BEGIN:VCALENDAR
VERSION:2.0
PRODID:Linklings LLC
BEGIN:VTIMEZONE
TZID:Europe/Stockholm
X-LIC-LOCATION:Europe/Stockholm
BEGIN:DAYLIGHT
TZOFFSETFROM:+0100
TZOFFSETTO:+0200
TZNAME:CEST
DTSTART:19700308T020000
RRULE:FREQ=YEARLY;BYMONTH=3;BYDAY=-1SU
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:+0200
TZOFFSETTO:+0100
TZNAME:CET
DTSTART:19701101T020000
RRULE:FREQ=YEARLY;BYMONTH=10;BYDAY=-1SU
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTAMP:20191206T142451Z
LOCATION:Analog 1\, 2
DTSTART;TZID=Europe/Stockholm:20180625T130500
DTEND;TZID=Europe/Stockholm:20180625T130900
UID:isc_hpc_ISC High Performance 2018_sess103_phd131@linklings.com
SUMMARY:(PhD01) Modelling of Dynamic Network Objects: New Approaches and A
 daptation Challenges for Future HPC Systems
DESCRIPTION:PhD Forum\n\n(PhD01) Modelling of Dynamic Network Objects: New
  Approaches and Adaptation Challenges for Future HPC Systems\n\nKushnarenk
 o\n\nModelling complex dynamic systems is a research topic for many years 
 and applied to design, automatization, monitoring, control, modelling and 
 simulation in many domains such as mechanical engineering, coal mining, me
 tallurgy. The high complexity of the simulation model is caused by their n
 onlinearity, spatial distribution and concurrence of the process parameter
 s. Furthermore, they typically grow large  and are highly time variant . A
 s a result of their complexity there are still sequential solvers used in 
 practice that cannot utilise at all modern or future HPC system architectu
 res and limit significantly the problem size and dynamism that can be addr
 essed.\nAs the existing solvers have been designed with homogeneous hardwa
 re and sequential processing in mind they cannot be easily parallelized. T
 his applies in particular for heterogeneous multi-processing systems. \nWi
 thin this PhD the goal is to apply new numerical methods to this problem s
 pace and target also for an increased abstraction from the underlying hard
 ware by utilising frameworks such as AnyDSL [1]. \nThe chosen use case are
  different models of coal mine air ventilation systems. Such systems can b
 e represented as dynamic network objects (DNO), allowing for a representat
 ion as graphs with different complexity levels. More specifically these ar
 e DNOs with concentrated or distributed parameters. \nIt is planned to bas
 e the parallel solver on block difference numerical methods (BDM). These m
 ethods allow to compute more than one simulation point per each iteration 
 step (e.g. 4 points, BDM4). The first experiments with sequential BDM-solv
 ers show high accuracy and convergence of these methods, less amount of co
 mputational steps, and good “accuracy-speed” relation, what brings a good 
 reason for further investigation in the direction of parallelization. An i
 nteresting research point is also a parallelization on the graph level, i.
 e. granularity of the parallel processes and their mapping to the physical
  compute resources (different parallelization levels).\nIn order to achiev
 e the anticipated simplicity it is proposed to express this algorithm as a
  domain specific language (DSL). The goal is to separate description (in c
 lear form for the end-user, an expert in the domain) and the adaptation to
  the specific architecture (which should be done by some hardware expert).
  As a starting point the AnyDSL framework will be investigated.\nThe poste
 r describes the simulation problem for complex dynamic systems, HPC-specif
 ic model adaptation challenges, and the basic concept for the proposed app
 roach using BDM and DSL. There are also first experimental results (for se
 quential BDM-solvers) and further planned research steps. \n[1] https://an
 ydsl.github.io/\n\nPasses: Conference Pass, Exhibition Pass, Heterogeneous
  Systems, Parallel Algorithms, Parallel Applications\n\nTag: Conference Pa
 ss, Exhibition Pass, Heterogeneous Systems, Parallel Algorithms, Parallel 
 Applications
URL:https://2018.isc-program.com/presentation/?id=phd131&sess=sess103
END:VEVENT
END:VCALENDAR