Upgrade to Pro
— share decks privately, control downloads, hide ads and more …
Speaker Deck
Features
Speaker Deck
PRO
Sign in
Sign up for free
Search
Search
Grafos com TinkerPop3 e TitanDB - TDC2016
Search
Mike Dias
July 08, 2016
Technology
0
95
Grafos com TinkerPop3 e TitanDB - TDC2016
Mike Dias
July 08, 2016
Tweet
Share
More Decks by Mike Dias
See All by Mike Dias
Elytics: Métricas em tempo real no Elo7
mikedias
0
120
Grafos distribuídos com Titan:db
mikedias
0
420
Index Activiti data on Elasticsearch
mikedias
0
390
TinkerPop Blueprints
mikedias
0
79
Other Decks in Technology
See All in Technology
学生・新卒・ジュニアから目指すSRE
hiroyaonoe
2
670
CDKで始めるTypeScript開発のススメ
tsukuboshi
1
500
小さく始めるBCP ― 多プロダクト環境で始める最初の一歩
kekke_n
1
470
We Built for Predictability; The Workloads Didn’t Care
stahnma
0
140
顧客の言葉を、そのまま信じない勇気
yamatai1212
1
360
ClickHouseはどのように大規模データを活用したAIエージェントを全社展開しているのか
mikimatsumoto
0
260
Introduction to Sansan, inc / Sansan Global Development Center, Inc.
sansan33
PRO
0
3k
登壇駆動学習のすすめ — CfPのネタの見つけ方と書くときに意識していること
bicstone
3
120
AIエージェントに必要なのはデータではなく文脈だった/ai-agent-context-graph-mybest
jonnojun
0
170
予期せぬコストの急増を障害のように扱う――「コスト版ポストモーテム」の導入とその後の改善
muziyoshiz
1
2k
StrandsとNeptuneを使ってナレッジグラフを構築する
yakumo
1
120
[CV勉強会@関東 World Model 読み会] Orbis: Overcoming Challenges of Long-Horizon Prediction in Driving World Models (Mousakhan+, NeurIPS 2025)
abemii
0
140
Featured
See All Featured
The Art of Programming - Codeland 2020
erikaheidi
57
14k
Bootstrapping a Software Product
garrettdimon
PRO
307
120k
The Director’s Chair: Orchestrating AI for Truly Effective Learning
tmiket
1
98
Conquering PDFs: document understanding beyond plain text
inesmontani
PRO
4
2.3k
Improving Core Web Vitals using Speculation Rules API
sergeychernyshev
21
1.4k
Color Theory Basics | Prateek | Gurzu
gurzu
0
200
The Illustrated Guide to Node.js - THAT Conference 2024
reverentgeek
0
260
Performance Is Good for Brains [We Love Speed 2024]
tammyeverts
12
1.4k
ピンチをチャンスに:未来をつくるプロダクトロードマップ #pmconf2020
aki_iinuma
128
55k
Put a Button on it: Removing Barriers to Going Fast.
kastner
60
4.2k
Agile Actions for Facilitating Distributed Teams - ADO2019
mkilby
0
120
Templates, Plugins, & Blocks: Oh My! Creating the theme that thinks of everything
marktimemedia
31
2.7k
Transcript
Titan:db TinkerPop3 Grafos com e @mike_dias
@mike_dias
None
Agenda • Grafos • TinkerPop3 • TitanDB
Grafos
mother name: hercules age: 30 type: demigod name: alcmene age:
24 type: human name: jupter age: 5000 type: god father
mother name: hercules age: 30 type: demigod name: alcmene age:
24 type: human name: jupter age: 5000 type: god father Vertex
mother name: hercules age: 30 type: demigod name: alcmene age:
24 type: human name: jupter age: 5000 type: god father Edge
mother name: hercules age: 30 type: demigod name: alcmene age:
24 type: human name: jupter age: 5000 type: god father Property
Property Graph
Por que usar grafos
Modelagem intuitiva
Queries expressivas
Casos de uso Recomendações personalizadas Influência social
Casos de uso Melhor rota Detecção de fraude
Quero usar grafos! #comofas?
Apache TinkerPop • Vendor-agnostic graph compute framework • Gremlin -
graph traversal language • Capacidades OLTP e OLAP
Core API Graph graph = TinkerGraph.open(); Vertex nosql =
graph.addVertex("trilha"); nosql.property("nome", "NoSQL"); Vertex tdc2016 = graph.addVertex("evento"); tdc2016.property("nome", "TDC 2016"); Edge e = nosql.addEdge("compoe", tdc2016); e.property("data", "08/07/2016");
Core API Graph graph = Neo4jGraph.open("..."); Vertex nosql =
graph.addVertex("trilha"); nosql.property("nome", "NoSQL"); Vertex tdc2016 = graph.addVertex("evento"); tdc2016.property("nome", "TDC 2016"); Edge e = nosql.addEdge("compoe", tdc2016); e.property("data", "08/07/2016");
TinkerPop-enabled Vendors TitanGraph.open(…) Neo4jGraph.open(…) OrientGraph.open(…) StardogGraph.open(…) DSEGraph.open(…) BlazeGraph.open(…)
Gremlin lang What are the names of Gremlin's friends' friends?
g.V().has("name","gremlin"). out("knows"). out("knows"). values("name")
Gremlin console
OLTP vs OLAP
Spark computer graph = GraphFactory.open(…); g = graph.traversal(). withComputer(SparkGraphComputer.class);
g.V().hasLabel("person").groupCount().by("age")
Giraph computer graph = GraphFactory.open(…); g = graph.traversal(). withComputer(GiraphGraphComputer.class);
g.V().hasLabel("person").groupCount().by("age")
Gremlin server REST WebSockets
Learn more http://sql2gremlin.com https://academy.datastax.com/courses/ds232-the-gremlin-graph-query-language https://academy.datastax.com/resources/getting-started-graph-databases http://tinkerpop.apache.org
Meu grafo é muito grande. #comofas?
Titan:db • Suporte nativo ao TinkerPop3 • Storages configuráveis •
Licença Apache 2
Backend flexível
Titan:db + • Disponibilidade contínua sem ponto único de falha
• Sem gargalos na leitura ou escrita no grafo • Escalabilidade elástica • Cache de dados frequentemente acessados
• Leituras e escritas estritamente consistentes • Escalabilidade linear •
Integração com o ecossistema Hadoop Titan:db +
g.V().has('name',CONTAINS,'Bob') g.V().has('place',WITHIN,Geoshape.circle(37.97,23.72,50)) Full text & Geo Search
Vertex centric index
Graph partitioning
Disclaimer
Quem está usando?
http://arli.us/edu-planet-scale
http://arli.us/magazinaluiza
http://arli.us/cisco-sec1
http://pt.slideshare.net/Hadoop_Summit/t-235p230-ctang
https://videos.dabcc.com/aws-reinvent-2015-dat203-building- graph-databases-on-aws/
Concluindo… • Vários casos de uso resolvidos com Grafos •
Quando for usar grafos, use TinkerPop3 • Titan transforma Big Data em Big "Graph" Data
Obrigado! @mike_dias
[email protected]
Dúvidas?