Post on 17-Jun-2020
University of Ljubljana Faculty of civil and geodetic engineering
mdolenc@itc.fgg.uni-lj.siECT 2012 Matevž Dolenc & Robert Klinc
Platform as a Service computing environment for earthquake engineering
Dubrovnik, Croatia, 4-7 September 2012
mdolenc@itc.fgg.uni-lj.siPlatform as a Service computing environment for earthquake engineering | ECT 2012, Dubrovnik, Croatia, 4-7 September 2012
Outline
‣ Introduction- Cloud computing overview
- PaaS: definition, characteristics, providers
‣ ICE4Risk project- Overview
- Cloud platform
‣ Summary
mdolenc@itc.fgg.uni-lj.siPlatform as a Service computing environment for earthquake engineering | ECT 2012, Dubrovnik, Croatia, 4-7 September 2012
Cloud computing
‣ Definition- “a paradigm in which information is permanently stored in servers
on the Internet and cached temporarily on clients that include desktops, entertainment centres, tablet computers, notebooks, wall computers, handhelds, sensors, monitors, etc.”
‣ Characteristics- Non-functional: elasticity, reliability, QoS, agility and
adaptability, availability, ...
- Economic: cost reduction, pay per use, improved time to market, ROI, ...
- Technological: virtualisation, security, privacy and compliance, data management, APIs, metering, ...
C. Hewitt, “ORGs for Scalable, Robust, Privacy-Friendly Client Cloud Computing”, IEEE Internet Computing, vol. 12, no. 5, pp. 96-99, Sep/Oct, 2008.
mdolenc@itc.fgg.uni-lj.siPlatform as a Service computing environment for earthquake engineering | ECT 2012, Dubrovnik, Croatia, 4-7 September 2012
Cloud computing: benefits & barriers to adoption
1. Access data/services at any place, from any device and at any time
2. Lower cost of entry
3. Reliability, scalability, security and sustainability
4. Minimize infrastructure risk
5. Reduce run time and response time
6. Increased pace of innovation
1. Availability of a service
2. Data lock-in
3. Data confidentiality and audibility
4. Data transfer bottlenecks
5. Performance unpredictability
6. Scalable storage
7. Bugs in large-scale distributed systems
8. Scaling quickly
9. Reputation, fate sharing
10. Software licensing
Benefits Barriers to adoption
mdolenc@itc.fgg.uni-lj.siPlatform as a Service computing environment for earthquake engineering | ECT 2012, Dubrovnik, Croatia, 4-7 September 2012
Cloud computing barriers to adoption
HUMAN BARRIERS
LEGAL BARRIERS
TECHNOLOGICAL BARRIERS
ORGANIZATIONAL BARRIERS
commercial constraints
lack of strategic planning
lack of commitment
lack of time
lack of energy
lack of resources
late implementation
traditional contacts
fragmented process
data protection
IPR
legal admissibility
fear of change
resistance to change
fear of failuredifferent cultures
different languages
different time zones
diverse company values
diverse company procedures
lack of understanding
lack of experience
lack of education
lack of training
lack of ability
human behaviour
lack of trust
multiple standards
poor standards
poor interoperability
information overload
vendor commercial interests
lack of investments
poor adoption rates
inaccessible information
skill shortage rapid ICT product change
mdolenc@itc.fgg.uni-lj.siPlatform as a Service computing environment for earthquake engineering | ECT 2012, Dubrovnik, Croatia, 4-7 September 2012
Cloud computing landscape
K. Jeffery and B. Neidecker-Lutz, The Future of Cloud Computing: Opportunities for European Cloud Computing Beyond 2010, http:// cordis.europa.eu/fp7/ict/ssai/docs/cloud-report-final.pdf
8 | P a g e
II. WHAT IS A “CLOUD” Various definitions and interpretations of “clouds” and / or “cloud computing” exist. With particular respect to the various usage scopes the term is employed to, we will try to give a representative (as opposed to complete) set of definitions as recommendation towards future usage in the cloud computing related research space. This report does not claim completeness with this respect, as it does not introduce a new terminology, but tries to capture an abstract term in a way that best represents the technological aspects and issues related to it.
FIGURE 1: NON-EXHAUSTIVE VIEW ON THE MAIN ASPECTS FORMING A CLOUD SYSTEM
A. TERMINOLOGY In its broadest form, we can define
a 'cloud' is an elastic execution environment of resources involving multiple stakeholders and providing a metered service at multiple granularities for a specified level of quality (of service).
In other words, clouds as we understand them in the context of this document are primarily platforms that allow execution in various forms (see below) across multiple resources (and potentially across enterprise boundaries, see below) – the main characteristics will be detailed in section II.B. We can distinguish different types of clouds (cf. section II.A.1), all of which have in common that they (directly or indirectly) enhance resources and services with additional capabilities related to manageability, elasticity and system platform independency.
