ScaleJoin: a Deterministic, Disjoint-Parallel and Skew-Resilient Stream Join

Vincenzo Gulisano
Vincenzo GulisanoAssociate Professor at Chalmers University of Technology
ScaleJoin: a Deterministic,
Disjoint-Parallel and Skew-Resilient
Stream Join
Vincenzo Gulisano, Yiannis Nikolakopoulos,
Marina Papatriantafilou, Philippas Tsigas
2015-10-31 1
Chalmers University
of technology
Agenda
• What is a stream join?
• Which are the challenges of a parallel stream join?
• Why ScaleJoin?
• How well does ScaleJoin addresses stream joins’
challenges?
• Conclusions
2015-10-31 2
Agenda
• What is a stream join?
• Which are the challenges of a parallel stream join?
• Why ScaleJoin?
• How well does ScaleJoin addresses stream joins’
challenges?
• Conclusions
2015-10-31 3
Motivation
Applications in sensor networks, cyber-physical
systems:
• large and fluctuating volumes of data generated
continuously
demand for:
• Continuous processing of data streams
• In a real-time fashion
Store-then-process is not feasible!!!
2015-10-31 4
What is a stream join?
2015-10-31 5
Data stream:
unbounded sequence of tuples
t1
t2
t3
t4
t1
t2
t3
t4
t1
t2
t3
t4
R S
Sliding
window Window
size WS
WSWR
Predicate P
Why parallel stream joins?
• WS = 600 seconds
• R receives 500 tuples/second
• S receives 500 tuples/second
• WR will contain 300,000 tuples
• WS will contain 300,000 tuples
• Each new tuple from R gets compared with
all the tuples in WS
• Each new tuple from S gets compared with
all the tuples in WR
… 300,000,000 comparisons/second!
t1
t2
t3
t4
t1
t2
t3
t4
R S
WSWR
2015-10-31 6
Agenda
• What is a stream join?
• Which are the challenges of a parallel stream join?
• Why ScaleJoin?
• How well does ScaleJoin addresses stream joins’
challenges?
• Conclusions
2015-10-31 7
Which are the challenges of a parallel stream join?
Scalability
High
throughput
Low latency
Disjoint
parallelism
Skew
resilience
Determinism
2015-10-31 8
Agenda
• What is a stream join?
• Which are the challenges of a parallel stream join?
• Why ScaleJoin?
• How well does ScaleJoin addresses stream joins’
challenges?
• Conclusions
2015-10-31 9
The 3-step procedure (sequential stream join)
For each incoming tuple t:
1. compare t with all tuples in opposite window given predicate P
2. add t to its window
3. remove stale tuples from t’s window
Add tuples to S
Add tuples to R
Prod
R
Prod
S
Consume resultsConsPU
2015-10-31 10
We assume each
producer delivers tuples
in timestamp order
The 3-step procedure, is it enough?
Scalability
High
throughput
Low latency
Disjoint
parallelism
Skew
resilience
Determinism
2015-10-31 11
t1
t2
t1
t2
R S
WSWR
t3
t1
t2
t1
t2
R S
WSWR
t4
t3
Enforcing determinism in sequential stream joins
• Next tuple to process = earliest(tS,tR)
• The earliest(tS,tR) tuple is referred to as the next ready tuple
• Process ready tuples in timestamp order  Determinism
PU
tS tR
2015-10-31 12
Deterministic 3-step procedure
Pick the next ready tuple t:
1. compare t with all tuples in opposite window given predicate P
2. add t to its window
3. remove stale tuples from t’s window
Add tuples to S
Add tuples to R
Prod
R
Prod
S
Consume resultsConsPU
2015-10-31 13
Shared-nothing parallel stream join
(state-of-the-art)
Prod
R
Prod
S
PU1
PU2
PUN
… Cons
Add tuple to PUi S
Add tuple to PUi R
Consume results
Pick the next ready tuple t:
1. compare t with all tuples in opposite window given P
2. add t to its window
3. remove stale tuples from t’s window
Chose a PU
Chose a PU
Take the next
ready output tuple
Scalability
High
throughput
Low latency
Disjoint
parallelism
Skew
resilience
Determinism
2015-10-31 14
Merge
Shared-nothing parallel stream join
(state-of-the-art)
Prod
R
Prod
S
PU1
PU2
PUN
…
2015-10-31 15
enqueue()
dequeue()
ConsMerge
From coarse-grained to fine-grained synchronization
Prod
R
Prod
S
PU1
PU2
PUN
…
Cons
2015-10-31 16
ScaleGate
2015-10-31 17
addTuple(tuple,sourceID)
allows a tuple from sourceID to be merged by ScaleGate in the
resulting timestamp-sorted stream of ready tuples.
getNextReadyTuple(readerID)
provides to readerID the next earliest ready tuple that has not been
yet consumed by the former.
https://github.com/dcs-chalmers/ScaleGate_Java
ScaleJoin
Prod
R
Prod
S
PU1
PU2
PUN
…
Cons
Add tuple SGin
Add tuple SGin
Get next ready
output tuple
from SGout
Get next ready input tuple from SGin
1. compare t with all tuples in opposite window given P
2. add t to its window in a round-robin fashion
3. remove stale tuples from t’s window
2015-10-31 18
SGin SGout
Steps for PU
2015-10-31 19
t1
t2
R S
WR
t3
t4
R S
t4
t1
WR
R S
t4
t2
WR
R S
t4
WR
t3
Sequential stream join:
ScaleJoin with 3 PUs:
ScaleJoin (example)
ScaleJoin
Prod
R
Prod
S
PU1
PU2
PUN
… Cons
Add tuple SGin
Add tuple SGin
Get next ready
output tuple
from SGout
2015-10-31 20
SGin SGout
Scalability
High
throughput
Low latency
Disjoint
parallelism
Skew
resilience
Determinism
Prod
S
Prod
S
Prod
R Get next ready input tuple from SGin
1. compare t with all tuples in opposite window given P
2. add t to its window in a round robin fashion
3. remove stale tuples from t’s window
Steps for PUi
Agenda
• What is a stream join?
• Which are the challenges of a parallel stream join?
• Why ScaleJoin?
• How well does ScaleJoin addresses stream joins’
challenges?
• Conclusions
2015-10-31 21
Evaluation setup
• Common benchmark
• Implemented in Java
• Evaluation platform
– NUMA architecture: 2.6 GHz AMD Opteron 6230 (48 cores over 4
sockets), 64 GB of memory
– Architecture with Hyper Threading: 2.0 GHz Intel Xeon E5-2650 (16
cores over 2 sockets), 64 GB of memory
2015-10-31 22
t1
t2
t3
t4
t1
t2
t3
t4
R S
R: <timestamp,x,y,z> S: <timestamp,a,b,c,d>
P: a−10≤x≤a+10 AND b−10≤y≤b+10
ScaleJoin Scalability – comparisons/second
2015-10-31 23
Number of PUs
ScaleJoin latency – milliseconds
2015-10-31 24
Number of PUs
ScaleJoin skew-resilience
Constant distinct rates with peaks
2015-10-31 25
Agenda
• What is a stream join?
• Which are the challenges of a parallel stream join?
• Why ScaleJoin?
• How well does ScaleJoin addresses stream joins’
challenges?
• Conclusions
2015-10-31 26
Conclusions
• ScaleJoin: a Deterministic, Disjoint-Parallel and
Skew-Resilient Stream Join
• Challenges of parallel
stream joins
• Fine-grained synchronization (ScaleGate)
• 4 billion comparisons/second, with latency lower
than 60 milliseconds
Scalability
High
throughput
Low latency
Disjoint
parallelism
Skew
resilience
Determinism
2015-10-31 27
ScaleJoin: a Deterministic,
Disjoint-Parallel and Skew-Resilient
Stream Join
Vincenzo Gulisano, Yiannis Nikolakopoulos,
Marina Papatriantafilou, Philippas Tsigas
Thank you! Questions?
2015-10-31 28
1 of 28

