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:20200227T164249Z
LOCATION:Panorama 1
DTSTART;TZID=Europe/Stockholm:20190619T144500
DTEND;TZID=Europe/Stockholm:20190619T151500
UID:isc_hpc_ISC High Performance 2019_sess221_inv_sp152@linklings.com
SUMMARY:Analysing and Tuning the Performance of Graph Processing Algorithm
s: a Statistical Modeling Approach
DESCRIPTION:Focus Session\n\nAnalysing and Tuning the Performance of Graph
Processing Algorithms: a Statistical Modeling Approach\n\nVarbanescu\n\nL
arge-scale and complex graph processing applications form a challenging do
main for high-performance computing. Despite graph processing algorithms b
eing considered parallelism-unfriendly, the use of parallel architectures
like multi-core CPUs and GPUs have proven revolutionary for these applic
ations. However, analysing and modeling the performance of these algorithm
s on parallel platforms remains a challenge: the tight dependencies betwee
n platform, algorithm, and dataset are proven difficult to analytically de
termine, model, and feed back into the algorithm design.\n\nIn this work,
we present a comprehensive framework for graph processing performance anal
ysis, and further demonstrate its use for performance modeling and tuning.
Our solution is based on a statistical approach, and combines efficient m
odel training with accurate predictions. We are further able to use these
predictions to improve algorithm execution. Finally, we present the perfo
rmance analysis and tuning of two case-studies (BFS and PageRank), and dem
onstrate how to use performance modeling to obtain better implementations,
which clearly outperform state-of-the-art implementations.\n\nPasses: Con
ference Pass, Graph Algorithms, HPC Accelerators, Parallel Algorithms, Per
formance Analysis and Optimization\n\nTag: Conference Pass, Graph Algorith
ms, HPC Accelerators, Parallel Algorithms, Performance Analysis and Optimi
zation
URL:https://2019.isc-program.com/presentation/?id=inv_sp152&sess=sess221
END:VEVENT
END:VCALENDAR