Oliver Gutsche

Oliver Gutsche


Telephone: (630) 840-8909    E-mail: gutsche@fnal.gov   Office Location: WH11W
Professional Affiliations: APS Member, DPG Member, Associate Editor of the Springer Journal “Computing and Software for Big Science


Scientific Interest

My current physics interests are focusing on proton-proton physics at the Large Hadron Collider (LHC) at CERN, Geneva, Switzerland. I am a member of the Compact Muon Solenoid (CMS) collaboration that built and operates one of the four detectors recording the world’s highest energy proton-proton collisions. I am interested in Beyond-Standard-Model (BSM) physics searches, especially in the area of dark matter. I find the techniques centered around long lived particles very intriguing, as they go beyond the default particle reconstruction that assumes that particles originate close to the primary collision point in the detector.

Since March 2019, I am the U.S. CMS Software and Computing Operations Program manager. My duties include enabling U.S. CMS physicists in their analysis activities by providing computing facilities at Fermilab and U.S. universities. The operations program also provides software for the CMS collaboration from the core software framework to execute the many simulation and reconstruction algorithms, to computing infrastructure software like data management and workflow management software.

Since 2020, I am also the technical lead for the Portable Parallelization Strategies (PPS) project in the DOE Center For Computational Excellence (CCE). The goal is to enable the HEP community to make efficient use of current and future HPC centers. These machines will be based on accelerators like GPUs, and experimental code has to run efficiently on them. This requires new programming paradigms. The PPS projects will evaluate technologies that allow to write accelerator code once and compile/execute on different architectures, may it be NVIDA, AMD, INTEL GPUs or even more exotic architectures.

In the longer term, I am planning to invest into machine learning techniques both for my analysis projects and to answer questions for the computing infrastructure. Machine learning is very intriguing to me, but needs to be treated carefully. One of the aspects that we need to understand is uncertainties related to using Machine Learning techniques. Paired with new techniques that allow for analysis of exabyte size datasets, my longer term goal is to enable a successful physics harvest in the HL-LHC era.