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:20191206T142523Z
LOCATION:Substanz 1\, 2
DTSTART;TZID=Europe/Stockholm:20180626T083000
DTEND;TZID=Europe/Stockholm:20180626T100000
UID:isc_hpc_ISC High Performance 2018_sess113@linklings.com
SUMMARY:Research Posters Session
DESCRIPTION:Research Poster\n\n(RP12) A user-controlled GGDML Code Transla
 tion Technique for Performance Portability of Earth System Models\n\nJum'a
 h, Kunkel\n\nDemand for high-performance computing is increasing in earth 
 system modeling, and in natural sciences in general. Unfortunately, automa
 tic optimizations done by compilers are not enough to make use of target m
 achines' capabilities. Manual code adjustments are mandatory to exploit ha
 rdware capabiliti...\n\n---------------------\n(RP05) Multicore Platform E
 fficiency Across Remote Sensing Applications\n\nTyutlyaeva, Moskovsky, Odi
 ntsov, Konyukhov, Poyda...\n\nA wide range of modern system architectures 
 and platforms targeted for different algorithms and application areas is n
 ow available. \nEven general-purpose systems have advantages in some compu
 tation areas and bottlenecks in another. Scientific applications on specif
 ic areas, on the other hand, have d...\n\n---------------------\n(RP02) Ea
 rly Evaluation of a New Vector Processor SX-Aurora TSUBASA\n\nKomatsu, Mom
 ose, Isobe, Sato, Musa...\n\nA brand-new vector supercomputer, SX-Aurora T
 SUBASA, has been launched. It has a newly developed Vector Engine(VE) proc
 essor. VE is developed to achieve a high sustained performance by powerful
  vector processing and a high memory bandwidth.\n\nVE is equipped with eig
 ht vector-cores, each of which exec...\n\n---------------------\n(RP09) iR
 eceptor: A Platform for Exploring and Analyzing Antibody/B-cell and T-cell
  Receptor Repertoire Data across Federated Repositories\n\nCorrie, Breden\
 n\nIndividuals produce an immense repertoire of antibody/B-cell and T-cell
  receptor sequences in order to recognize and destroy a diverse array of p
 athogens.  For example, it has been estimated that the number of possible 
 human B-cell sequences is greater than 10^13, and each person produces 10^
 8 - 10^...\n\n---------------------\n(RP18) BONSAI (Benchtesting OpeN Soft
 ware Autotuning Infrastructure)\n\nTsai, Bachstein, Luszczek, Kurzak, Anzt
 ...\n\nThe goal of the BONSAI (Benchtesting OpeN Software Autotuning Infra
 structure) project is to develop a software infrastructure for using paral
 lel hybrid systems at any scale to carry out large, concurrent autotuning 
 sweeps in order to dramatically accelerate the optimization process of com
 putational ...\n\n---------------------\n(RP13) Cross-architectural Modell
 ing of Power Consumption Using Neural Networks\n\nPuzovic, Lee, Elisseev\n
 \nOn the path to Exascale, goal of High Performance Computing (HPC) to ach
 ieve maximum performance, becomes a goal of achieving maximum performance 
 under strict power constraint. Novel approaches to hardware and software c
 o-design of modern HPC systems have to be developed to address such challe
 nges. ...\n\n---------------------\n(RP07) GPU-accelerated Simulation of E
 lastic Wave Propagation\n\nKadlubiak, Jaroš, Treeby\n\nModeling of ultraso
 und waves propagation in hard biological materials such as bones and the s
 kull has a rapidly growing area of applications, e.g., brain cancer treatm
 ent planning, deep brain neurostimulation and neuromodulation, and opening
  blood-brain barriers. Recently, we have developed a novel ...\n\n--------
 -------------\n(RP04) GPU-based Parallel PO-SWE Algorithm for the Design o
 f Large-sized Dual-Reflector Antennas\n\nMatsuo, Gocho, Ozaki, Katase, Ado
 \n\nLarge-sized dual-reflector antennas are analyzed to evaluate their per
 formance through long-time simulation using some accurate EM (electromagne
 tic) methods based on PO (physical optics). To reduce the time/term for th
 e simulation/design, we focused on the parallelization of the fast and acc
 urate hy...\n\n---------------------\n(RP15) Automatic Classification of S
 ystem Logs\n\nGhiasvand, Ciorba\n\nSystem logs (syslogs) are a valuable so
 urce of information for analyzing computing systems behavior.\nThe message
  part of each syslog entry includes detailed information about its respect
 ive event.\nSyslog is the de-facto logging protocol of high-performance co
 mputing systems.\nThe standard draft of th...\n\n---------------------\n(R
 P11) Task-Parallel Domain-wise Multifrontal Solver for High-Performance Fi
 nite Element Analysis on Multi-core Systems\n\nKim, Lee, Cho, Park, Kim\n\
 nThe domain-wise multifrontal solver (DMFS) is a high-performance sparse d
 irect solver especially designed for efficient finite element analysis and
  it has many successful applications to commercial and research finite ele
 ment analysis codes. However, the DMFS has limited scalability due to simp
 le st...\n\n---------------------\n(RP16) HuronFS : Hierarchical, User-lev
 el and On-demand Burst Buffer File System\n\nXu, Sato, Matsuoka\n\nIn HPC 
 centers, the performance of the parallel file systems (PFSs) cannot match 
 the dramatic leaps in computational performance, resulting in degradation 
 in the performance of data-intensive applications. Moreover, I/O interfere
 nce from other users further exacerbates the situation. Burst buffers, ...
