Optimizing Excited-State Electronic-Structure Codes for Intel Knights Landing: A Case Study on the BerkeleyGW Software

Derek Vigil-Fowler, Jack Deslippe, Felipe da Jornada, Taylor Barnes, Nathan Wichmann, Karthik Raman, Ruchira Sasanka, Steven Louie

Research output: Contribution to conferencePaperpeer-review

2 Scopus Citations


We profile and optimize calculations performed with the BerkeleyGW [2,3] code on the Xeon-Phi architecture. BerkeleyGW depends both on hand-tuned critical kernels as well as on BLAS and FFT libraries. We describe the optimization process and performance improvements achieved. We discuss a layered parallelization strategy to take advantage of vector, thread and node-level parallelism. We discuss locality changes (including the consequence of the lack of L3 cache) and effective use of the on-package high-bandwidth memory. We show preliminary results on Knights-Landing including a roofline study of code performance before and after a number of optimizations. We find that the GW method is particularly well-suited for many-core architectures due to the ability to exploit a large amount of parallelism over plane-wave components, band-pairs, and frequencies.


ConferenceInternational Workshops on High Performance Computing, ISC High Performance 2016 and Workshop on 2nd International Workshop on Communication Architectures at Extreme Scale, ExaComm 2016, Workshop on Exascale Multi/Many Core Computing Systems, E-MuCoCoS 2016, HPC I/O in the Data Center, HPC-IODC 2016, Application Performance on Intel Xeon Phi – Being Prepared for KNL and Beyond, IXPUG 2016, International Workshop on OpenPOWER for HPC, IWOPH 2016, International Workshop on Performance Portable Programming Models for Accelerators, P^3MA 2016, Workshop on Virtualization in High-Performance Cloud Computing, VHPC 2016, Workshop on Performance and Scalability of Storage Systems, WOPSSS 2016

Bibliographical note

Publisher Copyright:
© Springer International Publishing AG 2016.

NREL Publication Number

  • NREL/CP-5K00-67446


  • BerkeleyGW
  • optimization
  • performance


Dive into the research topics of 'Optimizing Excited-State Electronic-Structure Codes for Intel Knights Landing: A Case Study on the BerkeleyGW Software'. Together they form a unique fingerprint.

Cite this