Skip to content
Open
Show file tree
Hide file tree
Changes from all commits
Commits
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
33 changes: 33 additions & 0 deletions scramdb-tuned/README.md
Original file line number Diff line number Diff line change
@@ -0,0 +1,33 @@
# ScramDB (tuned)

`tuned: yes` companion to [`scramdb`](../scramdb). Same image, harness and
schema, only the config differs. See that README for the rest.

bash benchmark.sh

This configuration has been validated against all 43 queries on the full
dataset.

## Tuned

- `[storage.io] backend = "buffered"`: reads go through the OS page cache, so
a cold read pays disk once and every warm re-read of a hot byte is RAM
- `buffer_pool_percent = 0` (auto: 15% on the buffered backend, where the
page cache is the primary read cache and a large private pool would hold
the same bytes twice)
- `execution_memory_percent` 25, `working_set_percent` 85
- compiled-query cache 256MB memory / 1GB disk: 43 distinct query kernels
need only a few MiB; a larger cache takes memory from execution
- `morsel_size` 8MB, `arena_chunk_size` 8MB
- `copy_batch_rows` 131072, `copy_pipeline_depth` 8 (load path only)
- `spill_agg_partitions` 128 - the high-cardinality GROUP BY and exact
COUNT(DISTINCT) queries all spill
- `prefetch_depth` 32, `prefetch_queue_depth` 256, `bloom_max_bytes` 16KB
- `flush_threshold_rows` 65536
- background compaction, GC and auto-analyze pushed past the run length; the
driver restarts before all 43 queries, so nothing wakes mid-query
- `analyze_on_load` off; `./load` runs the one explicit `ANALYZE` instead
- UDF and GPU lanes off, so neither reserves memory from the budget
- shutdown drain 5s, for the harness's per-query cold restart cycle

Join knobs are untouched (no joins in the 43 queries).
12 changes: 12 additions & 0 deletions scramdb-tuned/benchmark.sh
Original file line number Diff line number Diff line change
@@ -0,0 +1,12 @@
#!/bin/bash

export BENCH_DOWNLOAD_SCRIPT="download-hits-csv"

export PGHOST="${PGHOST:-localhost}"
export PGPORT="${PGPORT:-5432}"
export PGUSER="${PGUSER:-scramdb}"
export PGDATABASE="${PGDATABASE:-scramdb}"
export PGCONNECT_TIMEOUT="${PGCONNECT_TIMEOUT:-10}"
export SCRAMDB_IMAGE="${SCRAMDB_IMAGE:-scramdb/scramdb:cb}"

