questdb-tsbs-benchmark

name	questdb-tsbs-benchmark
description	Run full TSBS (Time Series Benchmark Suite) benchmarks against QuestDB in Docker, including prerequisite checks, TSBS build, data generation, loading, query generation, and benchmark execution. Use when asked to set up or run QuestDB TSBS performance tests end-to-end.

TSBS Benchmark for QuestDB

Run the full TSBS (Time Series Benchmark Suite) against QuestDB running in Docker. Handle all prerequisites, data generation, loading, and query benchmarking.

Prerequisites Check And Install

1. Docker

Check if Docker is installed and the daemon is running:

docker --version && docker info --format '{{.ServerVersion}}'

If Docker is not installed, install it (Ubuntu/Debian):

sudo apt-get update -qq
sudo apt-get install -y -qq ca-certificates curl
sudo install -m 0755 -d /etc/apt/keyrings
sudo curl -fsSL https://download.docker.com/linux/ubuntu/gpg -o /etc/apt/keyrings/docker.asc
sudo chmod a+r /etc/apt/keyrings/docker.asc
echo "deb [arch=$(dpkg --print-architecture) signed-by=/etc/apt/keyrings/docker.asc] https://download.docker.com/linux/ubuntu $(. /etc/os-release && echo "$VERSION_CODENAME") stable" | sudo tee /etc/apt/sources.list.d/docker.list > /dev/null
sudo apt-get update -qq
sudo apt-get install -y -qq docker-ce docker-ce-cli containerd.io docker-buildx-plugin

2. Go (needed to build TSBS)

Check if Go is installed:

go version

If Go is not installed, install Go 1.22.5:

curl -fsSL https://go.dev/dl/go1.22.5.linux-amd64.tar.gz -o /tmp/go.tar.gz
sudo rm -rf /usr/local/go && sudo tar -C /usr/local -xzf /tmp/go.tar.gz && rm /tmp/go.tar.gz
export PATH=$PATH:/usr/local/go/bin:$HOME/go/bin

3. Build tools

sudo apt-get install -y -qq make gcc gzip

Step 1: Start QuestDB in Docker

Stop and remove any existing QuestDB container first to start clean:

sudo docker rm -f questdb 2>/dev/null || true

Pull latest image and start:

sudo docker pull questdb/questdb:latest
sudo docker run -d --name questdb \
  -p 9000:9000 -p 9009:9009 -p 8812:8812 -p 9003:9003 \
  questdb/questdb:latest

Verify it is running:

sudo docker ps --filter name=questdb --format '{{.Status}}'

Step 2: Clone And Build TSBS

cd /home/ubuntu
git clone https://github.com/questdb/tsbs.git
cd /home/ubuntu/tsbs
export PATH=$PATH:/usr/local/go/bin:$HOME/go/bin
make tsbs_generate_data tsbs_generate_queries tsbs_load_questdb tsbs_run_queries_questdb

Build only the four binaries needed for QuestDB benchmarking into bin/.

Step 3: Generate Benchmark Data

IMPORTANT: Do not compress (gzip) the data file. Compression/decompression adds CPU overhead that skews benchmark results, especially on smaller machines.

/home/ubuntu/tsbs/bin/tsbs_generate_data \
  --use-case="cpu-only" \
  --seed=123 \
  --scale=4000 \
  --timestamp-start="2016-01-01T00:00:00Z" \
  --timestamp-end="2016-01-02T00:00:00Z" \
  --log-interval="10s" \
  --format="questdb" \
  > /tmp/questdb-data.txt

This generates approximately 12GB of uncompressed data (34.5M rows, 345.6M metrics).

Step 4: Load Data Into QuestDB

Use as many workers as CPU cores available (up to 32). Capture the load summary and report ingestion throughput in rows/s (and metrics/s):

WORKERS=$(nproc)
if [ "$WORKERS" -gt 32 ]; then WORKERS=32; fi

/home/ubuntu/tsbs/bin/tsbs_load_questdb \
  --file=/tmp/questdb-data.txt \
  --workers=$WORKERS \
  2>&1 | tee /tmp/questdb-load.log

# Report ingestion throughput from TSBS summary
# Example output line includes: "overall row/s"
grep -E "loaded .*metrics|loaded .*rows" /tmp/questdb-load.log

Expected summary lines include the mean rates, e.g. overall row/s for rows/second and overall metric/s for metrics/second.

Step 5: Generate Query Files

Generate 1000 queries for all 16 cpu-only query types. Keep all query files uncompressed:

TSBS_BIN=/home/ubuntu/tsbs/bin
COMMON_ARGS="--use-case=cpu-only --seed=123 --scale=4000 --timestamp-start=2016-01-01T00:00:00Z --timestamp-end=2016-01-02T00:00:01Z --queries=1000 --format=questdb"

for QTYPE in \
  cpu-max-all-1 cpu-max-all-8 cpu-max-all-32-24 \
  single-groupby-1-1-1 single-groupby-1-1-12 single-groupby-1-8-1 \
  single-groupby-5-1-1 single-groupby-5-1-12 single-groupby-5-8-1 \
  double-groupby-1 double-groupby-5 double-groupby-all \
  high-cpu-1 high-cpu-all \
  lastpoint groupby-orderby-limit; do
  $TSBS_BIN/tsbs_generate_queries $COMMON_ARGS --query-type="$QTYPE" > "/tmp/questdb-queries-${QTYPE}.txt"
done

Step 6: Run Query Benchmarks

IMPORTANT: Use one worker for queries. QuestDB parallelizes queries internally (multi-threaded execution), so multiple client workers over-subscribe CPU and produce misleading results.

TSBS_BIN=/home/ubuntu/tsbs/bin

for QTYPE in \
  cpu-max-all-1 cpu-max-all-8 cpu-max-all-32-24 \
  single-groupby-1-1-1 single-groupby-1-1-12 single-groupby-1-8-1 \
  single-groupby-5-1-1 single-groupby-5-1-12 single-groupby-5-8-1 \
  double-groupby-1 double-groupby-5 double-groupby-all \
  high-cpu-1 high-cpu-all \
  lastpoint groupby-orderby-limit; do
  echo "=== $QTYPE ==="
  $TSBS_BIN/tsbs_run_queries_questdb \
    --file="/tmp/questdb-queries-${QTYPE}.txt" \
    --workers=1 \
    --print-interval=0
  echo ""
done

Cleanup

When done benchmarking:

sudo docker rm -f questdb
rm -f /tmp/questdb-data.txt /tmp/questdb-queries-*.txt

Notes

• cpu-only with scale=4000 and a 1-day window is a solid general benchmark.
• For --use-case, also try devops or iot (each has different query types).
• Run tsbs_generate_queries --help to see the full matrix of use-case + query-type combinations.
• Port 9000 is Web Console (HTTP), 9009 is ILP (line protocol), and 8812 is PostgreSQL wire.
• The data loading step uses ILP (9009); queries use PostgreSQL wire (8812).