big graph data
DESCRIPTION
The problems we are faced with in the 21st century require efficient analysis of ever more complex systems. This presentation outlines how such problems can be better understood and effectively solved if they are modeled as graphs or networks. We present two tools for to help solve such problems at scale: Titan, which is a real-time distributed graph database based on Apache Cassandra and Hbase and Faunus, which is a batch analytics framework for graphs based on Apache Hadoop. We discuss their current development status as of November 2012 and illustrate an example application for the GitHub coding network.TRANSCRIPT
![Page 1: Big Graph Data](https://reader034.vdocuments.us/reader034/viewer/2022051514/54b79e8b4a795993718b464b/html5/thumbnails/1.jpg)
AURELIUS THINKAURELIUS.COM
BIG GRAPH DATA Understanding a Complex World
Matthias Broecheler, CTO @mbroecheler November XIII, MMXII
![Page 2: Big Graph Data](https://reader034.vdocuments.us/reader034/viewer/2022051514/54b79e8b4a795993718b464b/html5/thumbnails/2.jpg)
AURELIUS THINKAURELIUS.COM
I Graph Foundation
![Page 3: Big Graph Data](https://reader034.vdocuments.us/reader034/viewer/2022051514/54b79e8b4a795993718b464b/html5/thumbnails/3.jpg)
Graph
name: Jupiter type: god
name: Pluto type: god
name: Neptune type: god
name: Hercules type: demigod
name: Cerberus type: monster
name: Alcmene type: god
name: Saturn type: titan
Vertex Property
![Page 4: Big Graph Data](https://reader034.vdocuments.us/reader034/viewer/2022051514/54b79e8b4a795993718b464b/html5/thumbnails/4.jpg)
Graph
name: Jupiter type: god
name: Pluto type: god
name: Neptune type: god
name: Hercules type: demigod
name: Cerberus type: monster
name: Alcmene type: god
name: Saturn type: titan
father father
mother brother
brother battled
pet
time:12
Edge
Edge Property
Edge Type
![Page 5: Big Graph Data](https://reader034.vdocuments.us/reader034/viewer/2022051514/54b79e8b4a795993718b464b/html5/thumbnails/5.jpg)
Path
name: Jupiter type: god
name: Pluto type: god
name: Neptune type: god
name: Hercules type: demigod
name: Cerberus type: monster
name: Alcmene type: god
name: Saturn type: titan
father father
mother brother
brother battled
pet
time:12
![Page 6: Big Graph Data](https://reader034.vdocuments.us/reader034/viewer/2022051514/54b79e8b4a795993718b464b/html5/thumbnails/6.jpg)
Degree
name: Jupiter type: god
name: Pluto type: god
name: Neptune type: god
name: Hercules type: demigod
name: Cerberus type: monster
name: Alcmene type: god
name: Saturn type: titan
father father
mother brother
brother battled
pet
time:12
![Page 7: Big Graph Data](https://reader034.vdocuments.us/reader034/viewer/2022051514/54b79e8b4a795993718b464b/html5/thumbnails/7.jpg)
AURELIUS THINKAURELIUS.COM
I Connected World
![Page 8: Big Graph Data](https://reader034.vdocuments.us/reader034/viewer/2022051514/54b79e8b4a795993718b464b/html5/thumbnails/8.jpg)
HEALTH
![Page 9: Big Graph Data](https://reader034.vdocuments.us/reader034/viewer/2022051514/54b79e8b4a795993718b464b/html5/thumbnails/9.jpg)
HEALTH
![Page 10: Big Graph Data](https://reader034.vdocuments.us/reader034/viewer/2022051514/54b79e8b4a795993718b464b/html5/thumbnails/10.jpg)
HEALTH
![Page 11: Big Graph Data](https://reader034.vdocuments.us/reader034/viewer/2022051514/54b79e8b4a795993718b464b/html5/thumbnails/11.jpg)
HEALTH
![Page 12: Big Graph Data](https://reader034.vdocuments.us/reader034/viewer/2022051514/54b79e8b4a795993718b464b/html5/thumbnails/12.jpg)
