Senior Solution Architect HPC

Role: Senior Solution Architect HPC, Cloud-Native Systems (ITAR-Restricted Role)

Location: Remote

Position Overview

We are seeking a high-performance Senior Solution Architect to lead the convergence of traditional High-Performance Computing (HPC) environments with modern cloud-native architectures. This position is designated as ITAR-restricted, requiring candidates that are legally authorized to access and handle U.S. export-controlled technical data.

The architect will design, integrate, and optimize large-scale, containerized, hybrid HPC environments using technologies such as Docker, Mirantis, ELK Stack, and advanced batch schedulers. This role requires deep technical leadership, architectural vision, and hands-on experience supporting mission-critical computational workloads in secure, compliant environments.

Core Responsibilities

1. Architecture & Design

• Architect end-to-end hybrid cloud solutions integrating Mirantis Container Cloud with dedicated HPC clusters.
• Balance performance, elasticity, and compliance requirements across on-prem and cloud environments.
• Produce architecture documentation in adherence with ITAR export-controlled standards and review practices.

2. HPC Orchestration

• Design and implement HPC job scheduling strategies using Slurm, Volcano, LAVA, or similar technologies.
• Support deterministic resource allocation for AI/ML analytics, physics simulations, and scientific workloads.
• Ensure schedulers meet ITAR-restricted workload isolation and audit requirements.

3. Optimization & Performance Tuning

• Apply best practices for high-performance containerization: multi-stage builds, minimal base images, and resource tuning (CPU, GPU, Memory).
• Implement strategies to minimize overhead, ensure stability, and eliminate noisy-neighbor issues.

4. Centralized Observability

• Architect and operate an enterprise-grade ELK Stack (Elasticsearch, Logstash, Kibana) tuned for HPC-scale environments.
• Manage Index Lifecycle Management (ILM) for massive log throughput while preserving traceability for compliance audits.

5. Full-Stack Automation

• Build IaC-driven automation pipelines using Terraform, Ansible, and GitOps workflows.
• Automate deployment of Mirantis Kubernetes Engine (MKE) and integrated HPC schedulers within an ITAR-secured environment.

6. CI/CD Automation

• Implement robust CI/CD workflows using Jenkins, GitLab CI, Argo Workflows, or similar tools.
• Ensure pipelines comply with ITAR policies, including artifact access control, secure registries, and encrypted transport.

7. Hybrid Integration

• Architect integration between Kubernetes and traditional HPC schedulers.
• Enable advanced workloads requiring high-speed interconnects such as InfiniBand, RDMA, or GPU-accelerated clusters.

Required Technical Skills

Containers & Mirantis

• Expertise in Docker Runtime, Mirantis Kubernetes Engine (MKE), and Lens Desktop management.
• Deep experience designing containerized workloads for HPC environments.

HPC Schedulers

• Hands-on experience with Slurm, PBS, or Kubernetes-native batch schedulers such as Volcano.
• Knowledge of hierarchical priority queues, gang scheduling, and resource fairness algorithms.

ELK Stack Mastery

• Strong understanding of Logstash pipeline performance optimization, Elasticsearch sharding strategies, and Kibana visualization design.

Performance Tools

• Experience with NVIDIA Enroot/Pyxis or equivalent technologies supporting near bare-metal container performance.

Security & Compliance

• Implement secure registry solutions, TLS encryption, RBAC, and identity-driven access controls.
• Demonstrated experience supporting compliance frameworks including ITAR, NIST 800-53, or similar.

Experience & Qualifications

Professional Background

• 10+ years in systems architecture or engineering roles.
• 5+ years in HPC, Cloud Infrastructure, or enterprise-scale DevOps environments.

HPC Knowledge

• Understanding of MPI (Message Passing Interface), GPU compute workloads, low-latency networks, and distributed parallel frameworks.

Certifications

Preferred certifications include:

• Certified Kubernetes Administrator (CKA)
• Mirantis Kubernetes certifications
• Relevant security/compliance certifications (a plus)

Cloud Platforms

• Experience with AWS HPC environments (EKS, AWS Batch, EKS for Lustre, EC2 GPU-accelerated instances).