You are on page 1of 23

Year

2012
2012
2012
2012
2012
2012
2012

2012

2012

2012
2012

2012

2012

2012

2012

2012

2012

2012
2012

2012
2012

2013

2013
2013

2013

2013

2013

2013

2013

2013

2013

2013

2013

2013

2013
2013
2013

2014

2014

2014

2014

2014

possvel mesclar os trabalhos

2014

2014

2014

2014

2014

2014
2012
2012
2012
2012
2012
2012
2012
2012
2012
2012
2012
2012
2012

2012

2012

2013
2013
2013
2013
2013

2013
Description Module

scalable context delivery platform


Algorhitm to allocate CEP rules to EP and guarantee operation STATE Deployment Model
A Multi-Core Based Message Broker with QoS Support Core
Augmented Reality with CEP Receiver
Query rules based on Location-aware Rules
QoS Aware: Data streams based on user context Decider
Fault classification for event-based systems General
Middleware for integration heteregeneous data sources Receiver / Forward
Parallel processing on multi-core machines based on IBM Stream
Processing language Receiver / Decider
Execution model and an algorithm to coordinated reconfiguration
CEP operators Deployment Model
Analyse how to group subscribers with the same interest in order to Forward
reduce receiving/forwarding traffic Deployment Model
encrypt message routed in a publish subscriber system General
Using CEP system to detect Air Traffic Control events, and analyse
them against physiological data. Application
Propose a technique to analyse message dependence in a
Distributed Event-Based system General
Propose a semantic matching to work with heterogenous events in
an event-based systems Decider

Using IBM InfoSphere Streams platform to perform CDRs analyse Application


Seq / Producer
Framework to share stream processing STATE between operators Deployment Model
Framework to share STATE PARALLEL processing between STATEFUL Seq / Producer
operators Deployment Model

Improve Event processing system scalability, routing and operator


cost through Data flow. Reduce Energy, infrastructure usage. Deployment Model
Use of background knowledge to improve CEP Knowledge base
Cloud architecture to provide scalability in Interactive TV
applications Software Archicteture
Tutorial: CEP Patterns archicteture (High Level archicteture) Software Archicteture
Pattern recognition using multi-core processors and Finite State
Machine Decider

Stream Load Balance. Adjust streams as input based on stream


metadata and requirements.

Dynamic and adaptive stream management (split, monitoring Receiver


admission) to improve system scalability and QoS Deployment Model
Reduce costs at runtime and recovery time (improve performance)
related to fault-tolerance mechanism (avoid interruption).
Do not use replication or checkpoints, but the new concept of
Savepoints.
The proposed approach relies on the observation that at certain
points in time, the execution of an event processing operator solely
depends on a distinct selection of events from the incoming
streams. General
Understand and reduce operators provenance in order to minimize
the computational and storage overhead Deployment Model

Exploits the knowledge of event traffic, user subscriptions


and topology of the underlying physical network
to perform efficient routing in a publish/subscribe system. Similar to
DEBS 2012 number 23. Detect subscriber proximity to create a low
cost traffic. Forward

Proposes a middleware to obtaing high-throughput and scalable


pub/sub for clusters and cloud deployment. Based on the principles Deployment Model
of MapReduce, a new architecture is proposed. Software Architecture

Pub/Sub system that provides efficient scalable event notifications


for DYNAMIC subscriptions based on Grouping users whenever the
users changes (Example of Earthquake).
Forward
Se DEBS 2012 number 23 and 2013 number 5 Software Architecture

use clustering as well as sampling techniques


to select the stream data to be processed, that will enable
us to identify core real-world events of interest over time in
an accurate and efficient manner. Receiver
Bring the concept of COMPLEX ACTIONS with temporal
dependencies into the CEP System Producer
Proposes an approach to cope with out-of-order events mixing
Buffering and speculative techniques Decider ?
Analysing, improve and create a scheduler to a specific Stream Core
language (Use StreamIT language) Deployment Model

Event enrichment are external to event processing


engines. This paper presents a unified model for event enrichment
that takes place
natively within the event processing engine Before Receiver...

