Title: MarketOriented Cloud and Grid Computing: A Vision, Hype and Reality of Delivering IT services as Com
1Market-Oriented Cloud (and Grid) Computing A
Vision, Hype and Reality of Delivering IT
services as Computing Utilities
Grid Computing and Distributed Systems (GRIDS)
LabDept. of Computer Science and Software
EngineeringThe University of Melbourne,
Australiawww.gridbus.org
Gridbus Sponsors
2The GRIDS Lab _at_ Melbourne
Education
R D
- Youngest and one of the rapidly growing research
labs in our School/University - Founded in 2002
- Houses 20 researchers consisting of
- Research Fellows/PostDocs
- Software Engineers
- PhD candidates
- Honours/Masters students
- Funding
- National and International organizations
- Australian Research Council DEST
- Many industries (Sun, StorageTek, Microsoft, IBM,
Microsoft) - University-wide collaboration
- Faculties of Science, Engineering, and Medicine
- Many national and international collaborations.
- Academics
- Industries
- Software
- Widely in academic and industrial users.
- Publication
Community Services e.g., IEEE TC for Scalable
Computing
3Books at Glance Co-authored/edited
4Outline
- 21st Century Vision of Computing
- Promising Computing Paradigms
- Cloud Computing and Related Paradigms
- Trends, Definition, Characteristics, Architecture
- Market-Oriented Cloud Architecture
- SLA-oriented Resource Allocation
- Global Cloud Exchange
- Emerging Cloud Platforms
- Megha Melbourne Cloud Computing Initiative
- Summary and Thoughts for Future
5Vision Implications of the Internet
- 1969 Leonard Kleinrock, ARPANET project
- As of now, computer networks are still in their
infancy, but as they grow up and become
sophisticated, we will probably see the spread of
computer utilities, which, like present
electric and telephone utilities, will service
individual homes and offices across the country - Computers Redefined
- 1984 John Gage, Sun Microsystems
- The network is the computer
- 2008 David Patterson, U. C. Berkeley
- The data center is the computer. There are
dramatic differences between of developing
software for millions to use as a service versus
distributing software for millions to run their
PCs - 2008 Cloud is the computer Buyya!
6Computing Paradigms and Attributes Realizing the
Computer Utilities Vision
?
- Web
- Data Centres
- Utility Computing
- Service Computing
- Grid Computing
- P2P Computing
- Market-Oriented Computing
- Cloud Computing
- -Ubiquitous access
- -Reliability
- Scalability
- Autonomic
- Dynamic discovery
- Composability
- -QoS
- -SLA
- -
- Trillion business
- Who will own it?
Paradigms
Attributes/Capabilities
7Outline
- 21st Century Vision of Computing
- Promising Computing Paradigms
- Cloud Computing and Related Paradigms
- Trends, Definition, Characteristics, Architecture
- Market-Oriented Cloud Architecture
- SLA-oriented Resource Allocation
- Global Cloud Exchange
- Emerging Cloud Platforms
- Megha Melbourne Cloud Computing Initiative
- Summary and Thoughts for Future
8Web Search Trends Hot News Items (ref Google)
Legend Cluster computing, Grid computing, Cloud
computing
- Nov 15 2007 IBM Introduces 'Blue Cloud'
Computing, CIO Today - Apr 14 2008 Google and Salesforce.com in cloud
computing deal, Siliconrepublic.com - Jun 27 2008 Yahoo realigns to support cloud
computing, 'core strategies', San Antonio
Business Journal - Jul 8 2008 Merrill Lynch Estimates "Cloud
Computing" To Be 100 Billion Market, SYS-CON
Media - Jul 23 2008 Cloud Computing Firm Closes 1.5m
Series A, SYS-CON Media
9What is Cloud ?
- Over 20 definitions
- http//cloudcomputing.sys-con.com/read/612375_p.ht
m - Buyyas definition?
- "A Cloud is a type of parallel and distributed
system consisting of a collection of
inter-connected and virtualised computers that
are dynamically provisioned and presented as one
or more unified computing resources based on
service-level agreements established through
negotiation between the service provider and
consumers. - Keywords Virtualisation (VMs), Dynamic
Provisioning (negotiation and SLAs), and Web 2.0
access interface
10Cloud Services
- Infrastructure as a Services (IaaS)
- CPU, Storage Amazon.com et. al
- Platform as a Services (PaaS)
- Google App Engine, Microsoft Azure, Aneka
- Software as a Service (SaaS)
- SalesForce.Com
11How is it different from other paradigms?
