Choosing a database: SQL, documents, graphs, and search
Search Engines (Database)
Pre-indexed for text β making impossible queries trivially fast
Search engines pre-build an inverted index β a lookup table mapping every word to every document containing it. When you search 'red running shoes', it doesn't scan 50 million products. It looks up 'red', 'running', and 'shoes' in the index and intersects the results in under 50ms. A SQL LIKE '%red running shoes%' would take minutes on the same dataset.
SQL's pattern matching was never designed for text search. LIKE queries scan every row. There's no concept of relevance ranking, fuzzy matching, or faceted filtering. Search engines solve this by pre-processing every field at write time β building indexes, normalizing text, computing term frequencies β so queries that would be impossible in SQL complete in milliseconds.
Elasticsearch
The industry-standard distributed search and analytics engine
Elasticsearch
The industry-standard distributed search and analytics engine
Mental model
Elasticsearch is like a search engine you can embed in your product. It ingests your data, breaks every text field into indexed tokens, and lets users search across millions of records with typo tolerance, relevance ranking, and instant faceting (filter by category: 50,000 results shown immediately).
Why it exists
Shay Banon built Elasticsearch in 2010 because he was writing a recipe search engine for his wife and Lucene (the underlying library) was impossible to run in a distributed way. He wrapped Lucene in a REST API with automatic clustering β making distributed text search accessible without a PhD in information retrieval.
Shines at
- Full-text search with relevance ranking
- Log analytics (the ELK stack)
- E-commerce product search with faceting
- Autocomplete and type-ahead suggestions
- Application monitoring and observability
Breaks down when
- Primary transactional data store
- Strict ACID requirements
- Cost at very large indexes (memory-heavy)
- Real-time writes that need immediate search visibility
Real-world use
Wikipedia powers its search with Elasticsearch. Netflix searches 100B+ log events with it. GitHub uses it for code search.
Typesense
Instant, typo-tolerant search built for developers
Typesense
Instant, typo-tolerant search built for developers
Mental model
Typesense is designed to feel instant and just work. Drop in your data, point your search bar at it, and get millisecond results with typo tolerance out of the box. Where Elasticsearch takes weeks to tune, Typesense is production-ready in hours with sane defaults engineers actually want.
Why it exists
Built by engineers who were frustrated that Elasticsearch required a dedicated search engineer to tune, provision, and maintain. Typesense is open-source, easy to self-host, RAM-first for speed, and designed with developer experience as a first-class goal.
Shines at
- E-commerce and catalog search for smaller/mid teams
- Documentation and knowledge base search
- Instant search (sub-10ms responses)
- Teams wanting search without Elasticsearch complexity
Breaks down when
- Billion-document scale (Elasticsearch is better)
- Complex log analytics pipelines
- Faceted aggregations on enormous datasets
Real-world use
Algolia (competitor), PostHog, and thousands of developer tools use Typesense for their search experience.
Your Path
Key Traits
Pick When
- Full-text search over large corpus
- E-commerce product search with filters
- Log analytics / observability
Used By
Core Tradeoff
Instant full-text search β index rebuild cost on writes
Browse database families
Decision explanations
What operational model best fits this workload?
A platform records thousands of financial operations per minute. Each operation touches multiple tables β deducting from one account, crediting another, updating a balance ledger. Partial writes must never succeed. Queries need to join across user, account, and transaction tables in real time.
Transactional (OLTP)
OLTP = individual, frequent writes where every single transaction must be correct. OLAP is the opposite β it scans huge amounts of historical data to answer questions like 'what was total revenue last quarter?' When every write matters, you need OLTP.
Analytical (OLAP): OLAP databases are built for scanning large amounts of old data all at once (like monthly reports). They trade away per-row correctness for bulk scan speed β terrible for thousands of individual write operations per second.
What storage architecture optimises this workload?
You run nightly BI reports: total revenue by region, cohort retention, anomaly detection. Queries scan hundreds of millions of rows but nothing is ever updated after it's written. Tables have 80+ columns per row but queries only need a few at a time.
Data Warehouse (Columnar)
Columnar storage is a direct response to analytical query patterns. When you query 2 out of 80 columns across 500M rows, a row-store reads all 80 columns for every row β wasting 97% of disk I/O. A columnar store reads only those 2 columns, compresses them heavily, and aggregates them in a single scan pass.