Mobile CEP - Try something similar to Migration VM to Mobile


networks to reduce latency and network utilization (placement and
migration method).
Migration user information near user location
FOG resource Core

Our system predicts future query regions for moving consumers and
starts processing events early so that the live situational information
is available when the consumer reaches the future location. Core
Storm: Static and dymanic approach to define/deploy a topology Deployment Model
Using Probabilistic Data Association to make decision faster Decider

An approach for general quality-aware sensor data processing


that can be used for many applications.
uniform meta data model to represent sensor data and information
quality at all levels of processing Receiver

a model, which estimates the latency


spike created by a set of operator movements
- built a latency-aware elastic operator placement algorithm Receiver

Processing stream through cloud computing (improving


performance). Propose a set of strategy for efficient transfers Receiver

Common vocabulary represented by a TEXT domain event model as


well as a reference architecture for text stream processing and
information filtering Receiver

iCEP, a novel framework that learns, from historical traces, the


hidden causality between the received events and the situations to
detect, and uses them to automatically generate CEP rules (detect
new rules based user defined rules) Rules

Reuse computations and this way reduces the resource


requirements of mobile CEP.
1- User in same area
2- Situation awareness applications do not require a
perfectly accurate set of input data to generate meaningful
situational information (ex. Speed of ten vehicles) Core

A complex event service description model that extends


OWL-S and captures the exact semantics of complex
events, including their patterns and attributes Rules

A DSL for CEP system Rules

Using aggregation to reduce data traffic from publishers and to


subscribers Deployment Model

Automated Qos Aware configuration (Design time) of Pub/Sub


based on domain, event type try to define some techniques that
will guarantee Quality of Service

It is a framework to automated design time configuration Deployment Model


A new protocol to discovery pub/sub operators (overcoming the
current limitations with DDS SDP) Deployment Model
DEBS Grand Challenge
DEBS Grand Challenge
DEBS Grand Challenge
Odysseus: DEBS Grand Challenge
DEMO:
Poster:
Poster:
Poster:
Poster:
Poster:
Demo:
Demo:
Demo:

Industry:
AdScorer: Improve advertisement through channel change(zap)
analisys Industry / Receiver / Decider
Industry:
BMPS and CEP integration Industry / Receiver

Industry: Proposes a system to be executed in a high-tech


manufacturing
equipment to detect anamolias/error pattern using raw data from
its sensors (help engineers analyses) Industry
Industry: Application to be proactive in transportation area Industry
Industry: validate data streams through a complex Model Industry
Industry: Analysing data from a mobile network Industry
Industry: Analysing data from Spotify system Industry

Intrusion detection through different event pattern analysing

See HUGO: 2013 08 (Industry) Industry


Keywords Country

Scalability, EP management, Deployment Model Japan


IoT, Publish/subscriber, Multi-core, Hardware, QoS China
Augmented Reality (AR), CEP, Context-aware Germany
WSN, Location-aware, algorithm Korea
QoS-Aware, context-aware, System Architecture Greece
Taxanomy, classification, fault, error Austria
JMS, MOM, Integration, Transformation, Heterogeneous data Germany

