Search & Caching
Shio CMS provides automatic content indexing with full-text search capabilities, and a distributed caching layer powered by Hazelcast for high-performance website delivery.
Full-Text Search
Automatic Indexing
Content in Shio CMS is automatically indexed when created or updated. This means your site has search functionality out of the box — no additional configuration required.
Built-in Search
The embedded search engine provides:
- Full-text search across all content
- Search API endpoint (
/api/v2/search) - Site-scoped search results
Viglet Turing ES Integration
For advanced search capabilities, Shio CMS integrates with Viglet Turing ES — an enterprise search platform with semantic navigation and generative AI.
What Turing ES adds
| Feature | Description |
|---|---|
| Faceted Search | Filter results by category, date, author, or any attribute |
| Autocomplete | Type-ahead suggestions as users type |
| Semantic Navigation | Configurable search experience with result ranking |
| Generative AI | RAG-powered AI answers grounded in your content |
| Spotlights | Curated results pinned to specific search terms |
Search Field Mapping
When configuring Post Type fields, you can map each field to a Turing Semantic Navigation field:
-
Search Field Association — map to a default Turing SN field:
- Title, Description, Text, Date, URL, Image
-
Create Additional Search Field — map to a custom field for facets or advanced filtering
Turing Search Endpoint
Shio CMS exposes a Turing search context at:
/__tur/sn/{siteName}
This endpoint proxies search requests to the configured Turing ES instance, allowing your site to use Turing's full search capabilities without direct client-server communication.
Configuration
Search providers are configured in the Configuration Console > Search Providers section. Define:
- Turing ES server URL
- Search site name
- Authentication credentials
Hazelcast Cache
Shio CMS uses Hazelcast as a distributed caching layer for website rendering. The cache dramatically improves response times by storing rendered page HTML and content objects.
How it works
Cache Behavior
| Aspect | Details |
|---|---|
| Scope | Page-level and object-level caching |
| TTL | Default 24 hours (configurable per region) |
| Invalidation | Automatic when content is created, updated, or deleted |
| Distribution | Hazelcast distributes cache across multiple Shio CMS nodes |
| Disable | Set TTL to 0 on a region to disable caching for that region |
Cache Configuration
Hazelcast is configured via hazelcast.xml in the application resources. Key settings:
- Cluster name — identifies the Hazelcast cluster
- Network — multicast or TCP/IP discovery for multi-node setups
- Map configuration — TTL, max size, and eviction policies per cache map
Multi-Node Caching
In a multi-node deployment, Hazelcast automatically discovers other nodes and distributes the cache. When content is updated on one node, the cache is invalidated across all nodes.
Related Pages
| Page | Description |
|---|---|
| Content Modeling | Search field mapping in Post Types |
| Website Development | Region caching and TTL configuration |
| Configuration Reference | Application properties |
| Architecture Overview | System components and deployment topologies |