event-listener-strategy,Apach
 evmap,https://github.com/jonhoo/evmap,MIT OR Apache-2.0,Jon Gjengset <jon@thesquareplanet.com>
 fail,https://github.com/tikv/fail-rs,Apache-2.0,The TiKV Project Developers
 fancy-regex,https://github.com/fancy-regex/fancy-regex,MIT,"Raph Levien <raph@google.com>, Robin Stocker <robin@nibor.org>, Keith Hall <keith.hall@available.systems>"
+fastant,https://github.com/fast/fastant,MIT,The fastant Authors
 fastdivide,https://github.com/fulmicoton/fastdivide,zlib-acknowledgement OR MIT,Paul Masurel <paul.masurel@gmail.com>
 fastrand,https://github.com/smol-rs/fastrand,Apache-2.0 OR MIT,Stjepan Glavina <stjepang@gmail.com>
 ff,https://github.com/zkcrypto/ff,MIT OR Apache-2.0,"Sean Bowe <ewillbefull@gmail.com>, Jack Grigg <thestr4d@gmail.com>"
@@ -323,6 +326,12 @@ foldhash,https://github.com/orlp/foldhash,Zlib,Orson Peters <orsonpeters@gmail.c
 foreign-types,https://github.com/sfackler/foreign-types,MIT OR Apache-2.0,Steven Fackler <sfackler@gmail.com>
 foreign-types-shared,https://github.com/sfackler/foreign-types,MIT OR Apache-2.0,Steven Fackler <sfackler@gmail.com>
 form_urlencoded,https://github.com/servo/rust-url,MIT OR Apache-2.0,The rust-url developers
+foyer,https://github.com/foyer-rs/foyer,Apache-2.0,MrCroxx <mrcroxx@outlook.com>
+foyer-common,https://github.com/foyer-rs/foyer,Apache-2.0,MrCroxx <mrcroxx@outlook.com>
+foyer-intrusive-collections,https://github.com/foyer-rs/intrusive-rs,Apache-2.0 OR MIT,Amanieu d'Antras <amanieu@gmail.com>
+foyer-memory,https://github.com/foyer-rs/foyer,Apache-2.0,MrCroxx <mrcroxx@outlook.com>
+foyer-storage,https://github.com/foyer-rs/foyer,Apache-2.0,MrCroxx <mrcroxx@outlook.com>
+foyer-tokio,https://github.com/foyer-rs/foyer,Apache-2.0,MrCroxx <mrcroxx@outlook.com>
 fraction,https://github.com/dnsl48/fraction,MIT OR Apache-2.0,dnsl48 <dnsl48@gmail.com>
 fragile,https://github.com/mitsuhiko/fragile,Apache-2.0,Armin Ronacher <armin.ronacher@active-4.com>
 fs4,https://github.com/al8n/fs4-rs,MIT OR Apache-2.0,"Dan Burkert <dan@danburkert.com>, Al Liu <scygliu1@gmail.com>"
@@ -409,6 +418,7 @@ inout,https://github.com/RustCrypto/utils,MIT OR Apache-2.0,RustCrypto Developer
 instant,https://github.com/sebcrozet/instant,BSD-3-Clause,sebcrozet <developer@crozet.re>
 integer-encoding,https://github.com/dermesser/integer-encoding-rs,MIT,Lewin Bormann <lbo@spheniscida.de>
 inventory,https://github.com/dtolnay/inventory,MIT OR Apache-2.0,David Tolnay <dtolnay@gmail.com>
+io-uring,https://github.com/tokio-rs/io-uring,MIT OR Apache-2.0,quininer <quininer@live.com>
 ipcrypt-rs,https://github.com/jedisct1/rust-ipcrypt2,ISC,Frank Denis <github@pureftpd.org>
 ipnet,https://github.com/krisprice/ipnet,MIT OR Apache-2.0,Kris Price <kris@krisprice.nz>
 ipnetwork,https://github.com/achanda/ipnetwork,MIT OR Apache-2.0,"Abhishek Chanda <abhishek.becs@gmail.com>, Linus Färnstrand <faern@faern.net>"
@@ -473,6 +483,7 @@ mea,https://github.com/fast/mea,Apache-2.0,The mea Authors
 measure_time,https://github.com/PSeitz/rust_measure_time,MIT,Pascal Seitz <pascal.seitz@gmail.com>
 memchr,https://github.com/BurntSushi/memchr,Unlicense OR MIT,"Andrew Gallant <jamslam@gmail.com>, bluss"
 memmap2,https://github.com/RazrFalcon/memmap2-rs,MIT OR Apache-2.0,"Dan Burkert <dan@danburkert.com>, Yevhenii Reizner <razrfalcon@gmail.com>, The Contributors"
+memoffset,https://github.com/Gilnaa/memoffset,MIT,Gilad Naaman <gilad.naaman@gmail.com>
 metrics,https://github.com/metrics-rs/metrics,MIT,Toby Lawrence <toby@nuclearfurnace.com>
 metrics-exporter-prometheus,https://github.com/metrics-rs/metrics,MIT AND Apache-2.0,Toby Lawrence <toby@nuclearfurnace.com>
 metrics-opentelemetry,https://github.com/DoumanAsh/metrics-opentelemetry,BSL-1.0,The metrics-opentelemetry Authors
@@ -485,6 +496,7 @@ minimal-lexical,https://github.com/Alexhuszagh/minimal-lexical,MIT OR Apache-2.0
 miniserde,https://github.com/dtolnay/miniserde,MIT OR Apache-2.0,David Tolnay <dtolnay@gmail.com>
 miniz_oxide,https://github.com/Frommi/miniz_oxide/tree/master/miniz_oxide,MIT OR Zlib OR Apache-2.0,"Frommi <daniil.liferenko@gmail.com>, oyvindln <oyvindln@users.noreply.github.com>, Rich Geldreich richgel99@gmail.com"
 mio,https://github.com/tokio-rs/mio,MIT,"Carl Lerche <me@carllerche.com>, Thomas de Zeeuw <thomasdezeeuw@gmail.com>, Tokio Contributors <team@tokio.rs>"
+mixtrics,https://github.com/foyer-rs/mixtrics,Apache-2.0,MrCroxx <mrcroxx@outlook.com>
 mockall,https://github.com/asomers/mockall,MIT OR Apache-2.0,Alan Somers <asomers@gmail.com>
 mockall_derive,https://github.com/asomers/mockall,MIT OR Apache-2.0,Alan Somers <asomers@gmail.com>
 moka,https://github.com/moka-rs/moka,(MIT OR Apache-2.0) AND Apache-2.0,The moka Authors
@@ -776,6 +788,7 @@ simple_asn1,https://github.com/acw/simple_asn1,ISC,Adam Wick <awick@uhsure.com>
 siphasher,https://github.com/jedisct1/rust-siphash,MIT OR Apache-2.0,Frank Denis <github@pureftpd.org>
 sketches-ddsketch,https://github.com/mheffner/rust-sketches-ddsketch,Apache-2.0,Mike Heffner <mikeh@fesnel.com>
 slab,https://github.com/tokio-rs/slab,MIT,Carl Lerche <me@carllerche.com>
+small_ctor,https://github.com/mitsuhiko/small-ctor,Apache-2.0,Armin Ronacher <armin.ronacher@active-4.com>
 smallvec,https://github.com/servo/rust-smallvec,MIT OR Apache-2.0,The Servo Project Developers
 snafu,https://github.com/shepmaster/snafu,MIT OR Apache-2.0,Jake Goulding <jake.goulding@gmail.com>
 snafu-derive,https://github.com/shepmaster/snafu,MIT OR Apache-2.0,Jake Goulding <jake.goulding@gmail.com>

