Trinity A Distributed Graph Engine on a Memory Cloud 1 Motivation - - PowerPoint PPT Presentation

trinity
SMART_READER_LITE
LIVE PREVIEW

Trinity A Distributed Graph Engine on a Memory Cloud 1 Motivation - - PowerPoint PPT Presentation

Trinity A Distributed Graph Engine on a Memory Cloud 1 Motivation Polarized view on previous approaches: Graph Query Graph Scalability Database Processing Analytics Neo4J Yes Yes No No HyperGraph Yes Yes No No DB MapReduce


slide-1
SLIDE 1

Trinity

A Distributed Graph Engine on a Memory Cloud

1

slide-2
SLIDE 2

Motivation

Polarized view on previous approaches:

2

Graph Database Query Processing Graph Analytics Scalability Neo4J Yes Yes No No HyperGraph DB Yes Yes No No MapReduce No No Yes Yes Pregel No No Yes Yes

slide-3
SLIDE 3

Concept

  • Globally shared memory
  • Distributed key/value store
  • Random access abstraction for online queries

(BFS, graph-matching)

  • Scalabililty through partitioning
  • Restrictions on message passing for improved

performance

3

slide-4
SLIDE 4

Architecture 1/2

4

slide-5
SLIDE 5

Architecture 2/2

5

slide-6
SLIDE 6

6

slide-7
SLIDE 7

Message passing model

  • Having all messages in memory: infeasible
  • Assumption: fixed set of messaging partners
  • Bipartite view effective but costly, better:

7

slide-8
SLIDE 8

Trinity Specification Language

Object oriented manipulation through cell accessor abstraction

8

slide-9
SLIDE 9

Evaluation - BFS

9

slide-10
SLIDE 10

Evaluation – Memory usage

10

slide-11
SLIDE 11

Conclusion

  • Unified approach for distributed graph

processing

  • Efficient shared memory abstraction
  • Fault recovery through checkpointing

(depending on task)

  • Library / coordination unclear
  • No ACID transactions

11