Skip to Content.
Sympa Menu

bd-palaver - [Bd-palaver] Dry Run Talks for ICAP

bd-palaver AT lists.psi.ch

Subject: Bd-palaver mailing list

List archive

[Bd-palaver] Dry Run Talks for ICAP


Chronological Thread 
  • From: "Adelmann Andreas (PSI)" <andreas.adelmann AT psi.ch>
  • To: bd-palaver <bd-palaver AT lists.psi.ch>, "Tarkeshian Roxana (PSI)" <roxana.tarkeshian AT psi.ch>, "Ischebeck Rasmus (PSI)" <rasmus.ischebeck AT psi.ch>, "Gsell Achim (PSI)" <achim.gsell AT psi.ch>, "Kolano Anna (PSI)" <anna.kolano AT psi.ch>, "Stadler Kleeb Hans-Christian (PSI)" <hans-christian.stadler AT psi.ch>, "Suter Andreas (PSI)" <andreas.suter AT psi.ch>, "Feichtinger Derek Heinrich (PSI)" <derek.feichtinger AT psi.ch>
  • Subject: [Bd-palaver] Dry Run Talks for ICAP
  • Date: Thu, 1 Oct 2015 16:56:24 +0000
  • Accept-language: en-US, de-CH
  • List-archive: <https://lists.web.psi.ch/pipermail/bd-palaver/>
  • List-id: <bd-palaver.lists.psi.ch>

Dear Colleagues Uldis and Matthias

both got an oral contribution for ICAP 2015. 

There will be a try run:

October 5 at 14.00 in WBBA 008

Thanks for stopping by,  enjoying the talks and 
provide feedback.

Andreas 


ID: 1286 - MODWC3  Dynamic Kernel Scheduler (DKS) - Accelerating the Object Oriented Particle Accelerator Library (OPAL)
My Schedule    Add
Speaker    Uldis Locans (PSI, Villigen PSI; University of Latvia, Riga)
Authors    Uldis Locans (PSI, Villigen PSI; University of Latvia, Riga), Andreas Adelmann, Andreas Suter (PSI, Villigen PSI) 
Abstract    Hardware accelerators, such as graphical processing units (GPU) and Intel Many Integrated Core (MIC) processors, provide a huge performance potential for HPC applications. However, due to different hardware architectures and development frameworks, developing fast and manageable code for these devices is a challenging task. This development is especially challenging when integrating the usage of co-processors in large applications, such as OPAL, where changing the code to support every device is not feasible. DKS is being developed to provide a software layer between the host application and different co-processors. DKS separates device specific code from the host application and provides a simple interface to communicate and schedule task execution on the device. Algorithms in DKS are implemented using CUDA, OpenCL, and OpenMP to target different devices. The first version of DKS was integrated in OPAL to speed up the FFT based Poisson solver and Monte Carlo simulations for particle matter interaction. The concepts of DKS and its integration with OPAL will be presented, as well as results showing speedups in the range of 4x to 50x for OPAL simulations using GPU or Intel MIC.

ID: 1302 - THCBC4  Stationary Distributions in Cyclotrons with Higher Order Moments of the Charge Distribution

My Schedule    Add
Speaker    Matthias Frey (PSI, Villigen, Villigen)
Authors    Andreas Adelmann (PSI, Villigen PSI), Matthias Frey (PSI, Villigen, Villigen) 
Abstract    This paper covers the topic of finding matched, i.e. stationary, distributions for coasting beams that takes into account higher order moments of the charge distribution. We present a general approach based on Lie-Algebra and Truncated-Power-Series Algebra, where the non-linear space-charge maps up to k-th order are constructed. The physical input parameters are intensity, projected emittances, and the energy. Our general approach is firstly compared and benchmarked with the well-known linear theory applied to the PSI Injector-2 and the Ring Cyclotron. Secondly we compared the non-linear approach to PIC calculations. Open questions such as constraining the higher order moments will be discussed as well as applications beyond cyclotrons.

 


  • [Bd-palaver] Dry Run Talks for ICAP, Adelmann Andreas (PSI), 10/01/2015

Archive powered by MHonArc 2.6.19.

Top of Page