Recommended

The benefits of fine-grained synchronization in deterministic and efficient ... by
The benefits of fine-grained synchronization in  deterministic and efficient ...The benefits of fine-grained synchronization in  deterministic and efficient ...
The benefits of fine-grained synchronization in deterministic and efficient ...Vincenzo Gulisano
581 views61 slides
Data Streaming (in a Nutshell) ... and Spark's window operations by
Data Streaming (in a Nutshell) ... and Spark's window operationsData Streaming (in a Nutshell) ... and Spark's window operations
Data Streaming (in a Nutshell) ... and Spark's window operationsVincenzo Gulisano
1.3K views31 slides
Tutorial: The Role of Event-Time Analysis Order in Data Streaming by
Tutorial: The Role of Event-Time Analysis Order in Data StreamingTutorial: The Role of Event-Time Analysis Order in Data Streaming
Tutorial: The Role of Event-Time Analysis Order in Data StreamingVincenzo Gulisano
151 views89 slides
The data streaming processing paradigm and its use in modern fog architectures by
The data streaming processing paradigm and its use in modern fog architecturesThe data streaming processing paradigm and its use in modern fog architectures
The data streaming processing paradigm and its use in modern fog architecturesVincenzo Gulisano
167 views64 slides
Crash course on data streaming (with examples using Apache Flink) by
Crash course on data streaming (with examples using Apache Flink)Crash course on data streaming (with examples using Apache Flink)
Crash course on data streaming (with examples using Apache Flink)Vincenzo Gulisano
255 views120 slides
20220201_semi dynamic STAQ application on BBMB.pptx by
20220201_semi dynamic STAQ application on BBMB.pptx20220201_semi dynamic STAQ application on BBMB.pptx
20220201_semi dynamic STAQ application on BBMB.pptxLuuk Brederode
60 views24 slides

More Related Content

Similar to ScaleJoin: a Deterministic, Disjoint-Parallel and Skew-Resilient Stream Join

Sequntial logic design by
Sequntial logic designSequntial logic design
Sequntial logic designPavan Mukku
841 views69 slides
Lic presentation by
Lic presentationLic presentation
Lic presentationMiguel Castaño
313 views39 slides
PosterFormatRNYF(1) by
PosterFormatRNYF(1)PosterFormatRNYF(1)
PosterFormatRNYF(1)Usman Khalid
45 views1 slide
Stream Loops on Flink - Reinventing the wheel for the streaming era by
Stream Loops on Flink - Reinventing the wheel for the streaming eraStream Loops on Flink - Reinventing the wheel for the streaming era
Stream Loops on Flink - Reinventing the wheel for the streaming eraParis Carbone
805 views37 slides
Flink Forward Berlin 2018: Paris Carbone - "Stream Loops on Flink: Reinventin... by
Flink Forward Berlin 2018: Paris Carbone - "Stream Loops on Flink: Reinventin...Flink Forward Berlin 2018: Paris Carbone - "Stream Loops on Flink: Reinventin...
Flink Forward Berlin 2018: Paris Carbone - "Stream Loops on Flink: Reinventin...Flink Forward
901 views37 slides
Practica1 digi2 by
Practica1 digi2Practica1 digi2
Practica1 digi2Juan Carlos Quispe Sano
46 views7 slides

Similar to ScaleJoin: a Deterministic, Disjoint-Parallel and Skew-Resilient Stream Join(20)

