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:20260422T000612Z
LOCATION:702
DTSTART;TZID=America/Denver:20231112T140000
DTEND;TZID=America/Denver:20231112T173100
UID:submissions.supercomputing.org_SC23_sess436@linklings.com
SUMMARY:IA^3 2023 - 13th Workshop on Irregular Applications: Architectures
  & Algorithms
DESCRIPTION:A New Sparse GEneral Matrix-Matrix Multiplication Method for L
 ong Vector Architecture by Hierarchical Row Merging\n\nVector processors h
 ave become essential to high-performance computing in scientific and engin
 eering applications, especially in numerical calculations that leverage da
 ta parallelism.  With escalating computational demands, the efficient exec
 ution of Sparse GEneral Matrix-matrix Multiplication (SpGEM...\n\n\nHikaru
  Takayashiki, Hotaka Yagi, Hiroki Nishimoto, and Naoki Yoshifuji (Fixstars
  Corporation)\n---------------------\nFiltering Wasteful Vertex Visits in 
 Breadth-First Search\n\nBreadth-First Search (BFS) is a common building bl
 ock for several graph processing algorithms today. In this work, we highli
 ght that a large fraction of vertex visits across the network in distribut
 ed BFS  results in wasteful work. We investigate methods to identify and f
 ilter such wasteful cross-ne...\n\n\nPrachatos Mitra and Alexandros Daglis
  (Georgia Institute of Technology)\n---------------------\nAccelerating De
 ep Neural Network Guided MCTS Using Adaptive Parallelism\n\nDeep Neural Ne
 twork guided Monte-Carlo Tree Search (DNN-MCTS) is a powerful class of AI 
 algorithms. The DNN operations are highly parallelizable, but tree search 
 operations are sequential and often become the system bottleneck.  Existin
 g MCTS parallel schemes on CPU platforms either exploit data par...\n\n\nY
 uan Meng, Qian Wang, Tianxin Zu, and Viktor Prasanna (University of Southe
 rn California (USC))\n---------------------\nIA^3 – Concluding Remarks\n\n
 Antonino Tumeo, John Feo, and Sanjukta Bhowmick (Pacific Northwest Nationa
 l Laboratory (PNNL)) and Nesreen Ahmed (Intel Corporation)\n--------------
 -------\nTowards a Massive-Scale Distributed Neighborhood Graph Constructi
 on\n\nGraph-based approximate nearest neighbor algorithms have shown high 
 performance and quality.  However, such approaches require a large amount 
 of memory and still take a long time to construct high-quality nearest nei
 ghbor graphs (NNGs).  Using distributed memory systems is important when d
 ata is lar...\n\n\nKeita Iwabuchi, Trevor Steil, Benjamin Priest, Roger Pe
 arce, and Geoffrey Sanders (Lawrence Livermore National Laboratory (LLNL))
 \n---------------------\nTANGO: A GPU-Optimized Traceback Approach for Seq
 uence Alignment Algorithms\n\nSequence alignment algorithms play a central
  role in most bioinformatics software. However, porting these algorithms t
 o GPUs can be challenging due to their reliance on irregular memory access
  patterns and integer-heavy operations. Here we present TANGO, an optimize
 d GPU implementation of the Smith-...\n\n\nLeAnn Lindsey (University of Ut
 ah), Muhammad Haseeb (Lawrence Berkeley National Laboratory (LBNL)), Hari 
 Sundar (University of Utah), and Muaaz Awan (Lawrence Berkeley National La
 boratory (LBNL))\n---------------------\nA Parallel Algorithm for Updating
  a Multi-Objective Shortest Path in Large Dynamic Networks\n\nIn dynamic n
 etworks, where continuous topological changes are prevalent, it becomes pa
 ramount to find and update different graph properties without the computat
 ional burden of recalculating from the ground up. However finding or updat
 ing a multi-objective shortest path (MOSP) in such a network is ch...\n\n\
 nArindam Khanda, S. M. Shovan, and Sajal K. Das (Missouri University of Sc
 ience and Technology)\n---------------------\nIA^3 – Invited Talk\n\nTony 
 Brewer (Micron Technology Inc)\n---------------------\nIA^3 – Welcome and 
 Introduction\n\nAntonino Tumeo and John Feo (Pacific Northwest National La
 boratory (PNNL)), Sanjukta Bhowmick (University of North Texas), and Nesre
 en Ahmed (Intel Corporation)\n---------------------\ncuAlign: Scalable Net
 work Alignment on GPU Accelerators\n\nGiven two graphs, the objective of n
 etwork alignment is to find a one-to-one mapping of vertices in one graph 
 to vertices in the other, such that the number of overlaps is maximized. N
 etwork alignment is an important optimization problem with several applica
 tions in bioinformatics, computer vision a...\n\n\nLizhi Xiang (University
  of Utah), Arif Khan (Meta), S. M. Ferdous (Pacific Northwest National Lab
 oratory (PNNL)), SR Aravind (Meta), and Mahantesh Halappanavar (Pacific No
 rthwest National Laboratory (PNNL))\n---------------------\nIA^3 2023 – Af
 ternoon Break\n\nTag: Algorithms, Applications, Architecture and Networks\
 n\nRegistration Category: Workshop Reg Pass\n\nSession Chairs: Sanjukta Bh
 owmick (University of North Texas), Vito Giovanni Castellana (Pacific Nort
 hwest National Laboratory (PNNL)), John Feo (Pacific Northwest National La
 boratory (PNNL)), Marco Minutoli (Advanced Micro Devices), and Antonino Tu
 meo (Pacific Northwest National Laboratory (PNNL))
END:VEVENT
END:VCALENDAR