Event Processing as a Service

Event Processing as a Service
Bob Marcus
bobmarcus1@comcast.net

 Cloud Computing (from Gartner Group) = “a style of
computing where massively scalable IT-enabled
capabilities are delivered 'as a service' to external
customers using Internet technologies”… The key
activity will be to determine which cloud services will be
viable, and when.
3
Cloud Definition

Service Oriented Levels
4
From OGF
Middleware
Architecture
Cloud

Cloud Computing Services Layers
Application Services (e.g. Software as a Service = SaaS)
Middleware and Data Services (e.g. Database as a Service)
Development and Deployment Services (e.g. Platform as a Service)
Resource Services (e.g. Infrastructure as a Service)
Computing Hardware and Physical Storage
Event, Processes, Interfaces (e.g. Event Processing as a Service)
Distributed Virtualization Services

Service-based Cloud Computing Architecture
Cloud-based Services
e.g. databases, queues
Software
as a Service
Platform
as a Service
Infrastructure
as a Service
Browser and
Internet Clients
Cloud Servers
Cloud Tools
e.g. development, monitoring
configuration
Cloud Middleware Services
e.g. dynamic allocation, security
Virtualized Resources Cloud
Applications and Services
Cloud Storage
and Data Sources

Enterprise Cloud Computing Group
 An Enterprise Cloud Computing Group was initiated
September 17 at an NCOIC.org meeting including industry,
government, standards groups, and vendors
 The goal of the Group will be to exchange information
about standardizations, critical requirements, best
practices, methodologies, success stories, and emerging
technologies
 There is a planned follow-on meeting at an Open Group
Conference in San Diego, February 2-5, 2009
7

Event Processing as a Service
 Complex event processing (CEP) is a key component for
many application areas (e.g. finance, supply chain,
intelligence, counter-terrorism, situation awareness)
 Required resources for CEP may fluctuate greatly over
time (e.g. emergency response)
 Deploying “Event Processing as a Service” on a Cloud
infrastructure could support dynamic requirements
 “Event Processing as Service” needs standardizations:
 Describe event patterns of interest
 Specify quality of services for reporting events
 Identifying event sources to be monitored

Event Processing Using Multiple Services
 Remote Monitoring as a Service
 Selection of sensing source and data feeds to monitor
 Ability to turn on sensing sources or data feeds
 QoS as a Service
 Event transmission and processing as a Service using priorities
 Middleware support (e.g. OMG Data Distribution Services)
 Network and compute resources to be applied
 Pattern Detection (CEP)
 Computing as a Service to obtain resources
 Software as a Service to utilize specific applications (e.g. CEP)

Select Sources Select Priorities Select Patterns Event Processing as a
Service (My inserts)

Map Reduce(Complex Event Processing)
Event ID,
Event Attributes
Pattern ID, Partial Match
Pattern ID,
Complete Match
by Pattern ID

Monitoring Data Sources as a Service
 Monitoring as a Service - is the provision of monitoring
(e.g. network monitoring, website monitoring, backup,
security, and compliance monitoring) as a service
 Monitoring as a service providers include:
 Redhat Command Center (Enterprise Systems)
 Hyperic CloudStatus (Amazon Web Services, Google App Engine)
 TrustSaaS (SaaS alerting and reporting)
 Just Uptime (Web site monitoring)
 Some Other Large Scale Sensing Projects
 Intel IrisNet (Internet-scale Resource-Intensive Sensor Network Service)
 Microsoft SenseWeb
 Sensor Web from GEOSS (Global Earth Observation System of Systems)

QoS as a Service
From RTI
Subscribe Publish
Future: Messaging Switch (e.g. Tervela)

15
QoS Policies Supported by OMG DDS
 Data-Centric Publish-Subscribe entities (e.g., topics, data
readers/writers) configurable via QoS policies
 QoS tailored to data distribution in tactical information systems
• DEADLINE
• Establishes contract regarding
rate at which periodic data is
refreshed
• LATENCY_BUDGET
• Establishes guidelines for
acceptable end-to-end delays
• TIME_BASED_FILTER
• Mediates exchanges between
slow consumers & fast producers
• RESOURCE_LIMITS
• Controls resources utilized by
service
• RELIABILITY (BEST_EFFORT,
RELIABLE)
• Enables use of real-time
transports for data
• HISTORY (KEEP_LAST,
KEEP_ALL)
• Controls which (of multiple)
data values are delivered
• DURABILITY (VOLATILE,
TRANSIENT, PERSISTENT)
• Determines if data outlives
time when they are written
…

CEP as a Service (Description Language)
Events Events
“Event Processing as a
Service” Engine
Complex Event
Description Language
Representation
Query Patterns
(possible standard)
Complex Events
(possible standard)
Complex Event
Processing Engine 1
(e.g. Rule-based)
Complex Event
Processing Engine 2
(e.g. SQL Extension)
Mapping to
CEP Engine 1
Patterns
Mapping from
CEP Engine 1
Events
Mapping to
CEP Engine 2
Patterns
Mapping from
CEP Engine 2
Events
Event Metamodel
(OMG standard)
Outputs to
external systems
Semantics
Input Patterns

Final Predictions
 In the next decade, there will be multiple Cloud
Event Processing Services (individual, enterprise)
 This will be a necessity for filtering the deluge of
information that will flood current resources
 These Services will input data sources, desired
level of service, and prioritized patterns of interest
 Output will be produced and delivered in an easy-
to-use structured format
 Data collected from the Event Processing
Services will provide the base for future advances
e.g. http://googleblog.blogspot.com/2008/09/intelligent-cloud.html

Event Processing as a Service

More Related Content

What's hot (19)

Similar to Event Processing as a Service (20)

Recently uploaded (20)

Event Processing as a Service