Sequntial logic design by Pavan Mukku
Sequntial logic designSequntial logic design
Sequntial logic design
Pavan Mukku841 views
Stream Loops on Flink - Reinventing the wheel for the streaming era by Paris Carbone
Stream Loops on Flink - Reinventing the wheel for the streaming eraStream Loops on Flink - Reinventing the wheel for the streaming era
Stream Loops on Flink - Reinventing the wheel for the streaming era
Paris Carbone805 views
Flink Forward Berlin 2018: Paris Carbone - "Stream Loops on Flink: Reinventin... by Flink Forward
Flink Forward Berlin 2018: Paris Carbone - "Stream Loops on Flink: Reinventin...Flink Forward Berlin 2018: Paris Carbone - "Stream Loops on Flink: Reinventin...
Flink Forward Berlin 2018: Paris Carbone - "Stream Loops on Flink: Reinventin...
Flink Forward901 views
Mba om 14_statistical_qualitycontrolmethods by Niranjana K.R.
Mba om 14_statistical_qualitycontrolmethodsMba om 14_statistical_qualitycontrolmethods
Mba om 14_statistical_qualitycontrolmethods
Niranjana K.R.152 views
ch08-Modeling & Simulation.ppt by LuckySaigon1
ch08-Modeling & Simulation.pptch08-Modeling & Simulation.ppt
ch08-Modeling & Simulation.ppt
LuckySaigon19 views
IRJET- Metastability Mitigation & Error Masking of High Speed Flip-Flop by IRJET Journal
IRJET- Metastability Mitigation & Error Masking of High Speed Flip-FlopIRJET- Metastability Mitigation & Error Masking of High Speed Flip-Flop
IRJET- Metastability Mitigation & Error Masking of High Speed Flip-Flop
IRJET Journal8 views
Keynote: Building and Operating A Serverless Streaming Runtime for Apache Bea... by Flink Forward
Keynote: Building and Operating A Serverless Streaming Runtime for Apache Bea...Keynote: Building and Operating A Serverless Streaming Runtime for Apache Bea...
Keynote: Building and Operating A Serverless Streaming Runtime for Apache Bea...
Flink Forward368 views
Dynamics of project-driven systems A production model for repetitive processe... by Ricardo Magno Antunes
Dynamics of project-driven systems A production model for repetitive processe...Dynamics of project-driven systems A production model for repetitive processe...
Dynamics of project-driven systems A production model for repetitive processe...
PERFORMANCE ANALYSIS OF D-FLIP FLOP USING CMOS, GDI, DSTC TECHNIQUES by IRJET Journal
PERFORMANCE ANALYSIS OF D-FLIP FLOP USING CMOS, GDI, DSTC TECHNIQUESPERFORMANCE ANALYSIS OF D-FLIP FLOP USING CMOS, GDI, DSTC TECHNIQUES
PERFORMANCE ANALYSIS OF D-FLIP FLOP USING CMOS, GDI, DSTC TECHNIQUES
IRJET Journal38 views
Process Mining Reloaded: Event Structures as a Unified Representation of Proc... by Marlon Dumas
Process Mining Reloaded: Event Structures as a Unified Representation of Proc...Process Mining Reloaded: Event Structures as a Unified Representation of Proc...
Process Mining Reloaded: Event Structures as a Unified Representation of Proc...
Marlon Dumas2.4K views
Javier Garcia - Verdugo Sanchez - Six Sigma Training - W2 Multi - vari Studies by J. García - Verdugo
Javier Garcia - Verdugo Sanchez - Six Sigma Training - W2 Multi - vari StudiesJavier Garcia - Verdugo Sanchez - Six Sigma Training - W2 Multi - vari Studies
Javier Garcia - Verdugo Sanchez - Six Sigma Training - W2 Multi - vari Studies
Self Managed and Automatically Reconfigurable Stream Processing - Vasiliki Ka... by Flink Forward
Self Managed and Automatically Reconfigurable Stream Processing - Vasiliki Ka...Self Managed and Automatically Reconfigurable Stream Processing - Vasiliki Ka...
Self Managed and Automatically Reconfigurable Stream Processing - Vasiliki Ka...
Flink Forward369 views
Self-managed and automatically reconfigurable stream processing by Vasia Kalavri
Self-managed and automatically reconfigurable stream processingSelf-managed and automatically reconfigurable stream processing
Self-managed and automatically reconfigurable stream processing
Vasia Kalavri638 views
Six sigma-in-measurement-systems-evaluating-the-hidden-factory (2) by Bibhuti Prasad Nanda
Six sigma-in-measurement-systems-evaluating-the-hidden-factory (2)Six sigma-in-measurement-systems-evaluating-the-hidden-factory (2)
Six sigma-in-measurement-systems-evaluating-the-hidden-factory (2)
DOE Full factorial by Monty Webb
DOE Full factorialDOE Full factorial
DOE Full factorial
Monty Webb1.1K views