exec ../lib/benchmark-common.sh
3 changes: 3 additions & 0 deletions scramdb-tuned/check
Original file line number Diff line number Diff line change
@@ -0,0 +1,3 @@
#!/bin/bash
set -e
psql -t -c 'SELECT 1' >/dev/null
108 changes: 108 additions & 0 deletions scramdb-tuned/create.sql
Original file line number Diff line number Diff line change
@@ -0,0 +1,108 @@
CREATE TABLE hits
(
WatchID BIGINT NOT NULL,
JavaEnable SMALLINT NOT NULL,
Title TEXT NOT NULL,
GoodEvent SMALLINT NOT NULL,
EventTime TIMESTAMP NOT NULL,
EventDate Date NOT NULL,
CounterID INTEGER NOT NULL,
ClientIP INTEGER NOT NULL,
RegionID INTEGER NOT NULL,
UserID BIGINT NOT NULL,
CounterClass SMALLINT NOT NULL,
OS SMALLINT NOT NULL,
UserAgent SMALLINT NOT NULL,
URL TEXT NOT NULL,
Referer TEXT NOT NULL,
IsRefresh SMALLINT NOT NULL,
RefererCategoryID SMALLINT NOT NULL,
RefererRegionID INTEGER NOT NULL,
URLCategoryID SMALLINT NOT NULL,
URLRegionID INTEGER NOT NULL,
ResolutionWidth SMALLINT NOT NULL,
ResolutionHeight SMALLINT NOT NULL,
ResolutionDepth SMALLINT NOT NULL,
FlashMajor SMALLINT NOT NULL,
FlashMinor SMALLINT NOT NULL,
FlashMinor2 TEXT NOT NULL,
NetMajor SMALLINT NOT NULL,
NetMinor SMALLINT NOT NULL,
UserAgentMajor SMALLINT NOT NULL,
UserAgentMinor VARCHAR(255) NOT NULL,
CookieEnable SMALLINT NOT NULL,
JavascriptEnable SMALLINT NOT NULL,
IsMobile SMALLINT NOT NULL,
MobilePhone SMALLINT NOT NULL,
MobilePhoneModel TEXT NOT NULL,
Params TEXT NOT NULL,
IPNetworkID INTEGER NOT NULL,
TraficSourceID SMALLINT NOT NULL,
SearchEngineID SMALLINT NOT NULL,
SearchPhrase TEXT NOT NULL,
AdvEngineID SMALLINT NOT NULL,
IsArtifical SMALLINT NOT NULL,
WindowClientWidth SMALLINT NOT NULL,
WindowClientHeight SMALLINT NOT NULL,
ClientTimeZone SMALLINT NOT NULL,
ClientEventTime TIMESTAMP NOT NULL,
SilverlightVersion1 SMALLINT NOT NULL,
SilverlightVersion2 SMALLINT NOT NULL,
SilverlightVersion3 INTEGER NOT NULL,
SilverlightVersion4 SMALLINT NOT NULL,
PageCharset TEXT NOT NULL,
CodeVersion INTEGER NOT NULL,
IsLink SMALLINT NOT NULL,
IsDownload SMALLINT NOT NULL,
IsNotBounce SMALLINT NOT NULL,
FUniqID BIGINT NOT NULL,
OriginalURL TEXT NOT NULL,
HID INTEGER NOT NULL,
IsOldCounter SMALLINT NOT NULL,
IsEvent SMALLINT NOT NULL,
IsParameter SMALLINT NOT NULL,
DontCountHits SMALLINT NOT NULL,
WithHash SMALLINT NOT NULL,
HitColor CHAR NOT NULL,
LocalEventTime TIMESTAMP NOT NULL,
Age SMALLINT NOT NULL,
Sex SMALLINT NOT NULL,
Income SMALLINT NOT NULL,
Interests SMALLINT NOT NULL,
Robotness SMALLINT NOT NULL,
RemoteIP INTEGER NOT NULL,
WindowName INTEGER NOT NULL,
OpenerName INTEGER NOT NULL,
HistoryLength SMALLINT NOT NULL,
BrowserLanguage TEXT NOT NULL,
BrowserCountry TEXT NOT NULL,
SocialNetwork TEXT NOT NULL,
SocialAction TEXT NOT NULL,
HTTPError SMALLINT NOT NULL,
SendTiming INTEGER NOT NULL,
DNSTiming INTEGER NOT NULL,
ConnectTiming INTEGER NOT NULL,
ResponseStartTiming INTEGER NOT NULL,
ResponseEndTiming INTEGER NOT NULL,
FetchTiming INTEGER NOT NULL,
SocialSourceNetworkID SMALLINT NOT NULL,
SocialSourcePage TEXT NOT NULL,
ParamPrice BIGINT NOT NULL,
ParamOrderID TEXT NOT NULL,
ParamCurrency TEXT NOT NULL,
ParamCurrencyID SMALLINT NOT NULL,
OpenstatServiceName TEXT NOT NULL,
OpenstatCampaignID TEXT NOT NULL,
OpenstatAdID TEXT NOT NULL,
OpenstatSourceID TEXT NOT NULL,
UTMSource TEXT NOT NULL,
UTMMedium TEXT NOT NULL,
UTMCampaign TEXT NOT NULL,
UTMContent TEXT NOT NULL,
UTMTerm TEXT NOT NULL,
FromTag TEXT NOT NULL,
HasGCLID SMALLINT NOT NULL,
RefererHash BIGINT NOT NULL,
URLHash BIGINT NOT NULL,
CLID INTEGER NOT NULL
);
109 changes: 109 additions & 0 deletions scramdb-tuned/create.sql.with-pk
Original file line number Diff line number Diff line change
