HPC Service and Management

The Research Office at MBZUAI delivers cutting-edge High-Performance Computing (HPC) capabilities—spanning both on-premises and cloud environments—to empower faculty and researchers in accelerating Research innovation.

Our on-premises HPC clusters feature diverse GPU and CPU architectures optimized for high-intensity and large-scale computational workloads, providing dedicated and reliable research environments. Complementing this, the Cloud HPC platform offers additional scalability, flexibility, and on-demand access to computing power, supporting diverse and dynamic research requirements.

Together, these platforms provide a unified, high-performance ecosystem that empowers researchers to compute without limits—enhancing efficiency, fostering collaboration, and driving breakthrough outcomes that position MBZUAI at the forefront of research excellence.

Mission To Serve

Empowering world-class research and innovation through powerful, accessible, and sustainable High-Performance Computing—across both on-premises and cloud platforms—enabling faculty and researchers to accelerate discovery, collaboration, and transformative outcomes, positioning MBZUAI at the forefront of research excellence.

Get in touch wih the teams

On-Perm HPC Cloud HPC

On-Premises HPC

Full-Lifecycle Cluster Management: Comprehensive deployment, configuration, optimization, and management of HPC clusters to ensure peak performance and uptime.

User and Research Project Onboarding: Personalized support to align computing environments with each project’s unique goals, ensuring rapid productivity from day one.

Performance Monitoring and Optimization: Continuous analysis and tuning to unlock peak efficiency and accelerate research outcomes.

System Maintenance and Hardware Evolution: Proactive maintenance, seamless upgrades, and expert hardware support to keep systems at the forefront of performance.

Data Stewardship: Secure, high-speed storage and resilient backup solutions that protect the integrity and longevity of research data.

Strategic Compute Consultation: Expert assessment and tailored recommendations to match research ambitions with the right computational power and architecture.

Technical Support and Troubleshooting: Dedicated technical assistance and rapid issue resolution to ensure uninterrupted research activity on the On-Prem compute resource.

Training Programs for Users and Faculty: Structured training sessions equip users and faculty with the skills to leverage HPC resources effectively, ensuring maximum productivity and research impact.

Quality Monitoring for Users and Faculty: We continuously monitor system performance and user experience, ensuring optimal resource allocation and identifying areas for improvement, fostering an efficient research environment.

Access Management for HPC Resources: Our team manages access permissions to ensure secure and appropriate use of HPC resources, supporting both individual and collaborative research efforts while maintaining system integrity.

Resource Scheduling and Job Queuing: We manage job scheduling and queuing, allowing users to submit and prioritize computational tasks efficiently, ensuring fair and balanced resource allocation across all users.

Custom HPC Solutions and Consultancy: We offer consultancy services to tailor HPC solutions to meet specific research requirements, helping users design and implement custom workflows and architecture of their unique needs.

Guidelines and Manuals

Available at: MBZUAI HPC Wiki (on-campus network required)

Get in Touch

Contact the On-Prem HPC team at hpc.admins@mbzuai.ac.ae
‍For access requests, GPU/CPU allocation, technical support and Data Center management.

Cloud HPC

Cloud HPC Infrastructure Deployment: End-to-end design, provisioning, and configuration of high-performance computing environments on leading cloud platforms, ensuring secure, scalable, and efficient operations

User and Research Project Onboarding: Tailored onboarding to align cloud compute environments with each project's research objectives, enabling rapid setup and seamless user & project onboarding

Performance Monitoring and Architecture Optimization: Continuous evaluation of cloud workloads to ensure optimal performance, resource utilization, and architectural efficiency for diverse research demands

Cost Estimation and Optimization: Data-driven cost analysis, forecasting, and proactive optimization strategies to maximize research output while minimizing cloud expenditure

Cloud Data Management, Security, and Compliance: Implementation of robust data governance frameworks ensuring data integrity, privacy, and adherence to institutional and global compliance standards

Technical Support and Troubleshooting: Dedicated technical assistance and rapid issue resolution to maintain uninterrupted research activity across cloud HPC environments