Recently uploaded

Exploring the nature and synchronicity of early cluster formation in the Larg... by
Exploring the nature and synchronicity of early cluster formation in the Larg...Exploring the nature and synchronicity of early cluster formation in the Larg...
Exploring the nature and synchronicity of early cluster formation in the Larg...Sérgio Sacani
910 views12 slides
Effect of Integrated Nutrient Management on Growth and Yield of Solanaceous F... by
Effect of Integrated Nutrient Management on Growth and Yield of Solanaceous F...Effect of Integrated Nutrient Management on Growth and Yield of Solanaceous F...
Effect of Integrated Nutrient Management on Growth and Yield of Solanaceous F...SwagatBehera9
5 views36 slides
Study on Drug Drug Interaction Through Prescription Analysis of Type II Diabe... by
Study on Drug Drug Interaction Through Prescription Analysis of Type II Diabe...Study on Drug Drug Interaction Through Prescription Analysis of Type II Diabe...
Study on Drug Drug Interaction Through Prescription Analysis of Type II Diabe...Anmol Vishnu Gupta
26 views12 slides
ELECTRON TRANSPORT CHAIN by
ELECTRON TRANSPORT CHAINELECTRON TRANSPORT CHAIN
ELECTRON TRANSPORT CHAINDEEKSHA RANI
10 views16 slides
별헤는 사람들 2023년 12월호 전명원 교수 자료 by
별헤는 사람들 2023년 12월호 전명원 교수 자료별헤는 사람들 2023년 12월호 전명원 교수 자료
별헤는 사람들 2023년 12월호 전명원 교수 자료sciencepeople
58 views30 slides
Applications of Large Language Models in Materials Discovery and Design by
Applications of Large Language Models in Materials Discovery and DesignApplications of Large Language Models in Materials Discovery and Design
Applications of Large Language Models in Materials Discovery and DesignAnubhav Jain
13 views17 slides

Recently uploaded(20)

Exploring the nature and synchronicity of early cluster formation in the Larg... by Sérgio Sacani
Exploring the nature and synchronicity of early cluster formation in the Larg...Exploring the nature and synchronicity of early cluster formation in the Larg...
Exploring the nature and synchronicity of early cluster formation in the Larg...
Sérgio Sacani910 views
Effect of Integrated Nutrient Management on Growth and Yield of Solanaceous F... by SwagatBehera9
Effect of Integrated Nutrient Management on Growth and Yield of Solanaceous F...Effect of Integrated Nutrient Management on Growth and Yield of Solanaceous F...
Effect of Integrated Nutrient Management on Growth and Yield of Solanaceous F...
SwagatBehera95 views
Study on Drug Drug Interaction Through Prescription Analysis of Type II Diabe... by Anmol Vishnu Gupta
Study on Drug Drug Interaction Through Prescription Analysis of Type II Diabe...Study on Drug Drug Interaction Through Prescription Analysis of Type II Diabe...
Study on Drug Drug Interaction Through Prescription Analysis of Type II Diabe...
ELECTRON TRANSPORT CHAIN by DEEKSHA RANI
ELECTRON TRANSPORT CHAINELECTRON TRANSPORT CHAIN
ELECTRON TRANSPORT CHAIN
DEEKSHA RANI10 views
별헤는 사람들 2023년 12월호 전명원 교수 자료 by sciencepeople
별헤는 사람들 2023년 12월호 전명원 교수 자료별헤는 사람들 2023년 12월호 전명원 교수 자료
별헤는 사람들 2023년 12월호 전명원 교수 자료
sciencepeople58 views
Applications of Large Language Models in Materials Discovery and Design by Anubhav Jain
Applications of Large Language Models in Materials Discovery and DesignApplications of Large Language Models in Materials Discovery and Design
Applications of Large Language Models in Materials Discovery and Design
Anubhav Jain13 views
How to be(come) a successful PhD student by Tom Mens
How to be(come) a successful PhD studentHow to be(come) a successful PhD student
How to be(come) a successful PhD student
Tom Mens524 views
application of genetic engineering 2.pptx by SankSurezz
application of genetic engineering 2.pptxapplication of genetic engineering 2.pptx
application of genetic engineering 2.pptx
SankSurezz14 views
Small ruminant keepers’ knowledge, attitudes and practices towards peste des ... by ILRI
Small ruminant keepers’ knowledge, attitudes and practices towards peste des ...Small ruminant keepers’ knowledge, attitudes and practices towards peste des ...
Small ruminant keepers’ knowledge, attitudes and practices towards peste des ...
ILRI5 views
A giant thin stellar stream in the Coma Galaxy Cluster by Sérgio Sacani
A giant thin stellar stream in the Coma Galaxy ClusterA giant thin stellar stream in the Coma Galaxy Cluster
A giant thin stellar stream in the Coma Galaxy Cluster
Sérgio Sacani17 views

