πŸ’¬
βœ‰οΈ
πŸ“Έ
▢️

Loading…

Choosing a database: SQL, documents, graphs, and search

Database Worlds
Database Worlds

Graph Databases

When relationships matter more than the data itself

UserFriendPostGroupLikenodes + edges Β· 2-hop traversal in <10ms

Every relational database stores 'User A follows User B' as a row in a separate follow_table. To find who @alice's second-degree connections are, SQL needs to JOIN three tables twice. A graph database stores that relationship as a physical edge directly connecting node A to node B. Traversing 6 levels of connections takes milliseconds instead of minutes.

Graph databases were invented because relational schemas become catastrophically slow when relationships are the data. Social networks, fraud rings, recommendation engines, knowledge graphs β€” these domains are fundamentally about traversing connections. Every hop in SQL requires a JOIN. Every hop in a graph database is a direct pointer follow. At 4+ degrees of separation, the performance gap becomes 100x to 10,000x.

Neo4j

The original and most widely used graph database

Mental model

Neo4j stores your data the way a whiteboard diagram looks β€” nodes connected by labeled, directional relationships. A user node connects to a product node with a PURCHASED relationship. That product connects to categories, tags, and similar products. Querying 'find products purchased by people who also bought X' is a single pattern match β€” not five JOINs.

Why it exists

Founded in 2007 when its creators were building a content management system in Java and realized that navigating deeply nested relationships in relational databases required enormous JOIN queries that became slower as the data grew. They needed a database where traversal cost was independent of total data size.

Shines at

  • Social network friend-of-friend queries
  • Fraud detection via relationship patterns
  • Product recommendation engines
  • Knowledge graphs and ontologies
  • Access control and permission hierarchies

Breaks down when

  • Simple key-value lookups (overkill)
  • Extremely high write throughput
  • Large-scale analytics over all nodes
  • Non-graph data models
Performance note: Relationship traversal is O(1) per hop regardless of database size. Cypher query language is expressive and optimized for graph patterns.

Real-world use

NASA uses Neo4j for spacecraft incident knowledge graphs. eBay uses it for their shopping graph. LinkedIn started their social graph on Neo4j before building their own.

Amazon Neptune

Managed graph database supporting property graphs and RDF

Mental model

Amazon Neptune is Neo4j's cloud-native managed counterpart. Instead of running and maintaining graph database servers, you provision a Neptune cluster and AWS handles replication, backups, and failover. It speaks both Gremlin (property graph) and SPARQL (RDF), making it flexible for different graph models.

Why it exists

AWS recognized that their customers were building graph-heavy applications (fraud detection, recommendation engines, compliance knowledge graphs) and wanted a managed solution that integrated with their existing AWS infrastructure without managing graph database servers.

Shines at

  • AWS-native graph applications
  • Compliance and regulatory knowledge graphs
  • Identity graph for fraud detection
  • Recommendation engines within AWS ecosystem

Breaks down when

  • Query flexibility vs Neo4j on complex traversals
  • Vendor lock-in to AWS
  • Cost at very large graph sizes
Performance note: Replicates across 3 AZs automatically. Supports both TinkerPop/Gremlin and SPARQL query languages.

Real-world use

Samsung uses Neptune for their knowledge graph. Siemens uses it for industrial compliance tracking.

Your Path

Database
└─
Direct Jump
└─
Graph DB

Key Traits

βœ“ O(1) per hop traversalβœ“ Native relationshipsβœ“ Pattern matchingβœ— Poor bulk analytics

Pick When

  • Social graph / friend-of-friend queries
  • Fraud detection via relationship patterns
  • Knowledge graph / recommendation paths

Used By

Neo4j: LinkedIn, eBayNeptune: Airbnb, Intuit

Core Tradeoff

Relationship traversal speed β€” weak outside graph queries

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 10–100Γ— 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 5–8 tables per request. No need to spread writes across multiple servers.

RDBMS

NewSQL’s 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 9–14 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 15–50ms. 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 15–40ms. 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.