Ajay Salve

Scientific Officer
Department of Theoretical Physics
Tata Institute of Fundamental Research

Since Oct 2003 - Current

Experienced in systems administration and management of High Performance Computing (HPC) clusters, supercomputing systems, and datacenter operations. Proven expertise in handling large-scale compute environments, ensuring high availability, performance optimization, and efficient resource utilization.


Key Responsibilities & Experience

• Managing 5 High Performance Computing (HPC) clusters with a combined capacity of 2300+ CPU cores, supporting parallel and distributed computing workloads.
• Administering MPI-based high-speed interconnects for efficient parallel job execution and low-latency communication.
• Overseeing parallel file system storage, ensuring high throughput, scalability, and data integrity for compute-intensive applications.
• Handling operations of a 750 TFLOPS Cray XC30 supercomputer, procured under the Indian Lattice Gauge Theory Initiative (ILGTI) project.
• Responsible for datacenter operations, including:
• Power and cooling management
• Rack layout and space optimization
• Monitoring uptime and system reliability
• Preventive and corrective maintenance coordination

Core Competencies

• HPC Cluster Deployment & Administration
• Supercomputing Systems Management
• MPI & Parallel Computing Environments
• Parallel File Systems (e.g., Lustre/GPFS)
• Datacenter Infrastructure & Operations
• Performance Monitoring & Optimization
• High Availability & Fault Tolerance

---