- Cluster Computing
- Tightly coupled
- Homogeneous
- Single System Image
- Distributed Computing
- Loosely coupled
- Heterogeneous
- Single administration
- Grid Computing
- Large scale
- Cross-organizational
- Geographical distribution
- Distributed management
- Cloud Computing
- Provisioned on demand
- Service guarantee
- VMs and Web 2.0-based
Derived from Hiro Kishimoto extended to Cloud
12Characteristics of Clusters, Grids, Clouds (1/2)
13Characteristics of Clusters, Grids, Clouds (2/2)
14Benefits of Clouds to Users
- No upfront infrastructure investment
- No procuring hardware, setup, hosting, power,
etc.. - On demand access
- Lease what you need and when you need..
- Efficient Resource Allocation
- Globally shared infrastructure, can always be
kept busy by serving users from different time
zones... - Nice Pricing
- Based on Usage, QoS, Supply and Demand, Loyalty,
- Application Acceleration
- Parallelism for large-scale data analysis,
what-if scenarios studies - High Availability
- Supports Creation of 3rd Party Services
Seamless offering - Builds on infrastructure and follows similar
Business model as Cloud
15(Layered) Cloud Architecture
Cloud applications Social computing, Enterprise,
ISV, Scientific, CDNs, ...
User level
Cloud programming environments and tools Web 2.0
Interfaces, Mashups, Concurrent and Distributed
Programming, Workflows, Libraries, Scripting
User-LevelMiddleware
Apps Hosting Platforms
QoS Negotiation, Admission Control, Pricing, SLA
Management, Monitoring, Execution Management,
Metering, Accounting, Billing
Autonomic / Cloud Economy
Adaptive Management
CoreMiddleware
Virtual Machine (VM), VM Management and
Deployment
Cloud resources
System level
16Outline
- 21st Century Vision of Computing
- Promising Computing Paradigms
- Cloud Computing and Related Paradigms
- Trends, Definition, Characteristics, Architecture
- Market-Oriented Cloud Architecture
- SLA-oriented Resource Allocation
- Global Cloud Exchange
- Emerging Cloud Platforms
- Megha Melbourne Cloud Computing Initiative
- Summary and Thoughts for Future
17Realizing the Computer Utilities Vision What
Consumers and Providers Want?
- Consumers minimize expenses, meet QoS
- How do I express QoS requirements to meet my
goals? - How do I assign valuation to my applications?
- How do I discover services and map applications
to meet QoS needs? - How do I manage multiple providers and get my
work done? - How do I outperform other competing consumers?
-
- Providers maximise Return On Investment (ROI)
- How do I decide service pricing models?
- How do I specify prices?
- How do I translate prices into resource
allocations? - How do I assign and enforce resource allocations?
- How do I advertise and attract consumers?
- How do I perform accounting and handle payments?
-
- Mechanisms, tools, and technologies
- value expression, translation, and enforcement
18Market-based Systems Self-managed and
self-regulated systems.
- Manage
- Complexity
- Supply and Demand
- Enhance Utility
1
2
3
penalty
19Market-oriented Cloud Architecture QoS
negotiation and SLA-based Resource Allocation
20InterCloud Global Cloud Exchange and Market Maker
21Outline
- 21st Century Vision of Computing
- Promising Computing Paradigms
- Cloud Computing and Related Paradigms
- Trends, Definition, Characteristics, Architecture
- Market-Oriented Cloud Architecture
- SLA-oriented Resource Allocation
- Global Cloud Exchange
- Emerging Cloud Platforms
- Megha Melbourne Cloud Computing Initiative
- Summary and Thoughts for Future
22Some Commercial-Oriented Cloud platforms/technolog
ies
23Outline
- 21st Century Vision of Computing
- Promising Computing Paradigms
- Cloud Computing and Related Paradigms
- Trends, Definition, Characteristics
- Market-Oriented Cloud Architecture
- SLA-oriented Resource Allocation
- Global Cloud Exchange
- Emerging Cloud Platforms
- Megha Melbourne Cloud Computing Initiative
- Summary and Thoughts for Future
24(No Transcript)
25GRIDS Labs Cloud Computing Initiatives
- Aneka A Software System Building enterprise
Cloud Computing - Market-Oriented Clouds
- SLA-based Resource Management
- Global Cloud Exchange Elements Brokers
- Building 3rd Party Cloud Services (MetaClouds)
- Building Content Delivery Networks using
different vendors Storage Clouds - A Grid of Clouds?
