Skip to main content

AI-powered Apache Spark job analyzer and configuration advisor

Project description

↩ spark-advisor

spark-advisor-cli

AI-powered Apache Spark job analyzer and configuration advisor. Standalone CLI tool.

Stop guessing Spark configs. Let data and AI tell you what's wrong.

Install

pip install spark-advisor-cli

Quick Start

# Analyze from event log file (rules-only, free)
spark-advisor analyze /path/to/event-log.json.gz --no-ai

# Analyze with AI recommendations
export ANTHROPIC_API_KEY=sk-ant-...
spark-advisor analyze /path/to/event-log.json.gz

# Analyze from History Server
spark-advisor analyze app-20250101120000-0001 -hs http://yarn:18080

# Agent mode (multi-turn AI analysis)
spark-advisor analyze /path/to/event-log.json.gz --agent

Commands

analyze — analyze a Spark job

# From event log file
spark-advisor analyze /path/to/event-log.json.gz

# From History Server
spark-advisor analyze app-20250101120000-0001 -hs http://yarn:18080

# With AI analysis (default if ANTHROPIC_API_KEY is set)
spark-advisor analyze /path/to/event-log.json.gz

# Without AI (rules only)
spark-advisor analyze /path/to/event-log.json.gz --no-ai

# Agent mode (multi-turn AI with tool use)
spark-advisor analyze /path/to/event-log.json.gz --agent

# Verbose mode (per-stage breakdown)
spark-advisor analyze /path/to/event-log.json.gz --verbose

# JSON output
spark-advisor analyze /path/to/event-log.json.gz --format json

# Save suggested config to file
spark-advisor analyze /path/to/event-log.json.gz -o spark-defaults.conf

# Use specific Claude model
spark-advisor analyze /path/to/event-log.json.gz --model claude-sonnet-4-6
Flag Short Default Description
source required App ID (with -hs) or path to event log file
--history-server -hs None Spark History Server URL
--no-ai False Disable AI analysis (rules only)
--agent False Use agent mode (multi-turn AI with tool use)
--model -m claude-sonnet-4-6 Claude model for AI analysis
--output -o None Write suggested config to file
--format -f text Output format: text or json
--verbose -v False Show per-stage breakdown

scan — list recent jobs from History Server

spark-advisor scan -hs http://yarn:18080 --limit 20

version

spark-advisor version
# spark-advisor v0.1.10

What it detects

11 deterministic rules: data skew, disk spill, GC pressure, shuffle partitions, executor idle, task failures, small files, broadcast join threshold, serializer choice, dynamic allocation, memory overhead.

All thresholds are configurable via Thresholds model.

See also

License

Apache 2.0

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

spark_advisor_cli-0.1.16.tar.gz (12.5 kB view details)

Uploaded Source

Built Distribution

If you're not sure about the file name format, learn more about wheel file names.

spark_advisor_cli-0.1.16-py3-none-any.whl (9.3 kB view details)

Uploaded Python 3

File details

Details for the file spark_advisor_cli-0.1.16.tar.gz.

File metadata

  • Download URL: spark_advisor_cli-0.1.16.tar.gz
  • Upload date:
  • Size: 12.5 kB
  • Tags: Source
  • Uploaded using Trusted Publishing? No
  • Uploaded via: uv/0.10.12 {"installer":{"name":"uv","version":"0.10.12","subcommand":["publish"]},"python":null,"implementation":{"name":null,"version":null},"distro":{"name":"Ubuntu","version":"24.04","id":"noble","libc":null},"system":{"name":null,"release":null},"cpu":null,"openssl_version":null,"setuptools_version":null,"rustc_version":null,"ci":true}

File hashes

Hashes for spark_advisor_cli-0.1.16.tar.gz
Algorithm Hash digest
SHA256 00c5b2b31b9e9517cd4058ac00f9d1bce2c56a3a260363026181445b30bd7542
MD5 3789b07302a894cd29e43bac5356bcf7
BLAKE2b-256 2bdf941a88aea1e8e2bd281db441c2bd993426c45b74c7800d169d9d0eb91ee4

See more details on using hashes here.

File details

Details for the file spark_advisor_cli-0.1.16-py3-none-any.whl.

File metadata

  • Download URL: spark_advisor_cli-0.1.16-py3-none-any.whl
  • Upload date:
  • Size: 9.3 kB
  • Tags: Python 3
  • Uploaded using Trusted Publishing? No
  • Uploaded via: uv/0.10.12 {"installer":{"name":"uv","version":"0.10.12","subcommand":["publish"]},"python":null,"implementation":{"name":null,"version":null},"distro":{"name":"Ubuntu","version":"24.04","id":"noble","libc":null},"system":{"name":null,"release":null},"cpu":null,"openssl_version":null,"setuptools_version":null,"rustc_version":null,"ci":true}

File hashes

Hashes for spark_advisor_cli-0.1.16-py3-none-any.whl
Algorithm Hash digest
SHA256 f88a351ade6c480964ec80b87178275ca509b9712674389bd1fdedb122b86f49
MD5 92e734d82196a71901090546feb90662
BLAKE2b-256 40b7ee03f53fc342be4013f7057e66d941546dd1130da7f6e5abc8b4ea41099c

See more details on using hashes here.

Supported by

AWS Cloud computing and Security Sponsor Datadog Monitoring Depot Continuous Integration Fastly CDN Google Download Analytics Pingdom Monitoring Sentry Error logging StatusPage Status page