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:20191206T142452Z
LOCATION:Analog 1\, 2
DTSTART;TZID=Europe/Stockholm:20180625T133300
DTEND;TZID=Europe/Stockholm:20180625T133700
UID:isc_hpc_ISC High Performance 2018_sess103_phd112@linklings.com
SUMMARY:(PhD09) Bringing Neuroscience to HPC
DESCRIPTION:PhD Forum\n\n(PhD09) Bringing Neuroscience to HPC\n\nChatzikon
 stantis, Strydis, Kachris, Soudris\n\nThe poster aims at presenting a coll
 aborative effort between the National Technical University of Athens (NTUA
 ) and the Erasmus Medical Centre Rotterdam (EMC Rotterdam) to develop a pl
 atform for the neuroscientific community that allows conducting of computa
 tionally-heavy experiments regarding the functionality of the human brain.
 \nThe author of the poster and NTUA's Ph.D. student, George Chatzikonstant
 is, develops accelerated backend solvers for typical workloads in the doma
 in of computational neuroscience, which aim in particular at tackling biop
 hysically accurate and complicated neuron models at a scale previously tho
 ught unapproachable. With the help of modern HPC fabrics, such as Intel's 
 Xeon Phi manycore (co-)processors, network solvers with impressive speed a
 re attainable, allowing carrying out neuroscientific experiments of meanin
 gful accuracy and massive scale, offering a solid alternative to usual bla
 ck-box approaches. Our efforts so far show simulations of networks scaling
  up to 1 million neurons and 1 billion synapses as possible even on a sing
 le accelerator.\nStemming from the collaboration between the author and ot
 her developers of neuroscientific solutions on HPC fabrics (chiefly from E
 rasmus MC Rotterdam), a more complete solution in the domain of computatio
 nal neuroscience is being developed. BrainFrame is an online system which 
 utilizes a heterogeneous cloud of backends, with simulators running on GPU
 s, FPGAs, Xeon Phis and even traditional high-end CPUs. The computational 
 power of the heterogeneous cloud allows BrainFrame to provide the correct 
 computing fabric depending on the experiment the user desires to submit to
  the online platform. In our studies, we stress the importance of utilizin
 g the correct backend technology depending on the characteristics of each 
 experiment (network size, connectivity density, neuron model level of deta
 il).\nThe entire system is being developed in a dockerized environment in 
 order to be made modular; the goal of BrainFrame is to continuously integr
 ate new neuron models and corresponding backend solvers. We envision a pot
 ential heterogeneous system which allows the simulation of any neuronal mo
 del in massive quantities, without forcing the neuroscientist (nor his lab
 ) to spend time and money in difficult-to-obtain equipment, hardware and c
 oding experience.\nThe BrainFrame system is currently under development. A
  proof of concept has been already published in the  Journal of Neural Eng
 ineering 2017 titled "BrainFrame: a node-level heterogeneous accelerator p
 latform for neuron simulations". The aforementioned partners are currently
  working at releasing an alpha version where neuronal network simulators w
 ill be available and easily usable online through our hardware, as well as
  providing accelerated simulator alternatives (featuring orders of magnitu
 de improvements in performance) for a subset of the supported neuron model
 s.\n\nPasses: Conference Pass, Exhibition Pass, Brain Simulations, Contain
 erized HPC, HPC Accelerators, Heterogeneous Systems, Performance Analysis 
 and Optimization\n\nTag: Conference Pass, Exhibition Pass, Brain Simulatio
 ns, Containerized HPC, HPC Accelerators, Heterogeneous Systems, Performanc
 e Analysis and Optimization
URL:https://2018.isc-program.com/presentation/?id=phd112&sess=sess103
END:VEVENT
END:VCALENDAR