ScaleJoin: a Deterministic, Disjoint-Parallel and Skew-Resilient Stream Join

  • 1. ScaleJoin: a Deterministic, Disjoint-Parallel and Skew-Resilient Stream Join Vincenzo Gulisano, Yiannis Nikolakopoulos, Marina Papatriantafilou, Philippas Tsigas 2015-10-31 1 Chalmers University of technology
  • 2. Agenda • What is a stream join? • Which are the challenges of a parallel stream join? • Why ScaleJoin? • How well does ScaleJoin addresses stream joins’ challenges? • Conclusions 2015-10-31 2
  • 3. Agenda • What is a stream join? • Which are the challenges of a parallel stream join? • Why ScaleJoin? • How well does ScaleJoin addresses stream joins’ challenges? • Conclusions 2015-10-31 3
  • 4. Motivation Applications in sensor networks, cyber-physical systems: • large and fluctuating volumes of data generated continuously demand for: • Continuous processing of data streams • In a real-time fashion Store-then-process is not feasible!!! 2015-10-31 4
  • 5. What is a stream join? 2015-10-31 5 Data stream: unbounded sequence of tuples t1 t2 t3 t4 t1 t2 t3 t4 t1 t2 t3 t4 R S Sliding window Window size WS WSWR Predicate P
  • 6. Why parallel stream joins? • WS = 600 seconds • R receives 500 tuples/second • S receives 500 tuples/second • WR will contain 300,000 tuples • WS will contain 300,000 tuples • Each new tuple from R gets compared with all the tuples in WS • Each new tuple from S gets compared with all the tuples in WR … 300,000,000 comparisons/second! t1 t2 t3 t4 t1 t2 t3 t4 R S WSWR 2015-10-31 6
  • 7. Agenda • What is a stream join? • Which are the challenges of a parallel stream join? • Why ScaleJoin? • How well does ScaleJoin addresses stream joins’ challenges? • Conclusions 2015-10-31 7
  • 8. Which are the challenges of a parallel stream join? Scalability High throughput Low latency Disjoint parallelism Skew resilience Determinism 2015-10-31 8
  • 9. Agenda • What is a stream join? • Which are the challenges of a parallel stream join? • Why ScaleJoin? • How well does ScaleJoin addresses stream joins’ challenges? • Conclusions 2015-10-31 9
  • 10. The 3-step procedure (sequential stream join) For each incoming tuple t: 1. compare t with all tuples in opposite window given predicate P 2. add t to its window 3. remove stale tuples from t’s window Add tuples to S Add tuples to R Prod R Prod S Consume resultsConsPU 2015-10-31 10 We assume each producer delivers tuples in timestamp order
  • 11. The 3-step procedure, is it enough? Scalability High throughput Low latency Disjoint parallelism Skew resilience Determinism 2015-10-31 11 t1 t2 t1 t2 R S WSWR t3 t1 t2 t1 t2 R S WSWR t4 t3
  • 12. Enforcing determinism in sequential stream joins • Next tuple to process = earliest(tS,tR) • The earliest(tS,tR) tuple is referred to as the next ready tuple • Process ready tuples in timestamp order  Determinism PU tS tR 2015-10-31 12
  • 13. Deterministic 3-step procedure Pick the next ready tuple t: 1. compare t with all tuples in opposite window given predicate P 2. add t to its window 3. remove stale tuples from t’s window Add tuples to S Add tuples to R Prod R Prod S Consume resultsConsPU 2015-10-31 13