RDBMS (Row Store): Row-oriented stores are fast at fetching complete records β but analytical queries scan billions of partial rows. An RDBMS touching 500M rows would deserialise all 80 columns per row even when you only need 2 β wasting enormous I/O bandwidth and making queries 10β100Γ slower than a columnar store.
Which transactional database family fits this workload?
Your app handles 8,000 transactions per second in a single region. Schema is stable. Queries join across 5β8 tables per request. No need to spread writes across multiple servers.
RDBMS
NewSQLβs distributed write scaling adds operational complexity only justified when you need to shard writes across regions β typically at 100k+ TPS or for global multi-master deployments. This workload fits well within primary-replica RDBMS capacity, making NewSQL overhead unnecessary.
NewSQL: NewSQL solves one problem: writing to multiple servers across regions simultaneously. Great at Google scale. But it adds latency overhead per write and is complex to operate. If one server handles your load, RDBMS is the simpler, faster choice.
What data organisation model fits a workload where every entity carries a different field set?
Your product catalog has 200M items. Electronics have wattage. Clothes have size and material. Furniture has dimensions. No two categories share the same fields. 40% of queries fetch a full item. 40% filter by attribute combinations. Schema changes every time a new category launches.
Document-Oriented
Document databases let each item carry exactly its own fields β no null columns for attributes a product doesn't have. You can add an index on any field and query it directly. Key-value stores can only look up by exact key β you'd need to build an entire query engine on top just to filter by attributes.
Key-Driven Access: Key-value = 'give me the item with key #12345'. That's the only query it supports. No 'show me all electronics under $100', no filtering by attributes. For a query-heavy product catalog you'd have to pre-compute every possible filter combination yourself β that's the application's job, not the database.
As multi-hop relationship traversal becomes the dominant query pattern, what architectural response is correct?
Your social app runs on MongoDB. A new feature needs: 'find friends-of-friends who bought the same product the user did' β that's 4 levels of relationships. These queries take 9β14 seconds and pin the CPU. The deeper the social graph grows, the worse it gets.
Graph Database
Graph databases store relationships as physical pointers. Each hop is O(1) β instant, no matter how big the database is. In MongoDB, every hop is another aggregation step that multiplies the work. At 4 hops on 10M users, that's billions of operations. Graph databases solve this in milliseconds.
Denormalised Documents: Embedding relationships in arrays works for simple things like 'a post's top 5 comments'. Social graphs are different β they're unbounded. A user's 3-hop network could be millions of people. Storing that as nested arrays uses gigabytes per user and the query is still slow. It doesn't fix the fundamental problem.
For this read-dominated, key-exact workload, which architecture serves the latency and access pattern?
A ride-hailing app tracks live GPS for 3M drivers (200k writes/sec, 1M reads/sec, must respond in under 2ms). Also serves surge pricing per city zone and manages session data per active ride. All lookups are exact key lookups β no filtering or joins needed.
In-Memory Key-Value Store
The bottleneck is read speed: under 2ms. In-memory key-value stores (like Redis) keep all data in RAM β reads are microseconds, not milliseconds. Column-family databases write to disk, so reads take 15β50ms. When you need blazing-fast reads, you need data in memory.
Distributed Column-Family: Cassandra excels at absorbing massive write storms β IoT sensors writing millions of events per second. The trade-off: reads hit disk, taking 15β40ms. Perfect for 'store everything and query later' data. Terrible for a live driver map that needs to respond in under 2ms.
When time is the organising dimension and every query is a time range, what database family is architecturally aligned?
80,000 connected vehicles each send 400 sensor readings per second (speed, battery, brakes, GPS...) = 32M data points per second. You need: 'plot battery over the last 4 hours for vehicle X', 'alert if brake pressure drops', 'fleet average fuel consumption over 7 days'. No text search. Every query is a time window.
Time-Series Database
Time-series databases are built exactly for sequential numeric measurements. They compress repeated similar values efficiently (battery at 80%, 81%, 80%... is stored as tiny deltas) and run time-window queries natively. Search engines index words β vehicle sensor data has no words. You'd pay enormous memory cost for indexes that are completely useless here.
Search / Log Analytics Engine: Search engines shine for text: log messages, error descriptions, user queries. They build word indexes. But vehicle telemetry is pure numbers β speed: 62, battery: 77%, brake: 94%. There's nothing to text-search. You'd waste memory on word indexes and miss the compression and time-range query features that time-series gives you.
