Senior HPC Applications Manager

Mountain View, CA, United States

Job Description

ASRC Federal is searching for a Senior HPC Applications Manager to support Inuteq LLC out of NASA AMES, CA

ASRC Federal, InuTeq proudly supports NASA's High Performance Computing Services program with our site in Mountain View, CA at the Ames Research Center. Make a DIFFERENCE on a program that supports 4 On-site Supercomputers totaling 18,000 nodes and 17+ combined petaflops. Our program provides High Performance Computing services throughout the HPC lifecycle for computational requirements, architecture, acquisition, and operations to federal government customers. Our employees embrace innovation and are committed to a culture of continuous, standards-driven, process improvement, and assimilation of industry best practices.

We have an immediate position for a Senior HPC Applications Manager to join our HPC team in Mountain View.US Citizenship is required.

Summary : The successful candidate will directly oversee four HPC related teams, known as subtasks, in the following areas:

HPC Application Services and Tools

HPC Cloud Computing

Data Science Applications supporting HPC Users

HPC Visualization

The Application Services and Tools technical subtask focuses on enhancing the performance and productivity of NASA's advanced modeling, simulation, and data analysis applications. The goal is to solve complex application problems, decrease time-to-solution, and increase the achievable scale and fidelity of the applications. The result is an increase of user and facility productivity, which in turn enhances NASA's science and engineering. This subtask requires high performance application expertise in many of the following areas: state-of-the-art parallel architectures; parallel programming paradigms in shared memory, distributed memory, and emerging systems with heterogeneous processors and multiple memory subsystems; scientific computing programming languages, such as C/C++ and Fortran; parallel computing models, such as Message Passing Interface (MPI) and OpenMP, as well as offloading models for accelerators, such as general purpose graphic processing units (GPGPUs); hybrid computational approaches using MPI with threads or other models; deep understanding of parallel software development and profiling tools; and related technologies and tools in support of data analysis, such as machine learning. The subtask also provides support for performance monitoring and benchmarking of various parallel systems available on site and evaluating emerging architectures and technologies available in various locations, including commercial cloud.

The Cloud sub-task seeks opportunities and capabilities beyond traditional high-performance computing (HPC), such as cloud computing, cloud application services, and machine learning, to meet the ever-increasing demand of various scientific and engineering requirements. The team continually evaluates its options for cost-effective delivery of HPC and other relevant services. This requirement focuses explicitly on cloud computing as a complementary approach for the traditional HPC

The purpose of the Data Science subtask is to evaluate and explore the use of Machine Learning (ML) technologies to identify methods that can be applied to science projects to enable scientific breakthroughs, to speed up the acquisition of science results, and/or to improve their fidelity.

The Visualization and Data Analysis subtask develops and employs advanced visualization and data analysis in both interactive and batch environments to enable users to derive increased value from their computations and observational data streams. The overarching goal of this technical area is to provide enhanced routes to discovery by way of scientifically rich visualizations, incisive data analysis techniques, interactive exploratory environments for large and complex data, and efficient data management strategies. The general approach will be application driven, using an evolving portfolio of customer science and engineering problems to develop and deploy new and existing visualization and data analysis techniques that maximally exploit the high-performance computational resources at the NAS.

Duties and Responsibilities:

Manage day-to-day performance of approximately 15 direct reports, providing technical leadership and oversight

Work with NASA customer counterparts to achieve contract objectives and requirements

Varying levels of direct technical contributions in support of HPC application and optimization activities

Plan and achieve project objectives; technically guide projects through completion and ensure all project objectives are met within target time frames.

Partner with development and operations teams to develop automation solutions

Location

NASA/AMES, MOFFETT FIELD-CA

Requirements

Advanced degree in Computer Science or related HPC dependent field

A minimum of 10 years of meaningful experience developing system software in heterogeneous, multi-platform HPC environments

Solid understanding of the software development process, including requirements, use cases, design, coding, documentation, and testing of scalable, distributed applications in a Linux environment

Demonstrated experience managing projects using various industry best practices and tools

Strong ability to analyze, debug and maintain the integrity of an existing code base

Experience working with HPC applications and proficiency in at least C, C++, or Fortran

Superior scripting skills and excellent attention to detail; proficiency in at least Python, Perl, or Bash

Strong ability to interact with customers to understand needs, elicit requirements, and get feedback on prototype solutions

Excellent communication and people skills; excellent time management and organizational skills

Track record of delivering commercial quality software on schedule with excellent quality through multiple release cycles

Proficiency at technical writing

Preferred Skills:

Ph.D. Computer Science or related HPC dependent field

Project Management Professional Certification

Background in rendering techniques e.g., ray tracing and graphics standards e.g., OpenGL

Proficiency with analysis and problem-solving skills for debugging and optimization of applications.

Proficiency with OpenMP and Message Passing Interface (MPI) programming.

Proficiency with Lustre and InfiniBand

Experience with Machine Learning technologies such as TensorFlow, PyTorch

Experience with cloud technologies (AWS, Azure, GCP), OpenStack or Kubernetes is a plus

GPU experience, e.g., CUDA programming a plus

EEO Statement

ASRC Federal and its Subsidiaries are Equal Opportunity / Affirmative Action employers. All qualified applicants will receive consideration for employment without regard to race, gender, color, age, sexual orientation, gender identification, national origin, religion, marital status, ancestry, citizenship, disability, protected veteran status, or any other factor prohibited by applicable law.

EEO Statement

ASRC Federal and its Subsidiaries are Equal Opportunity / Affirmative Action employers. All qualified applicants will receive consideration for employment without regard to race, gender, color, age, sexual orientation, gender identification, national origin, religion, marital status, ancestry, citizenship, disability, protected veteran status, or any other factor prohibited by applicable law.

#J-18808-Ljbffr