ECONOMY
![Page 13: Big Graph Data](https://reader034.vdocuments.us/reader034/viewer/2022051514/54b79e8b4a795993718b464b/html5/thumbnails/13.jpg)
ECONOMY
![Page 14: Big Graph Data](https://reader034.vdocuments.us/reader034/viewer/2022051514/54b79e8b4a795993718b464b/html5/thumbnails/14.jpg)
ECONOMY
![Page 15: Big Graph Data](https://reader034.vdocuments.us/reader034/viewer/2022051514/54b79e8b4a795993718b464b/html5/thumbnails/15.jpg)
ECONOMY
![Page 16: Big Graph Data](https://reader034.vdocuments.us/reader034/viewer/2022051514/54b79e8b4a795993718b464b/html5/thumbnails/16.jpg)
Social Systems
![Page 17: Big Graph Data](https://reader034.vdocuments.us/reader034/viewer/2022051514/54b79e8b4a795993718b464b/html5/thumbnails/17.jpg)
Social Systems
![Page 18: Big Graph Data](https://reader034.vdocuments.us/reader034/viewer/2022051514/54b79e8b4a795993718b464b/html5/thumbnails/18.jpg)
Social Systems
![Page 19: Big Graph Data](https://reader034.vdocuments.us/reader034/viewer/2022051514/54b79e8b4a795993718b464b/html5/thumbnails/19.jpg)
Social Systems
![Page 20: Big Graph Data](https://reader034.vdocuments.us/reader034/viewer/2022051514/54b79e8b4a795993718b464b/html5/thumbnails/20.jpg)
![Page 21: Big Graph Data](https://reader034.vdocuments.us/reader034/viewer/2022051514/54b79e8b4a795993718b464b/html5/thumbnails/21.jpg)
![Page 22: Big Graph Data](https://reader034.vdocuments.us/reader034/viewer/2022051514/54b79e8b4a795993718b464b/html5/thumbnails/22.jpg)
![Page 23: Big Graph Data](https://reader034.vdocuments.us/reader034/viewer/2022051514/54b79e8b4a795993718b464b/html5/thumbnails/23.jpg)
AURELIUS THINKAURELIUS.COM
III Titan Graph Database
![Page 24: Big Graph Data](https://reader034.vdocuments.us/reader034/viewer/2022051514/54b79e8b4a795993718b464b/html5/thumbnails/24.jpg)
Numerous Concurrent Users Many Short Transactions
read/write
Real-time Traversals (OLTP) High Availability Dynamic Scalability Variable Consistency Model
ACID or eventual consistency
Real-time Big Graph Data
Titan Features
![Page 25: Big Graph Data](https://reader034.vdocuments.us/reader034/viewer/2022051514/54b79e8b4a795993718b464b/html5/thumbnails/25.jpg)
Storage Backends
Partitionability
Availability Consistency
![Page 26: Big Graph Data](https://reader034.vdocuments.us/reader034/viewer/2022051514/54b79e8b4a795993718b464b/html5/thumbnails/26.jpg)
Titan Features
I. Data Management
II. Vertex-Centric Indices
![Page 27: Big Graph Data](https://reader034.vdocuments.us/reader034/viewer/2022051514/54b79e8b4a795993718b464b/html5/thumbnails/27.jpg)
Titan Features
III. Graph Partitioning
IV. Edge Compression
![Page 28: Big Graph Data](https://reader034.vdocuments.us/reader034/viewer/2022051514/54b79e8b4a795993718b464b/html5/thumbnails/28.jpg)
Titan Ecosystem
Native Blueprints Implementation
Gremlin Query Language
Rexster Server any Titan graph can be
exposed as a REST endpoint
Generic Graph API
Dataflow Processing
TraversalLanguage
Object-GraphMapper
GraphAlgorithms
GraphServer
![Page 29: Big Graph Data](https://reader034.vdocuments.us/reader034/viewer/2022051514/54b79e8b4a795993718b464b/html5/thumbnails/29.jpg)
AURELIUS THINKAURELIUS.COM
IV Github Network
![Page 30: Big Graph Data](https://reader034.vdocuments.us/reader034/viewer/2022051514/54b79e8b4a795993718b464b/html5/thumbnails/30.jpg)
Setup
$ ./titan-0.1.0/bin/gremlin.sh! ! ! !\,,,/! (o o)!-----oOOo-(_)-oOOo-----!gremlin> g = TitanFactory.open('/tmp/titan')!==>titangraph[local:/tmp/titan]!
