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:20250822T115805Z
LOCATION:Campussaal - Plenary Room
DTSTART;TZID=Europe/Stockholm:20250616T102000
DTEND;TZID=Europe/Stockholm:20250616T105000
UID:submissions.pasc-conference.org_PASC25_sess149_pos114@linklings.com
SUMMARY:P19 - A GPU-Accelerated Implementation of Spectrum Slicing for Pla
 ne-Wave Density Functional Theory in ABINIT
DESCRIPTION:Ioanna-Maria Lygatsika and Marc Torrent (CEA, Université Paris
 -Saclay)\n\nWe consider the problem of accelerating the iterative diagonal
 ization of Hamiltonian operators for electronic structure calculations in 
 plane-wave Density Functional Theory. The complexity bottleneck of existin
 g subspace iteration schemes is that the Rayleigh-Ritz procedure for extra
 cting eigenvectors from a subspace has limited scalability with the number
  of processors. On the other hand, polynomial filtering methods for constr
 ucting subspaces can greatly benefit from the parallel efficiency of Hamil
 tonian applications using batch processing. The latter feature becomes eve
 n more advantageous on Graphics Processing Unit architectures in particula
 r. For this reason, we focus on the spectrum slicing method, which is a sp
 ecial type of polynomial filtering allowing to apply the Rayleigh-Ritz ste
 p to fewer vectors. We present a new implementation of this scheme in the 
 ABINIT plane-wave code using GPUs. We also propose a complexity model that
  allows to achieve load balance in the number of filtering operations and 
 the number of vectors on which we apply the Rayleigh-Ritz step per slice. 
 The numerical performance of our implementation is compared to that of exi
 sting algorithms in ABINIT, in particular Chebyshev Filtering and Locally 
 Optimal Block Preconditioned Conjugate Gradient, for systems of up to ten 
 thousand bands.\n\nSession Chair: Chris Cantwell (Imperial College London)
 \n\n
END:VEVENT
END:VCALENDAR