 \n\n---------------------\n(RP03) HPC with Unstructured Meshes on Novel Ar
 chitectures\n\nLumi, Haase\n\nThis poster focuses on many-core parallel so
 lvers for discretized linear and non-linear partial differential equations
  in the context of cardiovascular applications. Analyses is performed on a
  set of applications with unstructured discretization based on tetrahedral
  meshes being considered as part o...\n\n---------------------\n(RP17) Eve
 nt Driven Asynchronous Tasks (EDAT)\n\nBrown\n\nIn task based models, by r
 ethinking parallelism in the paradigm of tasks one reduces synchronisation
  and decouples the management of parallelism from computation. However exi
 sting models typically rely on shared memory, where the programmer express
 es input and output dependencies of tasks based upon...\n\n---------------
 ------\n(RP14) Automatic Generation of Full-Set Batched BLAS\n\nHirota, Mu
 kunoki, Imamura\n\nBatched Basic Linear Algebra Subprograms (batched BLAS)
  is a new BLAS interface which computes multiple independent BLAS operatio
 ns as a single subroutine. On many-core processors, a small size problem m
 ay not utilize the computation power of all the cores. Batched BLAS is a s
 olution to utilize many...\n\n---------------------\nClosing & Transition 
 to the Poster Reception\n\nMüller, Chandrasekaran\n\n---------------------
 \n(RP01) Can Unified-Memory support on Pascal and Volta GPUs enable Out-of
 -Core DNN Training?\n\nAwan, Chu, Subramoni, Lu, Panda\n\nExisting Deep Ne
 ural Network (DNN) training frameworks like Caffe and CNTK\ncannot train l
 arge DNNs that do not fit the GPU memory without explicit memory\nmanageme
 nt schemes. In this poster, we propose OC-DNN - a novel\nOut-of-Core DNN t
 raining framework that\nexploits the new Unified Memory (UM) featu...\n\n-
 --------------------\n(RP10) Deep Learning Hardware Accelerates Fused Disc
 ontinuous Galerkin Simulations\n\nHeinecke, Breuer, Qiu, Cui\n\nIn recent 
 years the compute/memory balance of processors has been continuously shift
 ing towards compute. Especially the rise of Deep Learning, which is based 
 on matrix multiplication, accelerated this path, especially in terms of si
 ngle precision (FP32) and lower precision compute. An important rese...\n\
 n---------------------\n(RP08) Using GPU's FP16 Tensor Cores Arithmetic to
  Accelerate Mixed-Precision Iterative Refinement Solvers and Reduce Energy
  Consumption\n\nHaidar, Tomov, Abdelfattah, Zounon, Dongarra\n\nThe use of
  low-precision arithmetic in mixed-precision computing methods has been a 
 powerful tool to accelerate numerous scientific computing applications. Ar
 tificial intelligence (AI) in particular has pushed this to current extrem
 es, making use of half-precision floating-point arithmetic (FP16) in...\n\
 n---------------------\nIntroduction Research Posters Session\n\nMüller, C
 handrasekaran\n\n---------------------\n(RP06) A Deep Learning Tool for fa
 st Simulation\n\nVallecorsa\n\nMachine Learning techniques have been used 
 in different applications by the HEP community: in this talk, we discuss t
 he case of detector simulation. The need for simulated events, expected in
  the future for LHC experiments and their High Luminosity upgrades, is inc
 reasing dramatically and requires n...\n\n\nPasses: Conference Pass
END:VEVENT
END:VCALENDAR