![Page 31: Big Graph Data](https://reader034.vdocuments.us/reader034/viewer/2022051514/54b79e8b4a795993718b464b/html5/thumbnails/31.jpg)
Titan Storage Model
Adjacency list in one column family
Row key = vertex id
Each property and edge in one column Denormalized, i.e. stored twice
Direction and label/key as column prefix Use slice predicate for quick retrieval
5
5
![Page 32: Big Graph Data](https://reader034.vdocuments.us/reader034/viewer/2022051514/54b79e8b4a795993718b464b/html5/thumbnails/32.jpg)
USER
REPOSITORY
COMMENT
ISSUE COMMIT
PAGE
edited
pushed
in
opened
to on on
on
created
![Page 33: Big Graph Data](https://reader034.vdocuments.us/reader034/viewer/2022051514/54b79e8b4a795993718b464b/html5/thumbnails/33.jpg)
Defining Property Keys
gremlin> g.makeType().name(‘username’).!! ! ! dataType(String.class).!! ! ! functional().!! ! ! indexed().unique().!! ! ! makePropertyKey()!
gremlin> g.makeType().name(‘time’).!! ! ! dataType(Long.class).!! ! ! functional().makePropertyKey()!
![Page 34: Big Graph Data](https://reader034.vdocuments.us/reader034/viewer/2022051514/54b79e8b4a795993718b464b/html5/thumbnails/34.jpg)
Defining Edge Labels
gremlin> g.makeType().name(‘on’).!! ! ! makeEdgeLabel()!
gremlin> g.makeType().name(‘pushed’).!! ! ! primaryKey(time).!! ! ! makeEdgeLabel()!
gremlin> g.makeType().name(‘in’).!! ! ! unidirected().!! ! ! makeEdgeLabel()!
![Page 35: Big Graph Data](https://reader034.vdocuments.us/reader034/viewer/2022051514/54b79e8b4a795993718b464b/html5/thumbnails/35.jpg)
Create & Retrieve
gremlin> v = g.addVertex([username: ‘okram’])!==>v[4]!gremlin> v.map!==>{username=okram}!gremlin> g.V('username','okram')!==>v[4]!
![Page 36: Big Graph Data](https://reader034.vdocuments.us/reader034/viewer/2022051514/54b79e8b4a795993718b464b/html5/thumbnails/36.jpg)
Titan Locking
Locking ensures consistency when it is needed
Titan uses time stamped consistent reads and writes on separate CFs for locking
Uses Property uniqueness: .unique()
Functional edges: .functional()
Global ID management
5
9
name : Hercules
name : Hercules
name : Jupiter
name : Pluto
father
father
x
![Page 37: Big Graph Data](https://reader034.vdocuments.us/reader034/viewer/2022051514/54b79e8b4a795993718b464b/html5/thumbnails/37.jpg)
Titan Indexing
Vertices can be retrieved by property key + value
Titan maintains index in a separate column family as graph is updated
Only need to define a property key as .index()
5
9
name : Hercules
name : Jupiter
![Page 38: Big Graph Data](https://reader034.vdocuments.us/reader034/viewer/2022051514/54b79e8b4a795993718b464b/html5/thumbnails/38.jpg)
Basic Queries
gremlin> v.out(‘pushed’)!gremlin> v.out(‘pushed’).out(‘to’).name!gremlin> v.out(‘pushed’).out(‘to’).dedup.name!gremlin> v.out(‘pushed’).out(‘to’).dedup.!! ! ! name.sort{it}!
gremlin> v.outE(‘pushed’).has(‘time’,T.gt,1000).inV!
