Sistemas Informáticos EuropeosSistemas Informáticos Europeos
  • Products
    • Digital Platforms
    • Software
    • Solutions
  • Services
    • Social
    • Storage
    • Security
    • Virtualization
  • Projects
    • HPC developments
    • I + D + i projects
  • News
    • SIE News
    • Success stories
    • SIE With science
    • Other news
  • About us
    • About Us
    • Contact

AMBER TEST WITH NEW GTX 1080Ti

6 April 2017HPC, HOME, UncategorizedRaúl Díaz

AMBER TEST WITH NEW GTX 1080Ti

The researcher Jordi Bujons from the CSIC-CID has been able to test the new NVDIA GTX 1080Ti, on the SIE Ladon G-1P Workstation, verifying a significant improvement in the tests carried out.

“For the tests, Amber 16 + update 6, CUDA 8.0 and Nvidia driver 318.13 were used, with Persistence Mode and Exclusive Mode activated.”
Summary of Results:
JAC_PRODUCTION_NVE – 23,558 atoms PME 4fs


JAC_PRODUCTION_NPT – 23,558 atoms PME 4fs


JAC_PRODUCTION_NVE – 23,558 atoms PME 2fs


JAC_PRODUCTION_NPT – 23,558 atoms PME 2fs


FACTOR_IX_PRODUCTION_NVE – 90,906 atoms PME


FACTOR_IX_PRODUCTION_NPT – 90,906 atoms PME


CELLULOSE_PRODUCTION_NVE – 408,609 atoms

PME CELLULOSE_PRODUCTION_NPT – 408,609 atoms

PME STMV_PRODUCTION_NPT – 1,067,095 atoms

PME TRPCAGE_PRODUCTION – 304 atoms

GB MYOGLOBIN_PRODUCTION – 2,492 atoms

GB NUCLEOSOME_PRODUCTION – 25,095 atoms GB
“The 1080Ti has generally performed somewhat better (5-6%) or similar to the XP.”
sis. 1 Sis. 2


1080Ti: ns/day: 203.732 18.623


980Ti: ns/day: 116.918 11.106

The 1080Ti in general has given excellent performances. In fact, Amber considers it to be the best value for money GPU ever, so far. The reason is that the GTX 1080Ti gives a very similar performance to the GTX Titan XP, with a much more adjusted price.

In the Amber 16 benchmarks, a price of 1.12 $ ns/day is considered and SIE has launched a Workstation that contributes in the complete system with this GPU, less than €5 ns/day. It is probably the most affordable and powerful Workstation we have released to date.
The official results of Amber 16 with the GTX 1080Ti card can be checked here.

Screenshot 2017-04-06 00.18.45

Tags: GPU CALCULATION, cluster, GPU COMPUTE, GPU, GPU Computing, hpc, hpcsie, Parallel computing

Leave your comment Cancel reply

Your email address will not be published. Required fields are marked with *

  • Legal notice
  • Corporate Social Responsibility
  • Web Builder and AI images by Kosta TS
© 2020 Sistemas Informáticos Europeos, All rights reserved.
We use cookies to ensure that we give you the best experience on our website. If you continue to use this site, we will assume that you agree with it.