Parallel stream processing, multi-core, cluster, Partitiong USA

Deployment, reconfiguration, context-aware, CEP, algorithm Germany

Content-based, publish/subscriber, P2P, Spectral Graph Teory, Routing Germany


Privacy, Publish/subscriber, cryptography Switzerland

Human Performance, CEP, speficic scenario as Data source (stream) USA

Message dependence, Flow Analysis USA

Event matching, Semantic decoupling Ireland

Telecom (CDRs), IBM InfoSphere, Specific scenario Ireland

Distributed Shared State, Fault Tolerante, Scalability Switzerland

Parallelization, Shared State, Synchronization China

Stream Processing, Software Stack, Scalability, Static reconfiguration Greece


Semantic, Knowledge base Germany

Architecture, iTV, STB, Cloud, Middleware, publish/subscribe, scalability China


Architecture, Pattern Germany

FSM, Multi-core, Parallelism, CEP, scalability Switzerland

Scalability, Adaptative Stream management, Parallelism Switzerland


Fault Tolerance, checkpoints, shared state Germany

Provenance, Instrumentation, provenance-aware, Routing USA

Routing, Subscriber group, network Germany

Architecture, Publish/Subscribe, scalability, performance Switzerland

Architecture, Publish/Subscribe, scalability, performance USA

Crows, Mobile Systems, Event Detection, GPS Greece

Composite Actions, Temporal Dependence, Semantic Analysis Germany

CEP, Out-of-order, Publish/Subscribe, Fault-Tolerance Germany

Stream Language, Scheduler USA

Event Enrichment, Decider / Rules Ireland

Migration, VMs, Mobile Network, Fog resource Germany

Prediction, future Location, Location-aware USA


CEP, Static and dymanic deployment model Italy
Probabilistic Data Association, similar functions USA

Data Stream processing


- Raw sensor Data up to the derived information Germany

Distributed data stream processing, Load balance, Elasticity Germany

Stream processing, cloud computing, multiple data center Germany

Event processing, TEXT analysis... Germany

CEP, Rule generation, learning Netherlands

Mobility, Processing Reuse, Reduce latency Germany

Ireland

DSL, CEP, M2M Germany

Pub/Sub, aggreation, distributed adaptation Norway

Qos-Aware, Design-time configuration, publish/subscribe Germany

Discovery, Data Distribution Service (DDS), Pub/Sub, P2P USA


USA
Germany
Scotland
Germany

TV, Channel change 'score' Norway

Classification, integration BMPS, CEP Germany

Application, Monitoring data Germany


Application, Monitoring data, proactive Israel
Application, Validate Data Sweden
Application, Monitoring data USA
Norway

USA
Issue Return Priority

1
1
1

10

15

6
9

Improve performance

Improve performance 5
Overhead caused by reability in fail-free execution (run-time
concerning computational resources, event traffic, and event
detection time) 5

Overhead in provenance "metrics" 4

Reliabity to a specific filtering scheme for implementig


scalability 3

Performance 4

Subscribers: time-shortly and dynamically changes (users


location) 1

mobile devices often produce more


data than the network can deliver or the system can
process, but only a event subset can be sufficient.
How to detect correct Stream sample to analyse it. 6
Isolated actions can be a problema (Smoke and evacutaion
routes example) 1

Out of order events - high latency vs. Errors 6


Static vs. Dymanic stream processing
transfer rate vs. Optimization

Enrichment event process is difficult in a larga-scale


heterogenous system

High latency in Mobile CEP (between user after his changing


and fog node)

Users change their location frequently. Various processing


are wasted computation because user has left. How to be
oportunistc in this context?
Improve Storm scheduler (how to define current
infrastructure topology)
Improve events association

Sensor data is prone to errors.


Apps ignore it.
Apps are specific to one scenario (not reused). 3

Elastic Scaling comes with a latency penalt 3

Performance in stream processing 5

Stream processing widely used in areas like logistics,


financial, health care.. But not as TEXT streams (like twitter /
social networks) 10

Try to observe others events information to generate new


rules based on user defined rules

For each mobile consumer an individual set


of operators needs to be maintained and adapted to ensure
low latency paths (many resources). This massive balance
can reduce resource-efficiency in performing consumer-
centric CEP 2

Current service models are not suitable for describing


complex events and the requirements for enabling pattern-
based complex event service composition are not fully
addressed 7
Specify queries in CEP system is not in high level, e existents
EPL are not domain specific 7

Some pub/sub system can receive and delivery data from/to


many nodes. It can be reduced. Ex. Car traffic (speed can be
calculated as one segment) 4

Middleware configuration is manual and done by


developers. (prone to error) 3
Discovery publisher/subscriber in a large system it not
efficient. (SDP - protocol to discovery) 5
10

4
Title / File

13-A scalable complex event processing system and evaluations of its performance
14-Pub_Sub on stream_a multi-core based message broker with QoS support
15-An approach for using complex event processing for adaptive augmented reality in cultural heritage domain-experience rep
16-Location-aware event-driven query processing in sensor database management
17-Dynamic QoS-aware event sampling for community-based participatory sensing systems
19-Deriving a unified fault taxonomy for event-based systems
20-ACTrESS automatic context transformation in event-based software systems

