41
eScience challenges in levees monitoring - lessons from "flood" projects Marian Bubak Department of Computer Science AGH University of Science and Technology Kraków, Poland http://dice.cyfronet.pl/ eScience 2015, Munich, August 31 – September 4, 2015

EScience challenges in levees monitoring - lessons from "flood" projects Marian Bubak Department of Computer Science AGH University of Science and Technology

Embed Size (px)

Citation preview

Page 1: EScience challenges in levees monitoring - lessons from "flood" projects Marian Bubak Department of Computer Science AGH University of Science and Technology

eScience challenges in levees monitoring - lessons from "flood" projects

Marian BubakDepartment of Computer Science

AGH University of Science and TechnologyKraków, Poland

http://dice.cyfronet.pl/

eScience 2015, Munich, August 31 – September 4, 2015

Page 2: EScience challenges in levees monitoring - lessons from "flood" projects Marian Bubak Department of Computer Science AGH University of Science and Technology

• Bartosz Balis• Daniel Harezlak• Maciej Malawski• Piotr Nowakowski• Bartosz Wilk

• Tomasz Gubala• Marek Kasztelnik• Jan Meizner• Maciej Pawlik• ...

And colleagues from CrossGrid, K-WfGrid, UrbanFlood, ISMOP

Thanks to

Page 3: EScience challenges in levees monitoring - lessons from "flood" projects Marian Bubak Department of Computer Science AGH University of Science and Technology

Outline

• Motivation• Interactive system (person in a loop)• Exploitation of knowledge• Building early warming systems• IT support for levees monitoring• Summary

Page 4: EScience challenges in levees monitoring - lessons from "flood" projects Marian Bubak Department of Computer Science AGH University of Science and Technology

Motivation: our area of research

• Investigation of methods for complex scientific collaborative applications• Elaboration of environments and tools for eScience• Integration of large-scale distributed computing infrastructures• Knowledge-based approach to services, components, and their composition

Page 5: EScience challenges in levees monitoring - lessons from "flood" projects Marian Bubak Department of Computer Science AGH University of Science and Technology

Motivation: Krakow, May 2010

Page 6: EScience challenges in levees monitoring - lessons from "flood" projects Marian Bubak Department of Computer Science AGH University of Science and Technology

Flood - CrossGrid (2002- 5)

CrossGrid: Development of Grid Environment for Interactive Applicationshttp://cordis.europa.eu/project/rcn/63588_en.html ftp://ftp.cordis.europa.eu/pub/ist/docs/grids/crossgrid_achievement.pdf

L. Hluchy, V. D. Tran, O. Habala, B. Simo, E. Gatial, J. Astalos, M. Dobrucky: Flood Forecasting in CrossGrid Project, in Marios D. Dikaiakos (Eds): Grid Computing Second European AcrossGrids Conference, AxGrids 2004, Nicosia, Cyprus, January 28-30, 2004. Revised Papers, LNCS 3165, 51-60, 2004

This paper presents a prototype of flood forecasting system based on Grid technologies. The system consists of workflow system for executing simulation cascade of meteorological, hydrological and hydraulic models, data management system for storing and accessing different computed and measured data, and web portals as user interfaces. The whole system is tied together by Grid technology and is used to support a virtual organization of experts, developers and users.

Page 7: EScience challenges in levees monitoring - lessons from "flood" projects Marian Bubak Department of Computer Science AGH University of Science and Technology

Flood - K-WfGrid (2004-7)

K-WfGrid: Knowledge-based workflow system for Grid applications

http://cordis.europa.eu/publication/rcn/9410_en.htmlftp://ftp.cordis.europa.eu/pub/ist/docs/grids/k-wf-grid-interim-sheet_en.pdf

Ladislav Hluchý, Ondrej Habala, Martin Maliska, Branislav Simo, Viet D. Tran, Ján Astalos, Marian Babik: Grid Based Flood Prediction Virtual Organization. e-Science 2006, 4-6 December 2006, Amsterdam

This paper describes evolution of one such system -- a flood prediction application. The application consists of a set of simulation models, visualization tools, and various support components. During past six years it has evolved from a simple hydraulic modeling scenario into a sophisticated cascade of simulations, using state-of-the art grid, workflow and knowledge management technologies, and is one of the first applications of the SOKU [1]concept in the field of computer simulations.

