Unverified50% confidenceFactExact time
At data warehouse scale, re-embedding billions of rows of data is a long and expensive batch job, so mixed-version indexes are not an edge case but rather the default state during every model migration.
1
Sources
50%
Confidence
Long-term
Relevance
10/12/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedThe full rebuild approach is simple to implement and leaves no dirty data, but is costly at the million-vector scale and cannot meet near-real-time update requirements, making it suitable for scenarios where data changes infrequently.74% similarUnverifiedInference serving cost scales with query volume and is predictable and elastically scalable; re-embedding cost scales with the total historical data volume and only increases, and for a warehouse holding years of data, the backfill cost may far exceed the ongoing inference bill.73% similarUnverifiedSwitching embedding models means all historical data must have its vectors recomputed, resulting in extremely high migration costs.73% similarUnverifiedThe input scale of backfilling equals the full historical data, and its cost increases monotonically with the total data volume and cannot be compressed in total computation by adding concurrency, so in practice backfilling is difficult to complete within a few hours and inevitably overlaps with online queries.73% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1007794API
curl https://kongchang.com/api/v1/knowledge/claims/1007794MCP
get_claim(id=1007794)