Mastering Intelligent Clouds

1
Mastering Intelligent Clouds

• Engineering Intelligent Data Processing Services
  in the Cloud

Sergiy Nikitin, Industrial Ontologies
Group, University of Jyväskylä, Finland
Presented at ICINCO 2010 conference Funchal,
Madeira
2
Contents

• Background on Cloud Computing
• Extending cloud computing stack
• UBIWARE platform
• Data Mining services in the Cloud
• Conclusions

3
Cloud Computing already on the market

• SalesForce.com (SFDC)
• NetSuite
• Oracle
• IBM
• Microsoft
• Amazon EC2
• Google
• etc.
• (for a complete survey see Rimal et al., 2009)

4
Cloud Computing stack
Cloud computing stack
Application as a Service
SaaS
Application (business logic)
Services (Payment, Identity, Search)
Solution stack (Java, PHP, Python, .NET)
What add-value can we offer to the PaaS level?
PaaS
Structured storage (e.g. databases)
Raw data storage and network
OS-virtualization
IaaS
Virtualization Machine
Hardware configuration
5
Autonomic Computing

• A vision introduced by IBM in 2003 (Kephart et
  al.)
• software components get a certain degree of
  self-awareness
• self-manageable components, able to run
  themselves
• Why?
• To decrease the overall complexity of large
  systems
• To avoid a nightmare of ubiquitous computing
  an unprecedented level of complexity of
  information systems due to
• drastic growth of data volumes in information
  systems
• heterogeneity of ubiquitous components,
  standards, data formats, etc.

6
Intelligence as a Service in the cloud
Agent-driven service API
Services (Payment, Identity, Search)
Configuration management
Data adaptation
Solution stack (Java, PHP, Python, .NET)
PaaS
Intelligent services
Solution stack
Structured storage (e.g. databases)
Domain models

• Smoothly integrate with the infrastructure
• Build stack-independent solutions
• Automate reconfiguration of the solutions

7
UBIWARE platform
UBIWARE Agent
Beliefs storage
Role Script
Data
.class
RAB
RAB
RAB
RAB
Blackboard
8
API extension OS perspective
Cloud Platform Provider
Virtual machine
PCA
PMA
SW Platform
Customer applications and services
PCA Personal Customer Agent
PMA Platform Management Agent
9
Data Adaptation as a Service
Cloud Platform Provider
Virtual machine
PMA
SW Platform
PCA
Data Service
Customer applications and services
Files
Adapter Agent
DB/KB
PCA Personal Customer Agent
PMA Platform Management Agent
10
Platform-driven service execution in the cloud
Cloud Platform Provider
Virtual machine
Virtual machine
Service execution environment
SW Platform
PCA
Customer applications
PMA
API
PCA Personal Customer Agent
PMA Platform Management Agent
11
Agent-driven PaaS API extension
Agent-driven flexible intelligent service API
Agent-driven Adapters
Smart data source connectivity
Configurable data transformation
User applications in cloud
Proactive adapter management
Agent-driven intelligent services
Configurable model
Service mobility
Proactive self-management
Smart cloud stack
Stack control and updates
Failure-prone maintenance
Embedded and remote services
Smart Ontology
Domain models
Standards compatibility
System configuration and policies
12
Intelligent services PaaS API extension
Agent-driven flexible intelligent service API
User applications in cloud
Agent-driven intelligent services
Configurable model
Service mobility
Proactive self-management
13
Agent-driven data mining services

• Data mining applications are capabilities
• Agents can wrap them as services
• PMML language - a standard for DM-model
  representations
• Data Mining Group. PMML version 4.0. URL
  http//www.dmg.org/pmml-v4-0.html

14
PMML data mining model descriptions
PMML model
Header
Version and timestamp
Model development environment information
Data dictionary
Definition of variable types, valid, invalid and
missing values
Data Transformations
Normalization, mapping and discretization
Data aggregation and function calls
Model
Description and model specific attributes
Mining schema
Definition of usage type, outlier and missing
value treatment and replacement
Targets
Score post-processing - scaling
Definition of model architecture/parameters
PMML - Predictive Model Markup Language
(www.dmg.org/pmml-v3-0.html)
15
Data mining service types
Fixed model service
Model player service
Model construction service
16
A use case for data mining service stack

• A Web of Intelligence case

Input
Model
Output
Pattern of learning data to be collected ?V?p1,
?p2, ?p3
Distributed query planning and execution
A set of learning samples (vectors)
1
Learning samples and the desired model settings
Model M1 parameters
Model constructor
2
Set up a model M1
Model player
Model M1 assigned
3
Paper machine alarms classifier neural network
model (M1)
Vector class of V1 is Urgent Alarm according
to model M1
Vector to be classified alarm message V10.785,
High, node_23
4
17
Data Mining services in UBIWARE
Ontology construction
Data Mining service
Model construction service
Computational service
Fixed model service
Model player service
Data mining domain
Core DM service ontology
Problem domain
18
UBIWARE in cloud computing stack
Cloud computing stack
Example application
UBIWARE for control and management in cloud
Semantic Business Scenarios
DM model wrapped as a service for paper industry
Application as a Service
Domain-specific components as services
Applications and Software as a Service
Application (business logic)
Domain model (Ontology) components
DM model for paper industry
Services (Payment, Identity, Search)
Cross-domain Middleware components
Componentization Servicing
Data Mining service player
Platform as a service
Solution stack (Java, PHP, Python, .NET)
Cross-layer configuration management mechanisms
Connectors, Adapters
Agent-driven service API
Structured storage (e.g. databases)
RABs, Scripts
Raw data storage and network
OS-virtualization
Infrastructure as a service
Virtualization Machine
Hardware configuration
Technologies in cloud
19
Conclusions

• Web intelligence as a cloud service
• Ubiware is a cross-cutting management and
  configuration glue
• Advanced data adaptation mechanisms as cloud
  services
• A competitive advantage for cloud providers
• Seamless data integration for service consumption
  and provisioning
• Autonomous agents as a Service (A4S)
• Supply any resource with the autonomous manager

20
Thank you!