@@ -0,0 +1,109 @@
CREATE TABLE hits
(
WatchID BIGINT NOT NULL,
JavaEnable SMALLINT NOT NULL,
Title TEXT NOT NULL,
GoodEvent SMALLINT NOT NULL,
EventTime TIMESTAMP NOT NULL,
EventDate Date NOT NULL,
CounterID INTEGER NOT NULL,
ClientIP INTEGER NOT NULL,
RegionID INTEGER NOT NULL,
UserID BIGINT NOT NULL,
CounterClass SMALLINT NOT NULL,
OS SMALLINT NOT NULL,
UserAgent SMALLINT NOT NULL,
URL TEXT NOT NULL,
Referer TEXT NOT NULL,
IsRefresh SMALLINT NOT NULL,
RefererCategoryID SMALLINT NOT NULL,
RefererRegionID INTEGER NOT NULL,
URLCategoryID SMALLINT NOT NULL,
URLRegionID INTEGER NOT NULL,
ResolutionWidth SMALLINT NOT NULL,
ResolutionHeight SMALLINT NOT NULL,
ResolutionDepth SMALLINT NOT NULL,
FlashMajor SMALLINT NOT NULL,
FlashMinor SMALLINT NOT NULL,
FlashMinor2 TEXT NOT NULL,
NetMajor SMALLINT NOT NULL,
NetMinor SMALLINT NOT NULL,
UserAgentMajor SMALLINT NOT NULL,
UserAgentMinor VARCHAR(255) NOT NULL,
CookieEnable SMALLINT NOT NULL,
JavascriptEnable SMALLINT NOT NULL,
IsMobile SMALLINT NOT NULL,
MobilePhone SMALLINT NOT NULL,
MobilePhoneModel TEXT NOT NULL,
Params TEXT NOT NULL,
IPNetworkID INTEGER NOT NULL,
TraficSourceID SMALLINT NOT NULL,
SearchEngineID SMALLINT NOT NULL,
SearchPhrase TEXT NOT NULL,
AdvEngineID SMALLINT NOT NULL,
IsArtifical SMALLINT NOT NULL,
WindowClientWidth SMALLINT NOT NULL,
WindowClientHeight SMALLINT NOT NULL,
ClientTimeZone SMALLINT NOT NULL,
ClientEventTime TIMESTAMP NOT NULL,
SilverlightVersion1 SMALLINT NOT NULL,
SilverlightVersion2 SMALLINT NOT NULL,
SilverlightVersion3 INTEGER NOT NULL,
SilverlightVersion4 SMALLINT NOT NULL,
PageCharset TEXT NOT NULL,
CodeVersion INTEGER NOT NULL,
IsLink SMALLINT NOT NULL,
IsDownload SMALLINT NOT NULL,
IsNotBounce SMALLINT NOT NULL,
FUniqID BIGINT NOT NULL,
OriginalURL TEXT NOT NULL,
HID INTEGER NOT NULL,
IsOldCounter SMALLINT NOT NULL,
IsEvent SMALLINT NOT NULL,
IsParameter SMALLINT NOT NULL,
DontCountHits SMALLINT NOT NULL,
WithHash SMALLINT NOT NULL,
HitColor CHAR NOT NULL,
LocalEventTime TIMESTAMP NOT NULL,
Age SMALLINT NOT NULL,
Sex SMALLINT NOT NULL,
Income SMALLINT NOT NULL,
Interests SMALLINT NOT NULL,
Robotness SMALLINT NOT NULL,
RemoteIP INTEGER NOT NULL,
WindowName INTEGER NOT NULL,
OpenerName INTEGER NOT NULL,
HistoryLength SMALLINT NOT NULL,
BrowserLanguage TEXT NOT NULL,
BrowserCountry TEXT NOT NULL,
SocialNetwork TEXT NOT NULL,
SocialAction TEXT NOT NULL,
HTTPError SMALLINT NOT NULL,
SendTiming INTEGER NOT NULL,
DNSTiming INTEGER NOT NULL,
ConnectTiming INTEGER NOT NULL,
ResponseStartTiming INTEGER NOT NULL,
ResponseEndTiming INTEGER NOT NULL,
FetchTiming INTEGER NOT NULL,
SocialSourceNetworkID SMALLINT NOT NULL,
SocialSourcePage TEXT NOT NULL,
ParamPrice BIGINT NOT NULL,
ParamOrderID TEXT NOT NULL,
ParamCurrency TEXT NOT NULL,
ParamCurrencyID SMALLINT NOT NULL,
OpenstatServiceName TEXT NOT NULL,
OpenstatCampaignID TEXT NOT NULL,
OpenstatAdID TEXT NOT NULL,
OpenstatSourceID TEXT NOT NULL,
UTMSource TEXT NOT NULL,
UTMMedium TEXT NOT NULL,
UTMCampaign TEXT NOT NULL,
UTMContent TEXT NOT NULL,
UTMTerm TEXT NOT NULL,
FromTag TEXT NOT NULL,
HasGCLID SMALLINT NOT NULL,
RefererHash BIGINT NOT NULL,
URLHash BIGINT NOT NULL,
CLID INTEGER NOT NULL,
PRIMARY KEY (CounterID, EventDate, UserID, EventTime, WatchID)
);
3 changes: 3 additions & 0 deletions scramdb-tuned/data-size
Original file line number Diff line number Diff line change
@@ -0,0 +1,3 @@
#!/bin/bash
set -eu
sudo du -sb --exclude=scramvm_cache ./db | cut -f1
92 changes: 92 additions & 0 deletions scramdb-tuned/install
Original file line number Diff line number Diff line change
@@ -0,0 +1,92 @@
#!/bin/bash
set -eu

