This document summarizes Rafael Brundo Uriarte's doctoral thesis on supporting autonomic management of clouds. The thesis addresses service-level agreements (SLAs), cloud monitoring, and similarity learning. It presents the SLAC language for defining SLAs, the Panoptes monitoring framework, and a Random Forest and PAM approach (RF+PAM) for learning service similarities. The Polus framework integrates these contributions to provide knowledge for autonomic cloud management.
POGONATUM : morphology, anatomy, reproduction etc.
Supporting Autonomic Management of Clouds: Service-Level-Agreement, Cloud Monitoring and Similarity Learning
1. Supporting Autonomic Management of
Clouds: Service-Level-Agreement, Cloud
Monitoring and Similarity Learning
by
Rafael Brundo Uriarte
rafael.uriarte@imtlucca.it
Under the Supervision of:
Prof. Rocco De Nicola and Prof. Francesco Tiezzi
Doctoral Thesis Defense - March 30th, 2015 - Lucca, Italy
8. Knowledge for the Self-Management
Policies
Service Definition and Objectives
Status of the Cloud and Services
Specific Knowledge
Introduction Rafael Brundo Uriarte 7/51
10. Research Questions
Research Question 1
How to describe services and their objectives in the cloud
domain?
Research Question 2
What is data, information, knowledge and wisdom in the
autonomic cloud domain?
Research Question 3
How to collect and transform operational data into useful
knowledge without overloading the autonomic cloud?
Introduction Rafael Brundo Uriarte 9/51
11. Research Questions
Research Question 4
How to produce a robust measure of similarity for services in
the domain and how can this knowledge be used?
Research Question 5
How to integrate different sources of knowledge and feed the
autonomic managers?
Introduction Rafael Brundo Uriarte 10/51
14. SLA for Cloud Computing - SLAC
Domain Specific
Multi-Party
Deployment Models
Formalism
Ease-of-Use
SLA for Clouds Rafael Brundo Uriarte 13/51
15. Yet Another SLA Definition Language?
Features WSOL WSLA SLAng WSA SLA* SLAC
General
Deployment Models
Broker Support - - - - -
Business
Pricing Schemes -
Formal
Semantics - - - -
Verification - - - -
“ ” feature covered
“ ” feature partially covered
“-” no support
SLA for Clouds Rafael Brundo Uriarte 14/51
16. Main Concepts
Predefined Metrics - Involved Parties and Unit
Intervals for Metrics - Template and Variations
Groups - Multiple Service, Community Cloud
Constraint Solving Problem
SLA for Clouds Rafael Brundo Uriarte 15/51
18. Business Aspects
Business Actions
Flat and Variable Models
Pricing Schemes - Exchange, Auction, Tender,
Bilateral, Fixed, Posted
SLA for Clouds Rafael Brundo Uriarte 17/51
19. Implementation
Editor for SLAs (Ecplise-based using Xtext)
SLA Evaluator (Z3 Solver)
Integration with the Monitoring System
SLA for Clouds Rafael Brundo Uriarte 18/51
21. DIKW in the Domain
Data
Information
Knowledge
Wisdom
Cloud Monitoring Rafael Brundo Uriarte 20/51
22. Cloud Monitoring
The Role of the Monitoring System in Clouds:
Collect data and Provide Information and
Knowledge
No Wisdom - Related to Decision-Making
Sensor of MAPE-K Loop
Cloud Monitoring Rafael Brundo Uriarte 21/51
23. Related Works
Property PCMONS Monalytics Lattice Wang
Cloud - - -
Autonomic Integration - - - -
Scalability -
Adaptability - -
Resilience - - - -
Timeliness - -
Extensibility
“ ” feature covered
“ ” feature partially covered
“-” no support
Cloud Monitoring Rafael Brundo Uriarte 22/51
24. Panoptes
Multi-agent system
Monitoring in different levels
Monitoring Modules - What needs to be
monitored and how to process the data
Cloud Monitoring Rafael Brundo Uriarte 23/51
30. Specific Knowledge
Generation of Knowledge for a Specific Purpose, i.e.
not applicable in all clouds. For example, similarity.
But what is similarity?