![Page 39: Big Graph Data](https://reader034.vdocuments.us/reader034/viewer/2022051514/54b79e8b4a795993718b464b/html5/thumbnails/39.jpg)
Basic Queries
gremlin> v.out(‘pushed’)!gremlin> v.out(‘pushed’).out(‘to’).name!gremlin> v.out(‘pushed’).out(‘to’).dedup.name!gremlin> v.out(‘pushed’).out(‘to’).dedup.!! ! ! name.sort{it}!
gremlin> v.outE(‘pushed’).has(‘time’,T.gt,1000).inV!
Query Optimization
![Page 40: Big Graph Data](https://reader034.vdocuments.us/reader034/viewer/2022051514/54b79e8b4a795993718b464b/html5/thumbnails/40.jpg)
Vertex-Centric Indices
Sort and index edges per vertex by primary key Primary key can be composite
Enables efficient focused traversals Only retrieve edges that matter
Uses push down predicates for quick, index-driven retrieval
![Page 41: Big Graph Data](https://reader034.vdocuments.us/reader034/viewer/2022051514/54b79e8b4a795993718b464b/html5/thumbnails/41.jpg)
v
time: 1
fought fought father
mother
battled battled battled
battled
time: 3 time: 5
time: 9 v.query()!
![Page 42: Big Graph Data](https://reader034.vdocuments.us/reader034/viewer/2022051514/54b79e8b4a795993718b464b/html5/thumbnails/42.jpg)
v
time: 1
father
mother
battled battled battled
battled
time: 3 time: 5
time: 9 v.query()! .direction(OUT)!
![Page 43: Big Graph Data](https://reader034.vdocuments.us/reader034/viewer/2022051514/54b79e8b4a795993718b464b/html5/thumbnails/43.jpg)
v
time: 1
battled battled battled
battled
time: 3 time: 5
time: 9 v.query()! .direction(OUT)! .labels(‘battled’)!
![Page 44: Big Graph Data](https://reader034.vdocuments.us/reader034/viewer/2022051514/54b79e8b4a795993718b464b/html5/thumbnails/44.jpg)
v
time: 1
battled battled
time: 3
v.query()! .direction(OUT)! .labels(‘battled’)! .has(‘time,T.lt,5)!
![Page 45: Big Graph Data](https://reader034.vdocuments.us/reader034/viewer/2022051514/54b79e8b4a795993718b464b/html5/thumbnails/45.jpg)
Recommendation Engine
gremlin> v.out('pushed').out('to')[0..9].!! ! ! in('to').in('pushed')[0..500].!! ! ! except([v]).name.!! ! ! groupCount.cap.next().sort{-it.value}[0..4]!
![Page 46: Big Graph Data](https://reader034.vdocuments.us/reader034/viewer/2022051514/54b79e8b4a795993718b464b/html5/thumbnails/46.jpg)
Recommendation Engine
gremlin> v.out('pushed').out('to')[0..9].!! ! ! in('to').in('pushed')[0..500].!! ! ! except([v]).name.!! ! ! groupCount.cap.next().sort{-it.value}[0..4]!
v = g.V(‘username’,’okram’):!==>lvca=175!==>spmallette=56!==>sgomezvillamor=36!==>mbroecheler=33!==>joshsh=20!
![Page 47: Big Graph Data](https://reader034.vdocuments.us/reader034/viewer/2022051514/54b79e8b4a795993718b464b/html5/thumbnails/47.jpg)
Recommendation Engine
gremlin> v.out('pushed').out('to')[0..9].!! ! ! in('to').in('pushed')[0..500].!! ! ! except([v]).name.!! ! ! groupCount.cap.next().sort{-it.value}[0..4]!
v = g.V(‘username’,’torvalds’):!==>iksaif=90!==>rjwysocki=22!==>kernel-digger=20!==>giuseppecalderaro=16!==>groeck=15!
