AI's infrastructure boom is standing on porcelain legs
Reported by TechRadar1 source ↓
AI agents are exposing the costs and delays created when businesses use separate systems for search and analytics. A single request can trigger multiple queries and hand-offs across those systems.
Key points
- Separate search and analytics systems require data copies, synchronization pipelines and maintenance
- An AI agent can turn one user request into several database operations
- Hand-offs between systems can add latency, cost and risk of context loss
- Open data formats can let search and analytics tools use the same underlying data
Details
TechRadar reports that search systems are built to find relevant material in documents and other unstructured data, while analytical databases work with structured data to aggregate values, join records and identify trends. A request that needs both can require one system to retrieve information and another to calculate from it.
Businesses often connect these systems with data copies, extraction and transformation pipelines, synchronization jobs and application code. Some also use a separate vector-search layer. The article says these arrangements create storage and engineering costs, and can leave data out of date or slow queries.
TechRadar describes storing data in object storage, such as Amazon S3, in open formats such as Apache Parquet and Apache Iceberg as one way to reduce duplication. Search and analytics tools can use the same underlying data, though indexes and computing layers are still needed and some workloads may require specialized systems.
Written from coverage as of
Images
- TechRadar
Stocks in this story
Change since open: 20 Nov to 21 Nov 2026
Last refresh