Page 8: EScience challenges in levees monitoring - lessons from "flood" projects Marian Bubak Department of Computer Science AGH University of Science and Technology

From IJdijk to UrbanFlood (2008)

The IJkdijk consortium turns to 7FP to organize research on the development of • GeoSensing technology• Sensor network

telecommunication systems

• Sensor data processing facilities

• Smartness in sensors (sensor plug and play, data acquisition).

Robert Meijer, TNO ICT Groningenand University of Amsterdam

http://www.floodcontrolijkdijk.nl/en/

Page 9: EScience challenges in levees monitoring - lessons from "flood" projects Marian Bubak Department of Computer Science AGH University of Science and Technology

Smart levees

Page 10: EScience challenges in levees monitoring - lessons from "flood" projects Marian Bubak Department of Computer Science AGH University of Science and Technology

Stand-by mode

• Monitoring data collection (low frequency)

• Initial on-line analysis (trends, deviations in sensor readings)

• Presentation of external info: weather prediction, flood wave prediction, etc.

Threat assessment mode

• Increased frequency of sensor data collection

• Resource-intensive threat level evaluation

Monitoring and decisions

Page 11: EScience challenges in levees monitoring - lessons from "flood" projects Marian Bubak Department of Computer Science AGH University of Science and Technology

Monitor (AI)Decision SupportScenario ComputationVisualization

SSS

Control CentreSSS

Control Centre

AmsterdamRhine, De

Boston UK

Internet

AuthoritiesScience

Public

CISCISCIS

EWS1EWS2

EWS3

SoftwareHardware

UrbanFlood -Early Warning System

Page 12: EScience challenges in levees monitoring - lessons from "flood" projects Marian Bubak Department of Computer Science AGH University of Science and Technology

A platform facilitating development, deployment and execution of EWSs• EWS development

– EWS reference model– EWS development framework

• EWS deployment– EWS blueprints– EWS-factory-as-a-service

• EWS execution– CIS runtime services for resource allocation, self-monitoring,

self-healing, mission-critical operation, and urgent computing

Common Information Space

Page 13: EScience challenges in levees monitoring - lessons from "flood" projects Marian Bubak Department of Computer Science AGH University of Science and Technology

Flood simulation with CIS

Page 14: EScience challenges in levees monitoring - lessons from "flood" projects Marian Bubak Department of Computer Science AGH University of Science and Technology

Domain resources exposed as Basic ServicesData, sensors, apps wrapped as appliances and deployed

onto clouds, …

Composite Services (Parts)Building blocks for EWSs

Orchestrate domain resources towards complex application scenarios (e.g.

area flood simulation)

Early Warning SystemA number of Parts deployed,

connected, and configured for a specific setting (e.g. a dike

section)

Common Information Space

Page 15: EScience challenges in levees monitoring - lessons from "flood" projects Marian Bubak Department of Computer Science AGH University of Science and Technology

Flood EWS with CIS

Page 16: EScience challenges in levees monitoring - lessons from "flood" projects Marian Bubak Department of Computer Science AGH University of Science and Technology

CIS as a system factory

• On-demand resource provisioning (local resources, clouds)

• Horizontal scaling of infrastructure (more instances)

• Load balancing with lazy evaluation• On-line availability monitoring• Notifications about problems• Automatic restart of failed components

Page 17: EScience challenges in levees monitoring - lessons from "flood" projects Marian Bubak Department of Computer Science AGH University of Science and Technology

ISMOP: towards a levee monitoring system

Investigations on monitoring and assessment of levees:• Construction of an artificial levee• New sensors for levee instrumentation• Design and development of a sensor

communication infrastructure– Optimal collection and transmission of sensor

data• Levee modeling and simulation

– Comparison of simulated and real levee behavior

• Central System: software platforms for execution management, data management, visualization and decision support

Page 18: EScience challenges in levees monitoring - lessons from "flood" projects Marian Bubak Department of Computer Science AGH University of Science and Technology

