ECO-AIM-RAG-001¶
Name: Embedding regeneration without change detection
Category: AI/ML
Family: Retrieval-Augmented Generation
Primary layer: ai
System layers: ai
Description¶
Embeddings are regenerated for unchanged content, wasting compute and increasing pipeline latency.
Impact¶
- type: ai-compute
- confidence: 0.75
- notes: Added as part of the 0.3.0 expansion to capture cross-system sustainability and operational waste.
Detection¶
- method: static-or-runtime
- confidence: 0.55
- runtime_validation_required: Yes
Remediation¶
- guidance: Use content hashing, incremental indexing, and idempotent embedding pipelines.
- tradeoffs: May require architecture, product, or operations review rather than a local code change.
Cost Dimensions¶
- compute: high
- memory: medium
- network: medium
- storage: medium
- human_time: low
- carbon: high
- water: medium
Amplification¶
- scales_with_users: No
- scales_with_data_volume: Yes
- scales_non_linearly: No
Temporal Behavior¶
- startup_only: No
- steady_state: Yes
- burst_sensitive: Yes
- time_degradation: No
Runtime Evidence¶
- embedding job logs
- content hashes
- vector DB writes
Pattern examples¶
No pattern examples provided.
Remediation examples¶
No remediation examples provided.
Metadata¶
- catalog_version: 0.4.0
- status: draft
- source: catalog expansion recommendations applied 2026-05-21