usatlas-hllhc-computing-l AT lists.bnl.gov
Subject: US ATLAS HL-LHC computing discussion
List archive
[Usatlas-hllhc-computing-l] Fwd: Opportunity to trial new powerful data analysis system in CSI
- From: Torre Wenaus <wenaus AT gmail.com>
- To: usatlas-hllhc-computing-l AT lists.bnl.gov
- Subject: [Usatlas-hllhc-computing-l] Fwd: Opportunity to trial new powerful data analysis system in CSI
- Date: Sun, 7 Oct 2018 10:44:01 +0200
From: Kleese Van Dam, Kerstin <kleese AT bnl.gov>
Date: Sun, Oct 7, 2018 at 12:23 AM
Subject: Opportunity to trial new powerful data analysis system in CSI
To: CSI ML <CSIML AT bnl.gov>, Mueller, Berndt <bmueller AT bnl.gov>, Hill, John <hill AT bnl.gov>, Misewich, James <misewich AT bnl.gov>, McSweeney, Sean <smcsweeney AT bnl.gov>, Jakoncic, Jean <jjakoncic AT bnl.gov>, Rajagopalan, Srini <srinir AT bnl.gov>, Izubuchi, Taku <izubuchi AT bnl.gov>, Hybertsen, Mark <mhyberts AT bnl.gov>
Cc: Hoisie, Adolfy <ahoisie AT bnl.gov>, D'Imperio, Nicholas <dimperio AT bnl.gov>, Kleese Van Dam, Kerstin <kleese AT bnl.gov>
Dear all,
CSI is excited to announce that we have just taken delivery of one of the first IBM/NVIDIA DGX-2 systems to come from the factory . This is a GPU based system specifically for data intensive workloads with high I/O capacity, plenty of memory and 16 Volta GPU's. The system will deliver a peek performance of 2 Petaflop (all scientific computing systems at BNL currently deliver 4 PetaFlop in total).
We are keen to test the system and explore how its performance could help the different workloads around the lab. To this end we invite you to get in contact with us if you have a data intensive workflow - both large data and real time requirements - we would like to work with you to see what can be done.
The plan is to make the system available for testing for the foreseeable future, before we will take it into production.
Best wishes,
Kerstin
https://www.nvidia.com/en-us/data-center/dgx-2/
Experience new levels of AI speed and scale with NVIDIA ® DGX-2 ™, the first 2 petaFLOPS system that combines 16 fully interconnected GPUs for 10X the deep learning performance.It’s powered by NVIDIA ® DGX ™ software and a scalable architecture built on NVIDIA
NVSwitch, so you can take on the world’s most complex AI challenges.
|
- [Usatlas-hllhc-computing-l] Fwd: Opportunity to trial new powerful data analysis system in CSI, Torre Wenaus, 10/07/2018
Archive powered by MHonArc 2.6.24.