{"id":34686140,"url":"https://github.com/bieli/big-data-architecture-advisor","last_synced_at":"2026-05-22T20:33:16.035Z","repository":{"id":329031272,"uuid":"1117848894","full_name":"bieli/big-data-architecture-advisor","owner":"bieli","description":"Kind of \"expert system\" for BigData architects roles ;-) For fun \u0026 learning!","archived":false,"fork":false,"pushed_at":"2025-12-16T23:45:51.000Z","size":20,"stargazers_count":0,"open_issues_count":0,"forks_count":0,"subscribers_count":0,"default_branch":"main","last_synced_at":"2025-12-20T13:23:20.998Z","etag":null,"topics":["advisor","architecture","expert-system","rust"],"latest_commit_sha":null,"homepage":"","language":"Rust","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":"bsd-3-clause","status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/bieli.png","metadata":{"files":{"readme":"README.md","changelog":null,"contributing":null,"funding":null,"license":"LICENSE","code_of_conduct":null,"threat_model":null,"audit":null,"citation":null,"codeowners":null,"security":null,"support":null,"governance":null,"roadmap":null,"authors":null,"dei":null,"publiccode":null,"codemeta":null,"zenodo":null,"notice":null,"maintainers":null,"copyright":null,"agents":null,"dco":null,"cla":null}},"created_at":"2025-12-16T22:49:54.000Z","updated_at":"2025-12-16T23:45:54.000Z","dependencies_parsed_at":null,"dependency_job_id":null,"html_url":"https://github.com/bieli/big-data-architecture-advisor","commit_stats":null,"previous_names":["bieli/big-data-architecture-advisor"],"tags_count":null,"template":false,"template_full_name":null,"purl":"pkg:github/bieli/big-data-architecture-advisor","repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/bieli%2Fbig-data-architecture-advisor","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/bieli%2Fbig-data-architecture-advisor/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/bieli%2Fbig-data-architecture-advisor/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/bieli%2Fbig-data-architecture-advisor/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/bieli","download_url":"https://codeload.github.com/bieli/big-data-architecture-advisor/tar.gz/refs/heads/main","sbom_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/bieli%2Fbig-data-architecture-advisor/sbom","scorecard":null,"host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":286080680,"owners_count":33366521,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2026-05-21T12:23:38.849Z","status":"online","status_checked_at":"2026-05-22T02:00:06.671Z","response_time":265,"last_error":null,"robots_txt_status":"success","robots_txt_updated_at":"2025-07-24T06:49:26.215Z","robots_txt_url":"https://github.com/robots.txt","online":true,"can_crawl_api":true,"host_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub","repositories_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories","repository_names_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repository_names","owners_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners"}},"keywords":["advisor","architecture","expert-system","rust"],"created_at":"2025-12-24T21:47:43.458Z","updated_at":"2026-05-22T20:33:16.030Z","avatar_url":"https://github.com/bieli.png","language":"Rust","funding_links":[],"categories":[],"sub_categories":[],"readme":"# big-data-architecture-advisor\nKind of \"expert system\" for BigData architects roles ;-) For fun \u0026amp; learning!\n\n## Motivation\nModern data systems are complex, distributed, and often difficult to design. Architects must balance multiple factors such as **data volume, workload type, SLA requirements, budget constraints, and observability needs**. Traditional decision-making is often ad hoc and opaque, making it hard to justify choices or reproduce them later.\n\nThis project was motivated by the need for a **transparent, rule-based expert assistant** that:\n- Encodes architectural knowledge as declarative rules.\n- Produces clear recommendations with explanations.\n- Supports multiple criteria, including observability, cost, and compliance.\n- Allows dynamic weighting of priorities (e.g., SLA vs. budget vs. observability).\n- Provides reproducible, explainable decisions for engineers and architects.\n\n## How It Works?\nThe advisor is implemented in **Rust** using the [`crepe`](https://crates.io/crates/crepe) Datalog engine. It works as follows:\n\n```rust\ncrepe! {\n    @input\n    struct Volume(\u0026'static str);\n    @input\n    struct Workload(\u0026'static str);\n    @input\n    struct SLA(\u0026'static str);\n    @input\n    struct Budget(\u0026'static str);\n    @input\n    struct Observability(\u0026'static str);\n\n    @output\n    struct Recommendation(\u0026'static str, i32);\n    @output\n    struct Explanation(\u0026'static str);\n\n...