sudo apt-get update -y
sudo apt-get install -y postgresql-client docker.io bc

./stop || true
sudo docker rm -f scramdb 2>/dev/null || true
sudo rm -rf ./db
mkdir -p ./db
chmod 777 ./db

ram_kb=$(grep MemTotal /proc/meminfo | tr -dc 0-9)
ram_gb=$(( ram_kb / 1024 / 1024 ))

if [ "$ram_gb" -lt 16 ]; then
echo "WARNING: ScramDB targets machines with at least 16 GB of RAM; this one has ${ram_gb} GB." >&2
echo "WARNING: proceeding anyway, but this result is not representative of the product." >&2
fi

# The tuned configuration. The README lists every deviation from the product
# default and its reason.
cat > scramdb-config.toml <<EOF
[general]
jit_enabled = true
metrics_port = 9090
pg_address = "0.0.0.0:5432"
shutdown_jit_quiesce_timeout_secs = 10
shutdown_drain_timeout_secs = 5

[jit]
compiled_cache_memory = "256MB"
compiled_cache_disk = "1024MB"

[storage]
prod_name = "tundra"
shard_id = 0
basedir = "/var/lib/scramdb"
# 0 = auto: the share is derived from the io backend.
buffer_pool_percent = 0
buffer_pool_cap = "192GB"
execution_memory_percent = 25
execution_memory_cap = "64GB"

flush_threshold_rows = 65536
prefetch_depth = 32
prefetch_queue_depth = 256
bloom_max_bytes = "16KB"

[storage.memory]
working_set_percent = 85

[storage.io]
backend = "buffered"

[storage.wal.archive]
enabled = false

[storage.cold_config]
cold_enable = false
writeback_frequency_secs = 0
concurrent_requests = 0
bucket_name = ""
cloud_api = "GCP"
base_path = ""

[execution]
workers = "auto"
morsel_size = "8MB"
copy_batch_rows = 131072
copy_pipeline_depth = 8
spill_agg_partitions = 128
arena_chunk_size = "8MB"

[storage.compaction]
check_interval = "24h"

[storage.gc]
cycle_interval = "24h"

[statistics]
analyze_on_load = false
auto_analyze_check_interval = "24h"

[udf]
enabled = false

[gpu]
enabled = false
EOF

sudo docker pull "${SCRAMDB_IMAGE:-scramdb/scramdb:cb}"
10 changes: 10 additions & 0 deletions scramdb-tuned/load
Original file line number Diff line number Diff line change
@@ -0,0 +1,10 @@
#!/bin/bash
set -eu

psql -v ON_ERROR_STOP=1 -f create.sql
psql -v ON_ERROR_STOP=1 -c "COPY hits FROM '/bench/hits.csv' WITH (FORMAT CSV)"
psql -v ON_ERROR_STOP=1 -c "CHECKPOINT"
psql -v ON_ERROR_STOP=1 -c "ANALYZE hits"

rm -f hits.csv
sync
Loading