ISMOP: Consortium

• Department of Computer Science AGH• Department of Hydrogeology and Engineering

Geology AGH• Department of Geoinformatics and Applied

Computer Science AGH• NeoSentio, Kraków• Sweco Hydroprojekt Krakówin collaboration with the Czernichów Community

http://www.ismop.edu.pl/

Page 19: EScience challenges in levees monitoring - lessons from "flood" projects Marian Bubak Department of Computer Science AGH University of Science and Technology

ISMOP central system use cases

• Support for experiments on the artificial levee– Controlled flooding of the artificial levee and on-

line data collection– Validation of models of levees

• Elaboration of a decision support system– Continuous monitoring of levees – Automation data-driven and model-driven analyses– Prediction of breaches

Page 20: EScience challenges in levees monitoring - lessons from "flood" projects Marian Bubak Department of Computer Science AGH University of Science and Technology

Experimental levee 14.08.2015 (1/4)

Page 21: EScience challenges in levees monitoring - lessons from "flood" projects Marian Bubak Department of Computer Science AGH University of Science and Technology

Experimental levee 14.08.2015 (2/4)

Page 22: EScience challenges in levees monitoring - lessons from "flood" projects Marian Bubak Department of Computer Science AGH University of Science and Technology

Experimental levee 14.08.2015 (3/4)

Page 23: EScience challenges in levees monitoring - lessons from "flood" projects Marian Bubak Department of Computer Science AGH University of Science and Technology

Experimental levee 14.08.2015 (4/4)

Page 24: EScience challenges in levees monitoring - lessons from "flood" projects Marian Bubak Department of Computer Science AGH University of Science and Technology

Assessment of levee breach threat via scenario matching

Page 25: EScience challenges in levees monitoring - lessons from "flood" projects Marian Bubak Department of Computer Science AGH University of Science and Technology

ISMOP Decision Suport System

Page 26: EScience challenges in levees monitoring - lessons from "flood" projects Marian Bubak Department of Computer Science AGH University of Science and Technology

•Solution•Leveraging open standards (OGC, INSPIRE) for data & metadata models

Interoperability with external systems (e.g.

ISOK, regional flood protection agencies)

•Solution: research in progress…

Visualization of relevant information to

effectively support the decision making process

•Solution•Open domain-agnostic design (metadata and public APIs design are crucial)

Adaptability to other domains (e.g. monitoring

of communication infrastructure)

Challenges: visualization and decision support

Page 27: EScience challenges in levees monitoring - lessons from "flood" projects Marian Bubak Department of Computer Science AGH University of Science and Technology

Challenges: execution management

•Solution•Monitored area divided into sections •Managed by multiple instances of a Monitoring Application, dynamically deployed on-demand

Scale up to 100s-1000s

kilometers of levees

•Solution•Dynamic provisioning of resources from private or public clouds•Autoscaling algorithms and policies

Highly variable resource demands:

from very low in standby mode to

high in threat assessment mode

Page 28: EScience challenges in levees monitoring - lessons from "flood" projects Marian Bubak Department of Computer Science AGH University of Science and Technology

Challenges: data management

•Solution•Multiple data stores and models to address diverse needs

Diverse data sets (spatial, time series, binary,

metadata) and data usage patterns

•Solution•Big data infrastructure•Map-Reduce data search

Data-intensive processing

Threat level evaluation scenario: up to 130 GB of data to search per

1km of a levee

Page 29: EScience challenges in levees monitoring - lessons from "flood" projects Marian Bubak Department of Computer Science AGH University of Science and Technology

Urgent computing scenario

Goal: Assess flood risk for a large set levees by a specified deadlineSolution: dynamic provisioning of cloud resources

• A user:– Target area for flood threat assessment– Time window size for current measurements– Deadline to get results

• The system:– Generates workflow representing all required computations and data

dependencies– Plans the workflow execution so as to meet the deadline– Runs the workflow– Monitors its execution and reconfigures resource allocation if needed

Page 30: EScience challenges in levees monitoring - lessons from "flood" projects Marian Bubak Department of Computer Science AGH University of Science and Technology

Levee breach threat assessment

Page 31: EScience challenges in levees monitoring - lessons from "flood" projects Marian Bubak Department of Computer Science AGH University of Science and Technology

Implementation of urgent computing

Page 32: EScience challenges in levees monitoring - lessons from "flood" projects Marian Bubak Department of Computer Science AGH University of Science and Technology