How much an object (service) resembles other
Similarity Learning Rafael Brundo Uriarte 29/51
31. Applications of Similarity
Cluster Services:
Group Similar Services
Different Algorithms
(K-Means, PAM, EM)
Applications in the Domain:
Anomalous Behaviour Detections
Service Scheduling
Application Profiling
SLA Risk Assessment
Similarity Learning Rafael Brundo Uriarte 30/51
33. Random Forest
Clustering with Random Forest
Originally Developed for Classification
Calculate the Similarity
Clustering Algorithm (PAM)
Similarity Learning Rafael Brundo Uriarte 32/51
35. Problems
Similarity Matrix (Big Memory Footprint)
Re-cluster on Every New Observation
Cannot be Used in the Domain
Similarity Learning Rafael Brundo Uriarte 34/51
38. Experiments
Compared the performance of our algorithm to
other 2 methodologies
Compared the performance of RF+PAM with
the standard off-line similarity learning
Use Case:
Scheduler deploys together the most
dissimilar services
Similarity based on their SLAs
Similarity Learning Rafael Brundo Uriarte 37/51
46. Research Questions
Research Question 1
How to describe services and their objectives in the cloud
domain?
SLAC
Research Question 2
What is data, information, knowledge and wisdom in the
autonomic cloud domain?
DIKW Hierarchy
Research Question 3
How to collect and transform operational data into useful
knowledge without overloading the autonomic cloud?
Panoptes
Conclusions Rafael Brundo Uriarte 45/51
47. Research Questions
Research Question 4
How to produce a robust measure of similarity for services in
the domain and how can this knowledge be used?
RF+PAM
Research Question 5
How to integrate different sources of knowledge and feed the
autonomic managers?
Polus Framework
Conclusions Rafael Brundo Uriarte 46/51
49. Contributions
A theoretical and practical framework for the
generation and provision of knowledge for the
autonomic management of clouds (Polus
Framework):
SLAC - SLA Definition and Evaluation
Panoptes - Monitoring
RF+PAM - Similarity Learning
Conclusions Rafael Brundo Uriarte 48/51
50. Publications
1. R. B. Uriarte, S. Tsaftaris and F. Tiezzi. Service Clustering for
Autonomic Clouds Using Random Forest. In Proc. of the 15th
IEEE/ACM CCGrid [In Press], 2015.
2. R.B. Uriarte, F. Tiezzi, R. De Nicola, SLAC: A Formal
Service-Level-Agreement Language for Cloud Computing. In IEEE/ACM
7th International Conference on Utility and Cloud Computing (UCC),
2014.
3. R.B. Uriarte, C.B. Westphall, Panoptes: A monitoring architecture and
framework for supporting autonomic Clouds, In Proc. of the 16th
IEEE/IFIP Network Operations and Management Symposium (NOMS),
2014.
4. R.B. Uriarte, S.A. Chaves, C.B. Westphall, Towards an Architecture for
Monitoring Private Clouds. In IEEE Communications Magazine, 49,
pages 130-137, 2011.
Conclusions Rafael Brundo Uriarte 49/51
55. SLAC - Cloud Metrics
DTMF Cloud Computing Service Metrics
Description
Recent Document (Still a Draft)
Creation of a Model for the Definition of Metrics
The SLAC Metrics can be Adapted for this
Model
Conclusions Rafael Brundo Uriarte 51/51
56. SLAC Violation
Violation and Penalty are Separated Concepts
“Violation” Concept Flexible
Easy to Understand
Conclusions Rafael Brundo Uriarte 51/51
57. Panoptes - Scalability
Designed to be scalable
Adapt itself
Experiments suggest it is scalable
More experiments for future works
Conclusions Rafael Brundo Uriarte 51/51
58. Panoptes - Analysis of Apache Broklyn
Not Focused on Monitoring
Does Not Process the Data
Conclusions Rafael Brundo Uriarte 51/51
59. Panoptes - Analysis with CSPARQL
Data is not Decorated (e.g. RDF)
Impact of Decorated Monitoring Data
(Scalability)
Very Interesting Option
Conclusions Rafael Brundo Uriarte 51/51