![Page 48: Big Graph Data](https://reader034.vdocuments.us/reader034/viewer/2022051514/54b79e8b4a795993718b464b/html5/thumbnails/48.jpg)
Titan Embedding
Rexster RexPro lightweight Gremlin
Server
based on Grizzly
Titan Gremlin Engine
Embedded Storage Backend in-JVM method calls
![Page 49: Big Graph Data](https://reader034.vdocuments.us/reader034/viewer/2022051514/54b79e8b4a795993718b464b/html5/thumbnails/49.jpg)
Graph Partitioning
Goal: Vertex Co-location Titan maintains multiple
ID Pools
Ordered Partitioner in Storage Backend
Dynamically determines optimal partition and allocates corresponding IDs
ID Pool
![Page 50: Big Graph Data](https://reader034.vdocuments.us/reader034/viewer/2022051514/54b79e8b4a795993718b464b/html5/thumbnails/50.jpg)
What’s coming
Full-text indexing external index system integration
Bulk Loading integration with storage backend
utilities and Hadoop ingestion
240 Billion Edge Benchmark performance analysis and improvements
across the entire stack
![Page 51: Big Graph Data](https://reader034.vdocuments.us/reader034/viewer/2022051514/54b79e8b4a795993718b464b/html5/thumbnails/51.jpg)
AURELIUS THINKAURELIUS.COM
V Faunus Graph Analytics
![Page 52: Big Graph Data](https://reader034.vdocuments.us/reader034/viewer/2022051514/54b79e8b4a795993718b464b/html5/thumbnails/52.jpg)
Hadoop-based Graph Computing Framework
Graph Analytics
Breadth-first Traversals
Global Graph Computations
Batch Big Graph Data
Faunus Features
![Page 53: Big Graph Data](https://reader034.vdocuments.us/reader034/viewer/2022051514/54b79e8b4a795993718b464b/html5/thumbnails/53.jpg)
Faunus Architecture
g._()!
![Page 54: Big Graph Data](https://reader034.vdocuments.us/reader034/viewer/2022051514/54b79e8b4a795993718b464b/html5/thumbnails/54.jpg)
Faunus Work Flow
hdfs://user/ubuntu/
output/job-0/
output/job-1/
output/job-2/ { graph*
sideeffect*
g.V.out .out .count()
Compressed HDFS Graphs stored in sequence files variable length encoding prefix compression
![Page 55: Big Graph Data](https://reader034.vdocuments.us/reader034/viewer/2022051514/54b79e8b4a795993718b464b/html5/thumbnails/55.jpg)
Faunus Setup
$ bin/gremlin.sh !
\,,,/! (o o)!-----oOOo-(_)-oOOo-----!gremlin> g = FaunusFactory.open('bin/titan-hbase.properties')!==>faunusgraph[titanhbaseinputformat]!gremlin> g.getProperties()!==>faunus.graph.input.format=com.thinkaurelius.faunus.formats.titan.hbase.TitanHBaseInputFormat!==>faunus.graph.output.format=org.apache.hadoop.mapreduce.lib.output.SequenceFileOutputFormat!==>faunus.sideeffect.output.format=org.apache.hadoop.mapreduce.lib.output.TextOutputFormat!==>faunus.output.location=dbpedia!==>faunus.output.location.overwrite=true!
gremlin> g._() !12/11/09 15:17:45 INFO mapreduce.FaunusCompiler: Compiled to 1 MapReduce job(s)!12/11/09 15:17:45 INFO mapreduce.FaunusCompiler: Executing job 1 out of 1: MapSequence[com.thinkaurelius.faunus.mapreduce.transform.IdentityMap.Map]!12/11/09 15:17:50 INFO mapred.JobClient: Running job: job_201211081058_0003!
