RedisConf17 - Redis Graph
0 likes778 views
This document discusses Redis Graph, an open source graph database module for Redis. It provides concise summaries of Redis Graph's key features: - Redis Graph allows storing and querying graph-structured data using a graph data model with nodes, edges, and properties. It supports multi-hop queries across relationships. - Redis Labs provides commercial support for Redis Graph through products like Redis Enterprise Cloud and on-premise offerings. - The document demonstrates basic graph queries in Cypher and shows the query execution plan, highlighting Redis Graph's ability to efficiently query large graph datasets.
![MATCH (Michael:employee {name:michael
scott})-[boss]->(E:employee)-[visit]->(c:country)
WHERE (E.age > Michael.age AND c.continent = Europe)
RETURN E.name, count(c.name) AS countriesVisited
ORDER BY countriesVisited, E.age DESC
LIMIT 5
Query language Cypher](https://image.slidesharecdn.com/redisconfredisgraph-170619223813/85/RedisConf17-Redis-Graph-18-320.jpg)
![Execution plan
MATCH (Michael:employee {name:michael
scott})-[boss]->(E:employee)
WHERE E.age > Michael.age
RETURN E.name
ORDER BY E.age DESC
LIMIT 5](https://image.slidesharecdn.com/redisconfredisgraph-170619223813/85/RedisConf17-Redis-Graph-23-320.jpg)
![FilterExpand
Search hexastore for: SPO:Michael:Boss
MATCH (Michael:employee {name:michael
scott})-[boss]->(E:employee)](https://image.slidesharecdn.com/redisconfredisgraph-170619223813/85/RedisConf17-Redis-Graph-33-320.jpg)
![Features
● Multi hop, multi entry point
(A)-[R1]->(C)<-[R2]-(B)
● Aggregations, Group bys
RETURN E.gender, AVG(E.age) AS average_age
● Order bys, Distinct](https://image.slidesharecdn.com/redisconfredisgraph-170619223813/85/RedisConf17-Redis-Graph-40-320.jpg)
RedisConf17 - Redis Graph
- 1. Redis Graph Itamar Haber in lieu of Roi Lipman
- 2. Redis Labs – Home of Redis • Founded in 2011 • HQ in Mountain View CA, R&D center in Tel-Aviv IL The commercial company behind Open Source Redis Provider of the Redis Enterprise (Redise ) technology, platform and products
- 3. Redise Cloud Private Redis Labs Products Redise Cloud Redise Pack ManagedRedise Pack SERVICES SOFTWARE Fully managed Redise service in VPCs within AWS, MS Azure, GCP & IBM Softlayer Fully managed Redise service on hosted servers within AWS, MS Azure, GCP, IBM Softlayer, Heroku, CF & OpenShift Downloadable Redise software for any enterprise datacenter or cloud environment Fully managed Redise Pack in private data centers && &
- 6. Demo demo
- 7. Adding nodes GRAPH.CREATENODE <Graph_Name> <Label> <Attributes> GRAPH.CREATENODE the_office employee name "michael scott" age 53 gender male status single
- 8. 1. Give node a unique ID 2. Store node in “All node” store 3. If optional Label given, store Stores ALL employees
- 9. Attributes { name: michael scott age: 53 gender: male status: single } Redis hash Hget <node_id> <attribute> Hgetall <node_id>
- 10. Demo demo
- 11. GRAPH.ADDEDGE <Graph_Name> <SRC_NODE> <Relation> <DEST_NODE> <Attributes> GRAPH.ADDEDGE the_office michael scott boss jim halpert <Attributes> Adding edges
- 12. 1. All edge store 2. Label store Stores ALL BOSS
- 13. SPO OSP SOP PSO OPS POS Hexastore S Subject P Predicate O Object 6
- 15. Who does Michael manages? SPO:Michael:Boss:* Who manages Jim? OPS:Jim:Boss:* Who manages who? PSO:Boss:* Hexastore Michael S Jim O Boss P
- 16. Demo demo
- 17. Query language Cypher MATCH <Graph_name> <graph_pattern> WHERE <filters> RETURN <entities> ORDER BY <properties> LIMIT <count>
- 18. MATCH (Michael:employee {name:michael scott})-[boss]->(E:employee)-[visit]->(c:country) WHERE (E.age > Michael.age AND c.continent = Europe) RETURN E.name, count(c.name) AS countriesVisited ORDER BY countriesVisited, E.age DESC LIMIT 5 Query language Cypher
- 19. Query execution Query Parser AST Exec Plan
- 20. Tokenizer - Lex Parser - Lemon, SQLite LALR(1) parser generator for C Tokenizer & Parser
- 21. End to end AST Root Match Where Return Order
- 22. Execution plan GRAPH.EXPLAIN <Graph_Name> <Query>
- 23. Execution plan MATCH (Michael:employee {name:michael scott})-[boss]->(E:employee) WHERE E.age > Michael.age RETURN E.name ORDER BY E.age DESC LIMIT 5
- 24. Label Scan FilterExpandFilter Project Scan Employees store Yields employee ID
- 25. Label Scan FilterExpandFilter Project Finds Michael, Employee name should equal Michael Scott
- 26. Label Scan FilterExpandFilter Project Consult Hexastore SPO:Michael:boss: Finds Michael’s Employees
- 27. Label Scan FilterExpandFilter Project Finds employees who are older than Michael
- 28. Label Scan RETURN E.name ORDER BY E.age DESC LIMIT 5 Adds employee to result-set Order results by employees age Returns top 5 results FilterExpandFilter Project
- 31. Label Scan Filter Label scan emits Michael
- 32. Label Scan Filter Filter allows Michael to pass
- 33. FilterExpand Search hexastore for: SPO:Michael:Boss MATCH (Michael:employee {name:michael scott})-[boss]->(E:employee)
- 35. ExpandFilter Filter drops Jim, Jim is not older than Michael WHERE E.age > Michael.age
- 37. ExpandFilter Filter allows Kevin to pass Kevin is older than Michael
- 38. FilterProject Kevin Project adds Kevin to result set.
- 39. Demo demo
- 40. Features ● Multi hop, multi entry point (A)-[R1]->(C)<-[R2]-(B) ● Aggregations, Group bys RETURN E.gender, AVG(E.age) AS average_age ● Order bys, Distinct
- 41. Benchmark 150K inserts per second 15K queries per second
- 42. Roadmap ● Tinkerpop Gremlin ● Index entity attributes ● Extend query language
- 43. http://redisgraph.io - Thank You! Roi Lipman Senior Software Engineer Redis Labs @roilipman Itamar Haber Chief OSS Education Officer Redis Labs @itamarhaber