Research Computing Infrastructure Management

Complete HPC and Research Computing Visibility

Aravolta provides universities and research institutions with comprehensive visibility and management capabilities for their entire research computing infrastructure, from GPU clusters to traditional HPC systems. Integrate seamlessly with SLURM and other workload managers.

Faculty Research Lab Power Monitoring

Track power consumption at the individual lab and faculty level with our branch circuit monitoring system. Perfect for grant accounting, departmental chargebacks, and capacity planning:

• Per-Lab Power Metering:
  • Real-time power consumption by research group
  • Track GPU, CPU, and storage system usage
  • Automated billing for departmental chargebacks
  • Grant-specific cost allocation and reporting
• Faculty Self-Service Portal:
  • Dashboard for each research group
  • Real-time power and compute utilization
  • Historical usage trends and forecasting
  • Custom alerts for resource limits
• Department-Level Analytics:
  • Cross-lab resource utilization comparison
  • Energy efficiency benchmarking
  • Capacity planning for new faculty hires
  • Budget forecasting and cost optimization

SLURM Integration & HPC Workload Management

Aravolta integrates directly with SLURM and other popular HPC workload managers to provide comprehensive visibility into research computing resources:

• SLURM Integration Features:
  • Real-time job queue monitoring and analytics
  • Correlate power consumption with running jobs
  • Per-user and per-job energy accounting
  • Optimize job scheduling based on power availability
• Resource Optimization:
  • Identify underutilized nodes and resources
  • Balance workloads across compute resources
  • Power-aware job scheduling to reduce costs
  • GPU utilization tracking and optimization
• Research Impact Analytics:
  • Track compute resources by grant or project
  • Generate reports for funding agencies
  • Measure research output per watt
  • Support for federal reporting requirements (NSF, NIH, DOE)

Research Computing Facilities by Type

Manage all aspects of your research computing infrastructure with facility-specific monitoring and management:

• HPC Data Centers:
  • Traditional CPU clusters and supercomputers
  • GPU clusters for AI/ML research
  • High-speed networking and InfiniBand fabrics
  • Parallel file systems (Lustre, GPFS, BeeGFS)
• Department Server Rooms:
  • Individual lab servers and workstations
  • Department-specific compute resources
  • Legacy systems and specialized equipment
  • Environmental monitoring and cooling
• Edge Computing Labs:
  • Distributed research facilities across campus
  • Remote instrumentation and sensors
  • Field research data collection systems
  • Mobile computing infrastructure

Key Features for Higher Ed

• Grant and Project Tracking:
  • Per-grant resource usage and cost allocation
  • Automated reporting for funding agencies
  • Multi-year project tracking and forecasting
  • Compliance with federal grant requirements
• Multi-Tenant Support:
  • Department and lab-level isolation
  • Faculty self-service portals
  • Custom dashboards for different user roles
  • Secure data sharing between collaborators
• Research Continuity:
  • Predictive maintenance to prevent downtime
  • Automated backup and disaster recovery
  • Job checkpoint and restart capabilities
  • Resource availability forecasting
• Sustainability Reporting:
  • Carbon footprint tracking for research activities
  • Energy efficiency metrics and benchmarking
  • Support for campus sustainability initiatives
  • STARS and other green certification reporting