To be more specific, a cloud is a platform or infrastructure that enables execution of code (services, applications etc.), in a managed and elastic fashion, whereas “managed” means that reliability according to pre-defined quality parameters is automatically ensured and “elastic” implies that the
FEATURES MODES
LOCALITYBENEFITS
COMPARES TO STAKEHOLDERS
ReliabilityTYPESElasticity
Virtualisation
…
IaaS
PaaS
SaaS
Public
Private
Hybrid
…
Local
Remote
Distributed
Cost Reduction
Ease of use
…
Internet ofServices
Grid
Service-orientedArchitecture
ResellersProviders
Adopters
Users
…
Cloud Systems
mdolenc@itc.fgg.uni-lj.siPlatform as a Service computing environment for earthquake engineering | ECT 2012, Dubrovnik, Croatia, 4-7 September 2012
Cloud computing layers
Platform as a Service(PaaS)
Software as a Service(SaaS)
Infrastructure as a Service (IaaS)
Cloud computing
Utility computing
Grid computing
Cluster computing Super computing
Google App EngineHeroku
Microsoft Azure...
SalesforceGMail...
Amazon WSAkamaiJoyent...
mdolenc@itc.fgg.uni-lj.siPlatform as a Service computing environment for earthquake engineering | ECT 2012, Dubrovnik, Croatia, 4-7 September 2012
Cloud services in context
Users
ApplicationExtra
FunctionsApplication
ApplicationBrowser /Client
Application
Platform
Cloud
Local
Users Developers
Software as a service (SaaS) Attached services Platform as a Service (PaaS)
mdolenc@itc.fgg.uni-lj.siPlatform as a Service computing environment for earthquake engineering | ECT 2012, Dubrovnik, Croatia, 4-7 September 2012
Platform as a Service (PaaS)
‣ Definition- “provide computational resources via a platform upon which
applications and services can be developed and hosted”
‣ Characteristics- Multi-tenant architecture
- Customizable /Programmable User Interface
- Storage customization
- Workflow engine/capabilities
- Flexible “services-enabled” integration model
- Permission controlK. Jeffery and B. Neidecker-Lutz, The Future of Cloud Computing: Opportunities for European Cloud Computing Beyond 2010,
http:// cordis.europa.eu/fp7/ict/ssai/docs/cloud-report-final.pdf
‣ Google App Engine- developers.google.com/appengine/
- Java, Pyton, Go
‣ Heroku- heroku.com
- Ruby, Node.js, Clojure, Java, Python, Scala
‣ Microsoft Azure- windowsazure.com
- .NET, Node.js, Java, PHP, Python, ...
mdolenc@itc.fgg.uni-lj.siPlatform as a Service computing environment for earthquake engineering | ECT 2012, Dubrovnik, Croatia, 4-7 September 2012
Platform as a Service (PaaS) general offerings
mdolenc@itc.fgg.uni-lj.siPlatform as a Service computing environment for earthquake engineering | ECT 2012, Dubrovnik, Croatia, 4-7 September 2012
ICE4Risk project overview
‣ High-throughput computing environment for seismic risk assessment (ICE4Risk)
- ice4risk.slo-projekt.info
- supported by Slovenian Research Agency
‣ Goals- PBEE toolbox: a user-friendly and adaptable tool for
seismic performance assessment of structures based on Matlab and OpenSees (PEER Center)
- Web-based application for the prediction of approximate IDA curves
mdolenc@itc.fgg.uni-lj.siPlatform as a Service computing environment for earthquake engineering | ECT 2012, Dubrovnik, Croatia, 4-7 September 2012
Prediction of approximate IDA curves
‣ Motivation- Fast and accurate prediction of approximate Incremental
Dynamic Analysis (IDA) curves
‣ Possible applications- Determination of target displacement
- Collapse capacity
- Dispersions in demand and capacity
- Precedence list of ground motion records
- Accurate estimation of demand for seismic performance assessment of urban areas
mdolenc@itc.fgg.uni-lj.siPlatform as a Service computing environment for earthquake engineering | ECT 2012, Dubrovnik, Croatia, 4-7 September 2012
Prediction of approximate IDA curves
‣ Methodology- First process: Response Database
- Second process: n-dimensional linear interpolation
!
