SolrCloud and NoSQL
summary
Argues that SolrCloud is CP (favours consistency and partition tolerance) and uses ZooKeeper for cluster state, describes optimistic concurrency via client-obtained version numbers and transaction log durable writes for real-time gets, shows distributed indexing with leaders/replicas and a split‑shard feature (4.3.1) that splits shards without full reindex, contrasts Solr/Lucene as a searchable datastore vs MongoDB/Cassandra which usually hook an external search engine, and notes no built‑in streaming/river—use DIH or custom plugins.
AI-written summary · about
chapters (16)
- 0:00Start
- 1:18Big Data
- 3:02DB Rankings: Overall
- 3:51Search Engine Rankings
- 4:14MongoDB
- 5:23Cassandra
- 6:29The story so far...
- 7:36Adding NoSQL to Search
- 9:41SolrCloud Design Goals
- 14:32Distributed Indexing
- 16:02Optimistic Concurrency
- 18:11Document Routing
- 19:49Durable Writes
- 21:16Collections API
- 22:11Solr 4.3: Seamless Online Shard Splitting updates
- 28:28Bangalore Apache Lucene/Solr Meetup
related talks
33mHighly available and disaster ready Apache Solr
01h 01mNoSQL: An Analysis
08mSolr Cloud Manager
57mMulti Tenant NoSQL and NewSQL Cloud data Design Patterns Deep dive
29mModern Database Architecture: From SQL to NoSQL & Distributed SQL
57msolr-ruby: The Best Open Source Search Engine + Ruby
38mNot all data is created equal: NoSQL
34mData Storage: NoSQL Toasters and a Cloud of Kitchen Sinks