  • 14. Shared-nothing parallel stream join (state-of-the-art) Prod R Prod S PU1 PU2 PUN … Cons Add tuple to PUi S Add tuple to PUi R Consume results Pick the next ready tuple t: 1. compare t with all tuples in opposite window given P 2. add t to its window 3. remove stale tuples from t’s window Chose a PU Chose a PU Take the next ready output tuple Scalability High throughput Low latency Disjoint parallelism Skew resilience Determinism 2015-10-31 14 Merge
  • 15. Shared-nothing parallel stream join (state-of-the-art) Prod R Prod S PU1 PU2 PUN … 2015-10-31 15 enqueue() dequeue() ConsMerge
  • 16. From coarse-grained to fine-grained synchronization Prod R Prod S PU1 PU2 PUN … Cons 2015-10-31 16
  • 17. ScaleGate 2015-10-31 17 addTuple(tuple,sourceID) allows a tuple from sourceID to be merged by ScaleGate in the resulting timestamp-sorted stream of ready tuples. getNextReadyTuple(readerID) provides to readerID the next earliest ready tuple that has not been yet consumed by the former. https://github.com/dcs-chalmers/ScaleGate_Java
  • 18. ScaleJoin Prod R Prod S PU1 PU2 PUN … Cons Add tuple SGin Add tuple SGin Get next ready output tuple from SGout Get next ready input tuple from SGin 1. compare t with all tuples in opposite window given P 2. add t to its window in a round-robin fashion 3. remove stale tuples from t’s window 2015-10-31 18 SGin SGout Steps for PU
  • 19. 2015-10-31 19 t1 t2 R S WR t3 t4 R S t4 t1 WR R S t4 t2 WR R S t4 WR t3 Sequential stream join: ScaleJoin with 3 PUs: ScaleJoin (example)
  • 20. ScaleJoin Prod R Prod S PU1 PU2 PUN … Cons Add tuple SGin Add tuple SGin Get next ready output tuple from SGout 2015-10-31 20 SGin SGout Scalability High throughput Low latency Disjoint parallelism Skew resilience Determinism Prod S Prod S Prod R Get next ready input tuple from SGin 1. compare t with all tuples in opposite window given P 2. add t to its window in a round robin fashion 3. remove stale tuples from t’s window Steps for PUi
  • 21. Agenda • What is a stream join? • Which are the challenges of a parallel stream join? • Why ScaleJoin? • How well does ScaleJoin addresses stream joins’ challenges? • Conclusions 2015-10-31 21
  • 22. Evaluation setup • Common benchmark • Implemented in Java • Evaluation platform – NUMA architecture: 2.6 GHz AMD Opteron 6230 (48 cores over 4 sockets), 64 GB of memory – Architecture with Hyper Threading: 2.0 GHz Intel Xeon E5-2650 (16 cores over 2 sockets), 64 GB of memory 2015-10-31 22 t1 t2 t3 t4 t1 t2 t3 t4 R S R: <timestamp,x,y,z> S: <timestamp,a,b,c,d> P: a−10≤x≤a+10 AND b−10≤y≤b+10
  • 23. ScaleJoin Scalability – comparisons/second 2015-10-31 23 Number of PUs
  • 24. ScaleJoin latency – milliseconds 2015-10-31 24 Number of PUs
  • 25. ScaleJoin skew-resilience Constant distinct rates with peaks 2015-10-31 25
  • 26. Agenda • What is a stream join? • Which are the challenges of a parallel stream join? • Why ScaleJoin? • How well does ScaleJoin addresses stream joins’ challenges? • Conclusions 2015-10-31 26
  • 27. Conclusions • ScaleJoin: a Deterministic, Disjoint-Parallel and Skew-Resilient Stream Join • Challenges of parallel stream joins • Fine-grained synchronization (ScaleGate) • 4 billion comparisons/second, with latency lower than 60 milliseconds Scalability High throughput Low latency Disjoint parallelism Skew resilience Determinism 2015-10-31 27
  • 28. ScaleJoin: a Deterministic, Disjoint-Parallel and Skew-Resilient Stream Join Vincenzo Gulisano, Yiannis Nikolakopoulos, Marina Papatriantafilou, Philippas Tsigas Thank you! Questions? 2015-10-31 28