Resource provisioning model

• Bag-of-tasks model– Selection of dominating tasks– Uniform task runtimes

• Performance model: T = f (v, d, s, …)– T – total computing time– v – number of VMs– d – time window in days– s – number of tasks (sections)

Page 33: EScience challenges in levees monitoring - lessons from "flood" projects Marian Bubak Department of Computer Science AGH University of Science and Technology

Resource provisioning model

(1)

T – total computing timev – number of VMsd – time window in dayss – number of tasks (sections)

Parameters a, b, c to be determined experimentally

Solve eq. (1) to compute v (# of VMs) given T (deadline)

Page 34: EScience challenges in levees monitoring - lessons from "flood" projects Marian Bubak Department of Computer Science AGH University of Science and Technology

Simulations

• Setup: private cloud infrastructure– a node with 8 cores (Xeon E5-2650)– virtual machines (1VCPU, 512MB RAM)– data for simulated scenarios (244MB total) on local

disks• Simulations:

– 1-1024 sections– 1-16 VMs– 1-7 days time window

• Warmup tasks:

Page 35: EScience challenges in levees monitoring - lessons from "flood" projects Marian Bubak Department of Computer Science AGH University of Science and Technology

Resource provisioning - results

• Warmup tasks clearly separated as outliers

• Linear functions• Parameters a, b, c determined using non-

linear fit• The model fits well to the data

War

mup{

a = 6.53b = 9.41c = 31.71

1024 sections

128 sections

Page 36: EScience challenges in levees monitoring - lessons from "flood" projects Marian Bubak Department of Computer Science AGH University of Science and Technology

Clouds for urgent computing (1/2)

• Elasticity– On-demand provisioning of VMs – Job prioritization and preemption

• Reliability– Public cloud services are specifically designed to support systems with

high availability demands – Amazon: only five major outages in the years 2010-2013 (only one for

more than 6 h)• Safety

– Serious natural disaster may damage a local computing infrastructure– Public clouds as an emergency computing infrastructure– Data safety: public clouds as a reliable storage infrastructure for

important but not sensitive data (example: pre-simulated scenarios data sets)

Page 37: EScience challenges in levees monitoring - lessons from "flood" projects Marian Bubak Department of Computer Science AGH University of Science and Technology

Clouds for urgent computing (2/2)

• Cost-effectiveness– Decision support systems for natural disasters generate

‘spiky’ workloads: perfect cloud use case – Cheaper than maintenance of a dedicated infrastructure – Day-to-day operation can be handled by a relatively small,

low-cost on-premises infrastructure• Performance?

– Bag-of-tasks applications such as scenario identification perfectly fit the cloud

– What about CPU- and communication-intensive tightly-coupled simulations? HPC-in-the-Cloud is an emerging trend.

Page 38: EScience challenges in levees monitoring - lessons from "flood" projects Marian Bubak Department of Computer Science AGH University of Science and Technology

Summary (1/2)

• Environmental models results in complex applications – collaborative – multi-scale, multi-domain, – time-critical– With data and resource intensive scenarios

• We have contributed to – methods and tools for environmental computing – advanced problem solving environments, virtual

laboratories– compositions of resources into complex scenarios– ccommodation to ”spiky” behavior (variable workload)

Page 39: EScience challenges in levees monitoring - lessons from "flood" projects Marian Bubak Department of Computer Science AGH University of Science and Technology

Summary (2/2)

• We have addressed– Complex distributed systems– Coordination of execution (workflow)– Monitoring and management of services– allocation of resources to services– fault tolerance– provenance tracking

• Sustainability issue - supporting technologies

Page 40: EScience challenges in levees monitoring - lessons from "flood" projects Marian Bubak Department of Computer Science AGH University of Science and Technology

More at

http://www.urbanflood.eu

http://www.ismop.edu.pl

http://dice.cyfronet.pl

[email protected]

Page 41: EScience challenges in levees monitoring - lessons from "flood" projects Marian Bubak Department of Computer Science AGH University of Science and Technology

Acknowledgements

This research was supported by the National Centre for Research and

Development (NCBiR) under Grant No. PBS1/B9/18/2013.