![Page 56: Big Graph Data](https://reader034.vdocuments.us/reader034/viewer/2022051514/54b79e8b4a795993718b464b/html5/thumbnails/56.jpg)
Graph Analytics
gremlin> g.E.has('label',’followed').keep.!! ! !V.sideEffect('{it.degree = it.outE.count()}').!! ! !degree.groupCount!
gremlin> g.E.has('label','pushed').keep.!! ! !V.sideEffect('{it.degree = it.outE.count()}').!! ! !degree.groupCount!
![Page 57: Big Graph Data](https://reader034.vdocuments.us/reader034/viewer/2022051514/54b79e8b4a795993718b464b/html5/thumbnails/57.jpg)
Follow Degree Distribution
![Page 58: Big Graph Data](https://reader034.vdocuments.us/reader034/viewer/2022051514/54b79e8b4a795993718b464b/html5/thumbnails/58.jpg)
Follow Degree Distribution
P(k) ~ k-γ
γ = 2.2
![Page 59: Big Graph Data](https://reader034.vdocuments.us/reader034/viewer/2022051514/54b79e8b4a795993718b464b/html5/thumbnails/59.jpg)
Pushed Degree Distribution
![Page 60: Big Graph Data](https://reader034.vdocuments.us/reader034/viewer/2022051514/54b79e8b4a795993718b464b/html5/thumbnails/60.jpg)
Global Recommendations
gremlin> g.E.has('label','pushed','to').keep.!! ! !V.out('pushed').out('to').!! ! !in('to').in('pushed').!! ! !sideEffect('{it.score =it.pathCounter}').!! ! !score.order(F.decr,'name')!
# Top 5:!Jippi ! ! ! !60892182927!garbear ! ! !30095282886!FakeHeal ! ! !30038040349!brianchandotcom !24684133382!nyarla ! ! !15230275746!
![Page 61: Big Graph Data](https://reader034.vdocuments.us/reader034/viewer/2022051514/54b79e8b4a795993718b464b/html5/thumbnails/61.jpg)
What’s coming
Faunus 0.1
Bulk Loading loaded graph into Titan
loading derivations into Titan
Extending Gremlin Support currently only a subset is of
Gremlin implemented
Operational Tools
![Page 62: Big Graph Data](https://reader034.vdocuments.us/reader034/viewer/2022051514/54b79e8b4a795993718b464b/html5/thumbnails/62.jpg)
I Graph = Relationship Centric
![Page 63: Big Graph Data](https://reader034.vdocuments.us/reader034/viewer/2022051514/54b79e8b4a795993718b464b/html5/thumbnails/63.jpg)
II Graph = Agile Data Model
![Page 64: Big Graph Data](https://reader034.vdocuments.us/reader034/viewer/2022051514/54b79e8b4a795993718b464b/html5/thumbnails/64.jpg)
III Graph = Algebraic Data Model
![Page 65: Big Graph Data](https://reader034.vdocuments.us/reader034/viewer/2022051514/54b79e8b4a795993718b464b/html5/thumbnails/65.jpg)
Aurelius Graph Cluster
Stores a massive-scale property graph allowing real-time traversals and updates
Batch processing of large graphs with Hadoop
Runs global graph algorithms on large, compressed,
in-memory graphs
Map/Reduce Load & Compress
Analysis results back into Titan
Apache 2
![Page 66: Big Graph Data](https://reader034.vdocuments.us/reader034/viewer/2022051514/54b79e8b4a795993718b464b/html5/thumbnails/66.jpg)
The Graph Landscape Sp
eed
of T
rave
rsal
/Pro
cess
Size of Graph Illustration only, not to scale
![Page 67: Big Graph Data](https://reader034.vdocuments.us/reader034/viewer/2022051514/54b79e8b4a795993718b464b/html5/thumbnails/67.jpg)
TINKERPOP.COM
![Page 68: Big Graph Data](https://reader034.vdocuments.us/reader034/viewer/2022051514/54b79e8b4a795993718b464b/html5/thumbnails/68.jpg)
AURELIUS THINKAURELIUS.COM
Thanks!