- Extending existing Market Oriented Grids ideas
26Aneka A 3rd Gen enterprise Grid Technology ?
Cloud model
27ANEKA Product Overview (Alpha)
- .NET based service-oriented platform for grid /
cloud computing - Development and Run Time Environment
- Includes Development and Management Tools
- Suitable for
- Development of Enterprise Grid / Cloud
Applications - Grid / Cloud enabling legacy applications
- Ideal for Corporate Developers, Software, SaaS,
Hosting Vendors and Application / System
Integrators
ANEKA Product Architecture
28Aneka Deployment Models
- Enterprise/Private
- LAN connected resources
- Application Development, Testing
- Teaching and Learning
- Sensitive applications
- Public
- Amazon.com, Microsoft Azure
- SDK Framework
Enterprise/Private Clouds
Aneka
Public Clouds
29Building Aneka Enterprise Grids/Clouds
Aneka enterprise Cloud
work units
internet
work units
internet
30Aneka A dynamic platform driving cloud
applications
Executor
Grid Application
Manager
Manager / Executor
Grid Threads
31Anekaa SLA-View for Resource Allocation
Negotiation Protocol Engine
32QoS Negotiation in Aneka
Meta Negotiation Registry
3. Matching
DB
DB
DB
Registries
1. Publishing
2. Publishing, Querying
MN Middelware
MN Middelware
Gridbus Broker
Aneka
4. Session Establishment
Meta-Negotiation
Meta-Negotiation
Amadeus Workflow
Handshaking
Handshaking
Alternate Offers Negotiation Strategy
Alternate Offers Negotiation Strategy
Local SLA Template
WSDL
Local SLA Template
5. Negotiation
Party 1
Party 2
API
Service Consumer
Service Provider
6. Service Invocation
33Performance Evaluation (with UCSD SP2 workload
traces) Aneka Cloud node
34LibraAuto (Self-adjust Pricing Factor)
Algorithm Experimental Scenario
- Feitelsons Parallel Workload Archive
- Last 7 days in SDSC SP2 trace
35Dynamic Pricing Vs Revenue
36Case Study Protein Structure Prediction Portal
37Putting Aneka on Public Clouds A possible
exploration
- We like to building an Image of Aneka for
Deployment under (1) Amazon Windows Cloud
Infrastructure and (2) Microsoft Azure? - Position it as a Framework for Accelerating
Application Execution written for Windows/.NET
environments. - Two Business Model options
- 1. Amazon/Microsoft Cloud cost small fee for
using Aneka (need to make a deal with the to have
them manage it) - 2. We need to create a 3rd part service on
Amazon/Microsoft Cloud infrastructure and make
that as our own offering.
38Building 3rd Party Cloud Services
- A Case Study in Harnessing Storage Clouds for
Building Next-Gen Content Delivery Networks
39Internet/Web Flash Crowd Problem
- With the rapid growth of Internet and Web
- Services are competing each other for finite
network and computing resources - High availability and responsiveness are keys to
business Web sites - Large number of users are trying to
simultaneously access the same Web site, causing
Flash Crowd
40Centralized Web Flash Crowd Problem
- Slow
- content must traverse multiple backbones and long
distances - Unreliable
- delivery may be prevented by congestion or
backbone peering problems - Not scalable
- usage limited by bandwidth available at master
site - Inferior streaming quality
- packet loss, congestion, and narrow pipes degrade
stream quality
Source Bruce Maggs, CCGrid 2001 Keynote
41Content Delivery Network (CDN)
- Content Delivery Networks (CDNs) emerged as a
solution to Internet service degradation - Moving content to the edge of the Internet,
close to end-users - Alternatives
- Increased bandwidth, Web caching, Web
pre-fetching - CDN advantages
- Reduced server loads
- Distributed network traffic
- Reduced latency
42Motivations
- Content Delivery Networks (CDNs) such as Akamai
place web server clusters in numerous
geographical locations huge upfront
investment - to improve the responsiveness and locality of the
content it hosts for end-users. - However, their services are priced out of reach