\n```\n\n### Important points\n1. **Inputs**: The user provides facts about their scenario:\n   - Data Volume: `small`, `medium`, `large`\n   - Workload Type: `batch`, `streaming`, `mixed`\n   - SLA Requirement: `low`, `high`\n   - Budget: `low`, `high`\n   - Observability: `low`, `medium`, `high`\n\n2. **Rules**: The system encodes architectural knowledge as declarative rules. Examples:\n   - Large batch + low SLA + low budget → Hadoop/Spark\n   - Streaming + high SLA → Kafka/Flink\n   - Large mixed + high SLA + high budget → Lakehouse\n   - High observability → Prometheus/Grafana, OpenTelemetry/Jaeger, lineage tools\n\n3. **Scoring**: Each recommendation has a base score. The user can assign **weights** to criteria (SLA, budget, volume, workload, observability). Final scores are adjusted dynamically based on these weights.\n\n4. **CLI Interface**: The program runs interactively:\n   - Prompts the user for inputs.\n   - Prompts for weights (or uses defaults).\n   - Produces ranked recommendations.\n   - By default shows the top suggestion; with `--multi` flag shows all ranked suggestions.\n   - With `--history` flag, prints past saved decisions.\n\n\n## Multi Mode (`--multi`)\n\nBy default, the advisor uses **strict rule-based matching**: only architectures that exactly fit the provided inputs (volume, workload, SLA, budget, observability) are recommended.\n\nWith the `--multi` flag, the advisor switches to **fuzzy similarity scoring**:\n\n- Every architecture profile is evaluated against the user’s inputs.\n- Exact matches contribute full points; `\"any\"` or partial matches contribute partial points.\n- User-defined weights (SLA, budget, volume, workload, observability) influence the scoring.\n- All architectures are ranked and displayed, not just the top match.\n\nThis mode is useful when:\n- You want to explore **alternative architectures** beyond the strict best fit.\n- You need to see **trade-offs** between SLA, cost, observability, and workload types.\n- You’re experimenting with different weights to understand how priorities shift recommendations.\n\n\n## Example Run\n\n### Strict onr recommendation mode\n\n```bash\n$ cargo run\n=== Big Data Architecture Advisor CLI ===\n*** STRICT Rule-Based Mode (using Crepe library) ***\n\nEnter data volume (small/medium/large):\nmedium\nEnter workload type (batch/streaming/mixed):\nmixed\nEnter SLA requirement (low/high):\nlow\nEnter budget (low/high):\nhigh\nEnter observability requirement (low/medium/high):\nmedium\nEnter SLA weight (default 5):\n2\nEnter Budget weight (default 3):\n3\nEnter Volume weight (default 2):\n1\nEnter Workload weight (default 1):\n1\nEnter Observability weight (default 4):\n4\n\n=== Recommendations ===\n-\u003e Cloud-native Monitoring (CloudWatch/Stackdriver/Azure Monitor) (score: 82)\n\n=== Explanations ===\n- Cloud-native monitoring chosen for integrated observability in cloud ecosystems.\n```\n\n### Multi recommendations mode\n\n```bash\n$ cargo run -- --multi\n=== Big Data Architecture Advisor CLI ===\n*** MULTI Recommendations Mode ***\n\nEnter data volume (small/medium/large):\nlarge\nEnter workload type (batch/streaming/mixed):\nmixed\nEnter SLA requirement (low/high):\nlow\nEnter budget (low/high):\nhigh\nEnter observability requirement (low/medium/high):\nmedium\nEnter SLA weight (default 5):\n3\nEnter Budget weight (default 3):\n2\nEnter Volume weight (default 2):\n1\nEnter Workload weight (default 1):\n1\nEnter Observability weight (default 4):\n3\n\n=== Recommendations ===\n-\u003e Cloud-native Monitoring (CloudWatch/Stackdriver/Azure Monitor) (score: 150)\n-\u003e Lakehouse (score: 135)\n-\u003e ELK Stack (Elasticsearch/Logstash/Kibana) (score: 125)\n-\u003e ML Platform (Kubeflow/MLflow) (score: 123)\n-\u003e Data Mesh (score: 120)\n-\u003e Monitoring Stack (Prometheus/Grafana) (score: 120)\n-\u003e Data Lineage Tools (OpenLineage/Marquez) (score: 120)\n-\u003e Distributed Tracing (OpenTelemetry/Jaeger) (score: 118)\n-\u003e Hybrid Cloud (score: 110)\n-\u003e Orchestration (Airflow/Prefect) (score: 110)\n-\u003e Alerting \u0026 Incident Response (PagerDuty/OpsGenie) (score: 110)\n-\u003e Data Warehouse (score: 110)\n-\u003e Kafka/Flink (score: 105)\n-\u003e Data Lake (S3/ADLS) (score: 105)\n-\u003e FinOps Dashboards (score: 103)\n-\u003e Hadoop/Spark (score: 100)\n-\u003e Governance Layer (Collibra/Alation) (score: 95)\n-\u003e ETL Pipelines (score: 85)\n```\n\n## TODO list\n- [ ] Add --clear-history flag.\n- [ ] Extend rules for serverless, data fabric, AI pipelines.\n- [ ] Add YAML/JSON config for batch runs.\n- [ ] Integrate with web UI for interactive decision support.\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fbieli%2Fbig-data-architecture-advisor","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fbieli%2Fbig-data-architecture-advisor","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fbieli%2Fbig-data-architecture-advisor/lists"}