BEGIN:VCALENDAR
VERSION:2.0
PRODID:Linklings LLC
BEGIN:VTIMEZONE
TZID:America/Denver
X-LIC-LOCATION:America/Denver
BEGIN:DAYLIGHT
TZOFFSETFROM:-0700
TZOFFSETTO:-0600
TZNAME:MDT
DTSTART:19700308T020000
RRULE:FREQ=YEARLY;BYMONTH=3;BYDAY=2SU
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:-0600
TZOFFSETTO:-0700
TZNAME:MST
DTSTART:19701101T020000
RRULE:FREQ=YEARLY;BYMONTH=11;BYDAY=1SU
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTAMP:20240116T185911Z
LOCATION:E Concourse
DTSTART;TZID=America/Denver:20231114T100000
DTEND;TZID=America/Denver:20231114T170000
UID:submissions.supercomputing.org_SC23_sess290_drs115@linklings.com
SUMMARY:Interactive In-Situ Visualization of Large Distributed Volume Data
DESCRIPTION:Doctoral Showcase, Posters\n\nAryaman Gupta (Technical Univers
 ity Dresden, Center for Systems Biology Dresden (CSBD))\n\nLarge distribut
 ed volume data are routinely produced in numerical simulations and experim
 ents. In-situ visualization, the visualization of simulation or experiment
  data as it is generated, enables simulation steering and experiment contr
 ol, which helps scientists gain an intuitive understanding of the studied 
 phenomena. Such data exploration requires interactive visualization with s
 mooth viewpoint changes and zooming to convey depth perception and spatial
  understanding. As data sizes increase, this becomes increasingly challeng
 ing. \n\nThis thesis presents an end-to-end solution for interactive in-si
 tu visualization on distributed computers based on novel extensions to the
  Volumetric Depth Image (VDI) representation. VDIs are view-dependent, com
 pact representations of volume data that can be rendered faster than the o
 riginal data.\n\nWe propose the first algorithm to generate VDIs on distri
 buted 3D data, using sort-last parallel compositing to scale to large data
  sizes. Scalability is achieved by a novel compact in-memory representatio
 n of VDIs that exploits sparsity and optimizes performance. We also propos
 e a low-latency architecture for sharing data and hardware resources with 
 a running simulation. The resulting VDI is streamed for remote interactive
  visualization.\n\nWe provide a novel raycasting algorithm for rendering s
 treamed VDIs, significantly outperforming existing solutions. We exploit p
 roperties of perspective projection to minimize calculations in the GPU ke
 rnel and leverage spatial smoothness in the data to minimize memory access
 es.\n\nThe quality and performance of the approach are evaluated on multip
 le datasets, showing that the approach outperforms state-of-the-art techni
 ques for visualizing large distributed volume data. The contributions are 
 implemented as extensions to established open-source tools.\n\nTag: Data A
 nalysis, Visualization, and Storage\n\nRegistration Category: Tech Program
  Reg Pass, Exhibits Reg Pass
END:VEVENT
END:VCALENDAR