Introduction
Solr 9.x brings significant improvements including vector search, enhanced security, and better cloud support. This article explores the latest features and ecosystem evolution.
Vector Search
k-NN Search
// Enable vector search
{
"name": "embedding",
"type": "knn_vector",
"dimension": 384,
"method": {
"name": "hnsw",
"class": "solr.HnswVectorStrategy",
"engine": "lucene",
"M": 16,
"efConstruction": 200
}
}
// Query vector
{
"query": "{!knn topK=10}embedding:[0.1, 0.2, ...]"
}
Hybrid Search
// Combine keyword and vector
{
"query": "(title:search OR {!knn topK=10}embedding:[0.1, 0.2])"
}
Security Enhancements
Authentication
// PKI authentication
{
"authentication": {
"class": "solr.AuthenticationPlugin",
"blockUnknown": true
}
}
RBAC
// Role-based access
{
"authorization": {
"class": "solr.RuleBasedAuthorizationPlugin",
"permissions": [
{"name": "read", "role": "reader"},
{"name": "update", "role": "editor"}
]
}
}
SolrCloud Improvements
Auto-scaling
# Trigger auto-scaling
curl "http://localhost:8983/solr/admin/collections?action=CREATESHARD&shard=shard1&collection=products"
Distributed Tracing
// Enable tracing
{
"tracing": {
"enabled": true,
"sampler": "probability",
"param": 0.01
}
}
Performance Improvements
Faster Indexing
// Concurrent merge scheduling
{
"mergeScheduler": "solr.ConcurrentMergeScheduler"
}
Improved Caching
// Better cache eviction
{
"cache": {
"autowarmCount": "100%"
}
}
Conclusion
Solr 9.x brings vector search, enhanced security, and better cloud support. The platform continues to evolve for modern search requirements.
Comments