mdolenc@itc.fgg.uni-lj.siPlatform as a Service computing environment for earthquake engineering | ECT 2012, Dubrovnik, Croatia, 4-7 September 2012
First process: Response database
‣ Parametric study- One time analysis, very time consuming
- Condor + OpenSEES, 144 CPU cores, approx. 10 days
‣ 200 ground motion records
- 1.6 million IDA curves, 1 GB database (2.8 GB original output)
User
Amazon Elastic Compute Cloud (Amazon EC2)
Amazon Simple Storage Service (Amazon S3)
Storage
Retrieve /Store Data
Submit Jobs
StorageCondor worker
Condor worker
Condor worker
Condor central
manager
Condor sumbit host
Start EC2 virtual computers
Store / Retrieve Data Report Status / Run User Jobs
mdolenc@itc.fgg.uni-lj.siPlatform as a Service computing environment for earthquake engineering | ECT 2012, Dubrovnik, Croatia, 4-7 September 2012
Second Process: n-dimensional linear interpolation
‣ Web application- ice4risk.slo-projekt.info/analysis
Translates tasks and results to something which the user can understand by using an internet browser
Scripts (PHP) process the input parameters, interacting with the relational database, and preparing the output (X)HTML files. Fortran is used for computations.
IDA curves are stored in the MySQL relational database, which enable connectivity with the PHP scripts and the Apache web server.
mdolenc@itc.fgg.uni-lj.siPlatform as a Service computing environment for earthquake engineering | ECT 2012, Dubrovnik, Croatia, 4-7 September 2012
Second Process: n-dimensional linear interpolation
mdolenc@itc.fgg.uni-lj.siPlatform as a Service computing environment for earthquake engineering | ECT 2012, Dubrovnik, Croatia, 4-7 September 2012
Second Process: n-dimensional linear interpolation
mdolenc@itc.fgg.uni-lj.siPlatform as a Service computing environment for earthquake engineering | ECT 2012, Dubrovnik, Croatia, 4-7 September 2012
ICE4Risk platform
‣ Requirements- Enable parametric studies
- Easy integration of existing software
- Scalability
‣ Design principle- Ease of use
- 80/20 rule
mdolenc@itc.fgg.uni-lj.siPlatform as a Service computing environment for earthquake engineering | ECT 2012, Dubrovnik, Croatia, 4-7 September 2012
ICE4Risk platform architecture (old)
User
Amazon Elastic Compute Cloud (Amazon EC2)
Amazon Simple Storage Service (Amazon S3)
Storage
Retrieve /Store Data
Submit Jobs
StorageCondor worker
Condor worker
Condor worker
Condor central manager
Condor sumbit host
Start EC2 virtual computers
Store / Retrieve Data Report Status / Run User Jobs
mdolenc@itc.fgg.uni-lj.siPlatform as a Service computing environment for earthquake engineering | ECT 2012, Dubrovnik, Croatia, 4-7 September 2012
ICE4Risk platform architecture (new) - remote
User
Amazon Elastic Compute Cloud (Amazon EC2)
Amazon Simple Storage Service (Amazon S3)
Storage
Retrieve /Store Data
StorageCondor worker
Condor worker
Condor worker
Condor central manager
Start EC2 virtual computers
Store / Retrieve Data
Report Status /Run User Jobs
ICE4RISK
APIs
Web b
rowser
Condor Service
DM Service
VMs Service
...
In-house Apache server
mdolenc@itc.fgg.uni-lj.siPlatform as a Service computing environment for earthquake engineering | ECT 2012, Dubrovnik, Croatia, 4-7 September 2012
ICE4Risk platform architecture (new) - local
User
OpenStack Compute VMs / Local computer nodes
OpenStack Storage /Shared local file system
Storage
Retrieve /Store Data
StorageCondor worker
Condor worker
Condor worker
Condor central manager
Start/use virtual computers
Store / Retrieve Data
Report Status /Run User Jobs
ICE4RISK
APIs
Web b
rowser
Condor Service
DM Service
VMs Service
...
In-house Apache server
OpenStack - Open source software for building private and public clouds., openstack.org
mdolenc@itc.fgg.uni-lj.siPlatform as a Service computing environment for earthquake engineering | ECT 2012, Dubrovnik, Croatia, 4-7 September 2012
ICE4Risk platform architecture
‣ Virtualized computer infrastructure- Scalable local or remote compute nodes
- Transparent for the end-user
‣ ICE4Risk APIs- Supports base operations (VM, DM, Condor, ...)
- APIs use REST interface
- Enable vertical applications
mdolenc@itc.fgg.uni-lj.siPlatform as a Service computing environment for earthquake engineering | ECT 2012, Dubrovnik, Croatia, 4-7 September 2012
Summary
‣ Cloud computing- Clear benefits for future developments
‣ ICE4Risk platform- Better end-user/admin web interface
- Development of vertical applications
- New APIs
‣ ISES project (Intelligent Services for Energy-Efficient Design and Life Cycle Simulation)
- ises.eu-project.info
- BIM, vertical applications, ...