config/quickwit.yaml

@@ -210,6 +210,22 @@ indexer:
 #      max_num_bytes: 1G
 #      max_num_splits: 10000
 #      num_concurrent_downloads: 1
+#   # Process-wide Foyer disk cache for exact split footer and body ranges.
+#   # Omitted or null disables the cache. write_policy defaults to write-on-eviction.
+#   split_range_disk_cache:
+#     path: /var/cache/quickwit/split-range-v1
+#     disk_capacity: 300G
+#     memory_capacity: 1G
+#     buffer_pool_size: 512M
+#     submit_queue_size_threshold: 1G
+#     memory_eviction_policy: s3-fifo
+#     write_policy: write-on-eviction
+#     compression: lz4
+#     recover_mode: quiet
+#     block_size: 16M
+#     max_entry_size: 15M
+#     flushers: 4
+#     reclaimers: 4
 # -------------------------------- Jaeger settings --------------------------------
 
 jaeger:

quickwit/Cargo.toml

@@ -136,6 +136,7 @@ fail = "0.5"
 flate2 = "1.1"
 flume = "0.12"
 fnv = "1"
+foyer = { version = "=0.22.3", default-features = false, features = ["runtime-tokio"] }
 futures = "0.3"
 futures-util = { version = "0.3", default-features = false }
 glob = "0.3"
@@ -180,6 +181,7 @@ metrics-exporter-prometheus = { version = "0.18", default-features = false }
 metrics-util = "0.20"
 mime_guess = "2.0"
 mini-moka = "0.10"
+mixtrics = "=0.2.5"
 mockall = "0.14"
 mrecordlog = { git = "https://github.com/quickwit-oss/mrecordlog", rev = "3b3562ef" }
 new_string_template = "1.5"

quickwit/quickwit-config/src/lib.rs

@@ -77,10 +77,11 @@ pub use crate::metastore_config::{
     MetastoreBackend, MetastoreConfig, MetastoreConfigs, PostgresMetastoreConfig,
 };
 pub use crate::node_config::{
-    CacheConfig, CachePolicy, CompactorConfig, DEFAULT_QW_CONFIG_PATH, GrpcConfig, HealthConfig,
-    IndexerConfig, IngestApiConfig, JaegerConfig, KeepAliveConfig, LambdaConfig,
-    LambdaDeployConfig, MAX_GOSSIP_PROTOCOL_VERSION, NodeConfig, RestConfig, SearcherConfig,
-    SplitCacheLimits, StorageTimeoutPolicy, TlsConfig,
+    CacheConfig, CachePolicy, CompactorConfig, DEFAULT_QW_CONFIG_PATH, DiskCompression, GrpcConfig,
+    HealthConfig, IndexerConfig, IngestApiConfig, JaegerConfig, KeepAliveConfig, LambdaConfig,
+    LambdaDeployConfig, MAX_GOSSIP_PROTOCOL_VERSION, NodeConfig, RecoverMode, RestConfig,
+    SearcherConfig, SplitCacheLimits, SplitRangeCacheWritePolicy, SplitRangeDiskCacheConfig,
+    StorageTimeoutPolicy, TlsConfig,
 };
 pub use crate::serde_utils::HumanDuration;
 use crate::source_config::serialize::{SourceConfigV0_7, SourceConfigV0_8, VersionedSourceConfig};