Vadas Gintautas @vadasg
Marko Rodriguez @twarko
Stephen Mallette @spmallette
Daniel LaRocque
![Page 69: Big Graph Data](https://reader034.vdocuments.us/reader034/viewer/2022051514/54b79e8b4a795993718b464b/html5/thumbnails/69.jpg)
AURELIUS THINKAURELIUS.COM
![Page 70: Big Graph Data](https://reader034.vdocuments.us/reader034/viewer/2022051514/54b79e8b4a795993718b464b/html5/thumbnails/70.jpg)
AURELIUS THINKAURELIUS.COM
XVX Benchmark Results
![Page 71: Big Graph Data](https://reader034.vdocuments.us/reader034/viewer/2022051514/54b79e8b4a795993718b464b/html5/thumbnails/71.jpg)
AURELIUS THINKAURELIUS.COM
XVX - I Titan Performance Evaluation on Twitter-like Benchmark
![Page 72: Big Graph Data](https://reader034.vdocuments.us/reader034/viewer/2022051514/54b79e8b4a795993718b464b/html5/thumbnails/72.jpg)
Twitter Benchmark
1.47 billion followship edges and 41.7 million users Loaded into Titan using BatchGraph
Twitter in 2009, crawled by Kwak et. al
4 Transaction Types Create Account (1%)
Publish tweet (15%)
Read stream (76%)
Recommendation (8%) Follow recommended user (30%)
Kwak, H., Lee, C., Park, H., Moon, S., “What is Twitter, a Social Network or a News Media?,” World Wide Web Conference, 2010.
![Page 73: Big Graph Data](https://reader034.vdocuments.us/reader034/viewer/2022051514/54b79e8b4a795993718b464b/html5/thumbnails/73.jpg)
Benchmark Setup
6 cc1.4xl Cassandra nodes in one placement group
Cassandra 1.10
40 m1.small worker machines repeatedly running transactions
simulating servers handling user requests
EC2 cost: $11/hour
![Page 74: Big Graph Data](https://reader034.vdocuments.us/reader034/viewer/2022051514/54b79e8b4a795993718b464b/html5/thumbnails/74.jpg)
Benchmark Results
Transaction Type Number of tx Mean tx time Std of tx time
Create account 379,019 115.15 ms 5.88 ms
Publish tweet 7,580,995 18.45 ms 6.34 ms
Read stream 37,936,184 6.29 ms 1.62 ms
Recommendation 3,793,863 67.65 ms 13.89 ms
Total 49,690,061
Runtime 2.3 hours 5,900 tx/sec
![Page 75: Big Graph Data](https://reader034.vdocuments.us/reader034/viewer/2022051514/54b79e8b4a795993718b464b/html5/thumbnails/75.jpg)
Peak Load Results
Transaction Type Number of tx Mean tx time Std of tx time
Create account 374,860 172.74 ms 10.52 ms
Publish tweet 7,517,667 70.07 ms 19.43 ms
Read stream 37,618,648 24.40 ms 3.18 ms
Recommendation 3,758,266 229.83 ms 29.08 ms
Total 49,269,441
Runtime 1.3 hours 10,200 tx/sec
![Page 76: Big Graph Data](https://reader034.vdocuments.us/reader034/viewer/2022051514/54b79e8b4a795993718b464b/html5/thumbnails/76.jpg)
Benchmark Conclusion
Titan can handle 10s of thousands of concurrent users with short response 5mes even for complex traversals on a simulated social networking applica5on based on real-‐world network data with billions of edges and millions of users in a standard EC2 deployment.
For more informa5on on the benchmark: hDp://thinkaurelius.com/2012/08/06/5tan-‐provides-‐real-‐5me-‐big-‐graph-‐data/