21-Partition and compose_parallel complex event processing

22-Moving range queries in distributed complex event processing

23-Distributed spectral cluster management a method for building dynamic publish subscribe systems
24-Thrifty privacy_efficient support for privacy-preserving publish_subscribe

25-Characterizing human performance and potential for injury in air traffic control using complex event processing

26-Impact analysis for distributed event-based systems

27-Approximate semantic matching of heterogeneous events

28-Processing 6 billion CDRs day - from research to production (experience report)

29-CAPSULE language and system support for efficient state sharing in distributed stream processing systems

30-Parallelizing stateful operators in a distributed stream processing system how, should you and how much

31-Understanding and improving the cost of scaling distributed event processing


32-Fusion of background knowledge and streams of events

33-A highly efficient cloud-based architecture for large-scale STB event processing industry article
34-Tutorial on advanced design patterns in event processing

01-RIP run-based intra-query parallelism for scalable complex event processing

02-Adaptive input admission and management for parallel stream processing


03-Rollback-recovery without checkpoints in distributed event processing systems

04-Ariadne managing fine-grained provenance on data streams

05-Efficient content-based routing with network topology inference

06-StreamHub a massively parallel architecture for high-performance content-based publish subscribe

07-DYNATOPS a dynamic topic-based publish_subscribe architecture

12-Efficient event detection by exploiting crowds

13-Towards complex actions for complex event processing

14-Reliable speculative processing of out-of-order event streams in generic publish_subscribe middlewares

15-Dynamic expressivity with static optimization for streaming languages

16-Towards unified and native enrichment in event processing systems

17-MigCEP operator migration for mobility driven distributed complex event processing

18-Opportunistic spatio-temporal event processing for mobile situation awareness


19-Adaptive online scheduling in storm
20-Real-time probabilistic data association over streams

01-STREAM_Quality matters_supporting quality-aware pervasive applications by probabilistic data stream management

02-STREAM_Latency-aware elastic scaling for distributed data stream processing systems

03-STREAM_JetStream_enabling high performance event streaming across cloud data-centers

04-STREAM_An event processing approach to text stream analysis_basic principles of event based information filtering

05_EVENT_PROCESSING_Learning from the past_automated rule generation for complex event processing

06_EVENT_PROCESSING_RECEP selection-based reuse for distributed complex event processing

07_EVENT_PROCESSING_Complex event service provision and composition based on event pattern matchmaking

08-EVENT_PROCESSING_DS-EPL domain-specific event processing language

09-PUB_SUBS_FILTER_Distributed event aggregation for content-based publish-subscribe systems

11-PUB_SUBS_FILTER_Automated quality-of-service-aware configuration of publish-subscribe systems at design-time

12-PUB_SUBS_FILTER_Content-based filtering discovery protocol (CFDP) scalable and efficient OMG DDS discovery protocol
35-Solving manufacturing equipment monitoring through efficient complex event processing DEBS grand challenge
36-Where event processing grand challenge meets real-time web PLAY event marketplace
37-Glasgow automata illustrated DEBS grand challenge
38-Odysseus as platform to solve grand challenges DEBS grand challenge
39-Context-model generation for safe autonomous transport vehicles
40-Odysseus a highly customizable framework for creating efficient event stream management systems

43-Notifying medical data in health information systems

47-Data Stream Application Manager (DSAM)

09-AdScorer_an event-based system for near real-time impact analysis of television advertisements

18-Integrating CEP and BPM how CEP realizes functional requirements of BPM applications (industry article)

08-HUGO real-time analysis of component interactions in high-tech manufacturing equipment (industry article)
09-Proactive event processing in action a case study on the proactive management of transport processes (industry article)
10-Model-based validation of streaming data(industry article)
11-Mobile QoS management using complex event processing (industry article)
21-The hidden pub_sub of spotify (industry article)

22-Event stream database based architecture to detect network intrusion (industry article)
n-experience report
nagement

n filtering

overy protocol
dustry article)

You might also like