Editor's Notes

  1. Welcome…
  2. Agenda for this talk
  3. Since this is about stream joins, it is legitimate to start asking ourselves what is data streaming?
  4. This is the IEEE Big Data conference, so I get you heard before about Big Data What we know is the we have applications … and store-then-process is not always an option Traditional way DB … replaced by data streaming, main memory first the query than the data continuously
  5. So, why are we in need of scalable parallelization approaches for stream joins? Present example
  6. OK then, why Scalejoin? Let’s look in detail at what is the state of the art and what we did
  7. So, let’s see how stream joins are actually implemented Example But wait, what happens if we get tuples in another order? Mmmmhhh
  8. So, let’s see how stream joins are actually implemented Example But wait, what happens if we get tuples in another order? Mmmmhhh
  9. Ok, so deterministic 3-step procedure looks like this Now let’s try to parallelize this, and let’s do it as it has been done before
  10. Ok, so deterministic 3-step procedure looks like this Now let’s try to parallelize this, and let’s do it as it has been done before First, we need to do more operations, this affects the latency for sure But what’s worst is that we introduce a new bottleneck, the output thread and ready tuples And this actually breaks disjoint parallelism too… Finally, is not really skew-resilient So, what’s the problem? Are we doing it in the wrong way or are we forgetting something? Look at the data structures, we parallelize by parallelizing the computation, but what about the communication?
  11. The queues! We parallelized the computation, but overlooked the communication We are still using a queue with its methods enqueue and dequeue
  12. Let’s be creative, let’s assume they actually share something more powerful, that let’s them communicate and synchronize in a more efficient way What do we want from such communication and synchronization ds?
  13. Then we can do something like that…
  14. Here we can basically discuss why this addresses the different challenges, one by one… It gets even better, you can even have multiple physical producers for S and R!!!! And this is actually important because in the real world it will be like that!
  15. OK, so this is the benchmark we used… Implemented in Java And we evaluated it with 2 different systems (SAY WHY TWO SYSTEMS IF THEY ASK OR JUST SAY IT?)…
  16. Here we want to check the number of comparisons per second sustained by ScaleJoin After checking the ones obtained for a single thread, we computed the expected max and then observed ones for 3 different window sizes As you can see… Up to 4 billion comparison/second!
  17. This is the processing latency we get (in milliseconds) As you can see, even when we have 48 PUs (and notice that this means more threads than cores, since we have also injectors and receivers…) less than 60 – Actually, when we do not spawn too many threads we are talking of 30 milliseconds Might seem counterintuitive that latency grows with PU, but that’s because of determinism!
  18. In this case we have two different rates for R and S (notice actually the multiple physical streams!) and then peaks over time As you can see, comparisons of course increase when we have a peak, but nevertheless the overall work is very well balanced, the standard deviation among the Pus is less than 0.2% even during the spikes!!! Skew-resilience!