for all but the largest enterprise customers. - Hence, we have developed an alternative approach
to content delivery by leveraging infrastructure
Storage Cloud providers at a fraction of the
cost of traditional CDN providers pay as you
go
43Commercial Storage Clouds Pricing
44MetaCDN.org Harnessing Storage Clouds for
Content Delivery Network
45Market-Oriented Grids
- Thoughts for Building a Grid of Clouds
46Building a Grid of Clouds ? Global Utility
Computing
Grid Information Service
Grid Resource Broker
Application
R2
R3
R4
R5
RN
Grid Resource Broker
R6
R1
Resource Broker
Grid Information Service
47Grid Service Broker (GSB)
- A resource broker for scheduling task farming
data Grid applications with static or dynamic
parameter sweeps on global Grids. - It uses computational economy paradigm for
optimal selection of computational and data
services depending on their quality, cost, and
availability, and users QoS requirements
(deadline, budget, T/C optimisation) - Key Features
- A single window to manage control experiment
- Programmable Task Farming Engine
- Resource Discovery and Resource Trading
- Optimal Data Source Discovery
- Scheduling Predications
- Generic Dispatcher Grid Agents
- Transportation of data sharing of results
- Accounting
48workload
Gridbus User Console/Portal/Application Interface
App, T, , Optimization Preference
Gridbus Broker
Gridbus Farming Engine
Schedule Advisor
Trading Manager
RecordKeeper
Grid Dispatcher
Grid Explorer
TM TS
GE GIS, NWS
Core Middleware
Grid Info Server
RM TS
G
Data Catalog
Data Node
C
U
G
Unicore enabled node.
Globus enabled node.
L
A
49Gridbus Broker Separating applications from
different remote service access enablers
Application Development Interface
Single-sign on security
Alogorithm1
SchedulingInterfaces
AlogorithmN
Plugin Actuators
Data Store
Access Technology
SRB
Grid FTP
50Gridbus Services for eScience applications
- Application Development Environment
- XML-based language for composition of task
farming (legacy) applications as parameter sweep
applications. - Task Farming APIs for new applications.
- Web APIs (e.g., Portlets) for Grid portal
development. - Threads-based Programming Interface
- Workflow interface and Gridbus-enabled workflow
engine. - Grid Superscalar in cooperation with BSC/UPC
- Resource Allocation and Scheduling
- Dynamic discovery of optional computational and
data nodes that meet user QoS requirements. - Hide Low-Level Grid Middleware interfaces
- Globus (v2, v4), SRB, Aneka, Unicore, and
ssh-based access to local/remote resources
managed by XGrid, PBS, Condor, SGE.
51Click Here for Demo
Drug Design Made Easy!
52Adaptive Scheduling Steps
Discover More Resources
Discover Resources
Establish Rates
Evaluate Reschedule
Compose Schedule
Meet requirements ? Remaining Jobs, Deadline,
Budget ?
Distribute Jobs
53Deadline (D) and Budget (B) Constrained
Scheduling Algorithms
54s
55Case Study High Energy Physics and Data Grid
- The Belle Experiment
- KEK B-Factory, Japan
- Investigating fundamental violation of symmetry
in nature (Charge Parity) which may help explain
why do we have more antimatter in the universe
OR imbalance of matter and antimatter in the
universe?. - Collaboration 1000 people, 50 institutes
- 100s TB data currently
56Case Study Event Simulation and Analysis
B0-gtDD-Ks
- Simulation and Analysis Package - Belle Analysis
Software Framework (BASF) - Experiment in 2 parts Generation of Simulated
Data and Analysis of the distributed data -
?Analyzed 100 data files (30MB each) that were
distributed among the five nodes within
Australian Belle DataGrid platform.
57Australian Belle Data Grid Testbed
VPACMelbourne
58Belle Data Grid (GSP CPU Service Price G/sec)
G4
NA
G4
G6
VPACMelbourne
G2
Datanode
59Belle Data Grid (Bandwidth Price G/MB)
32
33
36
G4
31
30
34
NA
38
31
G4
G6
VPACMelbourne
G2
Datanode
60Deploying Application Scenario
- A data grid scenario with 100 jobs and each
accessing remote data of 30MB - Deadline 3hrs.
