BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//pretalx//pretalx.com//pyconde-pydata-2026//speaker//BELFDP
BEGIN:VTIMEZONE
TZID:CET
BEGIN:STANDARD
DTSTART:20001029T040000
RRULE:FREQ=YEARLY;BYDAY=-1SU;BYMONTH=10
TZNAME:CET
TZOFFSETFROM:+0200
TZOFFSETTO:+0100
END:STANDARD
BEGIN:DAYLIGHT
DTSTART:20000326T030000
RRULE:FREQ=YEARLY;BYDAY=-1SU;BYMONTH=3
TZNAME:CEST
TZOFFSETFROM:+0100
TZOFFSETTO:+0200
END:DAYLIGHT
END:VTIMEZONE
BEGIN:VEVENT
UID:pretalx-pyconde-pydata-2026-K9LCNQ@pretalx.com
DTSTART;TZID=CET:20260415T173500
DTEND;TZID=CET:20260415T180500
DESCRIPTION:Python’s scientific stack (NumPy/SciPy) is often confined to 
 single-node execution. When datasets exceed local memory\, researchers fac
 e a steep learning curve\, typically choosing between complex manual distr
 ibution or the overhead of task-parallel frameworks.\n\nIn this talk\, we 
 introduce [Heat](https://github.com/helmholtz-analytics/heat)\, an open-so
 urce distributed tensor framework designed to bring high-performance compu
 ting (HPC) capabilities to the scientific Python ecosystem. Built on PyTor
 ch and mpi4py\, Heat implements a data-parallel model that allows users to
  process massive datasets across multi-node\, multi-GPU clusters (includin
 g AMD GPUs) with minimal code changes.\n\nWe will discuss the design and a
 rchitecture enabling "transparent distribution":\n\n- Heat’s distributed
  n-dimensional array for data partitioning and communication under the hoo
 d\;\n- The synergy of PyTorch as a high-performance compute engine and MPI
  for efficient\, low-latency communication\;\n- Scaling efficiency\, encom
 passing both strong and weak scaling for memory-intensive operations\;\n- 
 Fundamental building blocks—from linear algebra to machine learning—re
 -implemented for distributed memory space.\n\nAttendees will learn how to 
 leverage the cumulative RAM of supercomputers without leaving the familiar
  NumPy-like interface\, effectively removing the "memory wall" for large-s
 cale scientific analytics.
DTSTAMP:20260412T141624Z
LOCATION:Europium [3rd Floor]
SUMMARY:Heat: scaling the Python scientific stack to HPC systems - Claudia 
 Comito\, Thomas Saupe
URL:https://pretalx.com/pyconde-pydata-2026/talk/K9LCNQ/
END:VEVENT
END:VCALENDAR