- Budget G 60K
- Scheduling Optimisation Scenario
- Minimise Time
- Minimise Cost
- Results
61Time Minimization in Data Grids
62Results Cost Minimization in Data Grids
63Observation
64Outline
- 21st Century Vision of Computing
- Promising Computing Paradigms
- Cloud Computing and Related Paradigms
- Trends, Definition, Characteristics, Architecture
- Market-Oriented Cloud Architecture
- SLA-oriented Resource Allocation
- Global Cloud Exchange
- Emerging Cloud Platforms
- Megha Melbourne Cloud Computing Initiative
- Summary and Thoughts for Future
65Summary
- Several Computing Platforms/Paradigms are
promising to deliver Computing Utilities vision - Cloud Computing is the most recent kid promising
to turn vision into reality - Clouds built on SOA, VMs, Web 2.0 technologies
- Market Oriented Clouds are getting real
- Need to move from static pricing to dynamic
pricing - Need strong support for SLA-based resource
management - 3rd party Composed Cloud services starting to
emerge - Building Grids using Clouds is much more
realistic. - Extension of idea can lead to ? Global Cloud
Exchange
66Many Challenges Remains to be Solved
ManagementPerspective
EngineeringPerspective
67Convergence of Competing Paradigms/Communities
Needed
?
- Web
- Data Centres
- Utility Computing
- Service Computing
- Grid Computing
- P2P Computing
- Cloud Computing
- Market-Oriented Computing
- Ubiquitous access
- Reliability
- Scalability
- Autonomic
- Dynamic discovery
- Composability
- QoS
- SLA
- Trillion business
- Who will own it?
Paradigms
Attributes/Capabilities
68References
- Keynote Paper
- R. Buyya, C. S. Yeo, and S. Venugopal,
Market-oriented Cloud Computing Vision, Hype,
and Reality for Delivering IT Services as
Computing Utilities, Proceedings of the 10th
IEEE International Conference on High Performance
Computing and Communications (HPCC 2008), Dalian,
China, Sept. 2008. - Extended version of Keynote Paper
- R. Buyya, C. S. Yeo, S. Venugopal, J. Broberg, I.
Brandic, Cloud Computing and Emerging IT
Platforms Vision, Hype, and Reality for
Delivering Computing as the 5th Utility, Future
Generation Computer Systems (FGCS) Journal, 2009.
(in press) - Aneka Paper
- X. Chu, K. Nadiminti, C. Jin, S. Venugopal, and
R. Buyya, Aneka Next-Generation Enterprise Grid
Platform for e-Science and e-Business
Applications, Proceedings of the 3rd IEEE
International Conference on e-Science and Grid
Computing (e-Science 2007), Dec. 10-13, 2007,
Bangalore, India. - The Grid Economy Paper
- R. Buyya, D. Abramson, S. Venugopal, The Grid
Economy, Proceedings of the IEEE, No. 3, Volume
93, IEEE Press, 2005.
69Thanks for your attention!
- Are there any
- Questions?
- Comments/ Suggestions
We Welcome Cooperation in RD and Business!
http/www.gridbus.org www.Manjrasoft.com rbuyya_at_
unimelb.edu.au raj_at_manjrasoft.com
70Backup Slides
71Aneka a multi-model platform for parallel and
distributed applications
Applications
Container
Thread Model
Task Model
Dataflow Model
MPI Model
Map Reduce
Other Models
SLANegotiation
Allocation Manager
Persistence
Message Handler / Dispatcher
Security
Communication Layer
72Research Goal SLA-oriented Resource Allocation
for Clouds
- Services Science
- Focusing on computational approaches and computer
systems to improve service efficiency and
catalyse service innovation - Towards utility computing vision of computer
utilities - Software as a Service (SaaS)
- Enterprise, scientific personal applications
- Infrastructure from IT industry
- Amazon EC2/S3, Sun Network.com
- Service Level Agreement (SLA)-oriented allocation
- SLA contract between consumer provider
- Specific service needs of consumer
- Risk management of provider
73Cloud Computing Architecture
74Cloud Architecture Explained
75What are Grid benefits?
- Resource sharing across multiple administrative
boundaries - Effective utilisation of the (existing) resources
- Dynamic provisioning
- Application Acceleration
- Scalability
- Reliability
- Virtualisation
- applications, services, resources,
76Aneka System Architecture
Container
Security
Services
Information Indexing
Scheduling
SLANegotiation
Thread Scheduler
Membership Catalog
Dataflow Scheduler
Application Catalog
Remote interactions
Authorization
MPI Scheduler
Data Catalog
Mapping Scheduler
Authentication
Message Handler / Dispatcher
Communication Layer
Auditing
Execution
Storage
Others
Dataflow Executor
Banking Service
MPI Executor
File Server
Thread Executor
Persistence
Optional
Compulsory
77LibraAutoSelf-adjust Pricing Factor
78Interactions