Skip to main content

Compute Providers

Vantage supports three main categories of compute providers, each offering distinct advantages and use cases:

On-Premises Compute Providers​

On-premises compute providers allow you to integrate your existing physical or virtualized infrastructure with Vantage's workload management capabilities. This includes bare metal servers, private cloud environments, and traditional HPC clusters.

Key Features:

  • Direct hardware control and customization
  • Integration with existing SLURM, PBS, LSF, and SGE schedulers
  • Support for high-performance interconnects like InfiniBand
  • Custom security and compliance configurations

Best For:

  • Organizations with existing HPC infrastructure
  • Workloads requiring specific hardware configurations
  • Strict data residency and compliance requirements
  • High-performance computing with specialized networking

Public Cloud Compute Providers​

Public cloud providers offer scalable, on-demand computing resources from major cloud platforms including AWS, Azure, and Google Cloud Platform. Vantage integrates with these providers to enable seamless multi-cloud workload management.

Key Features:

  • Virtually unlimited scalability
  • Pay-per-use and reserved capacity pricing models
  • Wide variety of instance types (CPU, GPU, memory-optimized)
  • Global availability with multiple regions and zones

Best For:

  • Variable or unpredictable workloads
  • Rapid scaling requirements
  • Geographic distribution of workloads
  • Cost optimization through spot/preemptible instances

Vantage Partner Compute Providers​

Vantage Partners are certified providers offering pre-configured, managed compute environments with guaranteed performance, support, and seamless integration with the Vantage platform.

Key Features:

  • Turnkey HPC solutions with expert support
  • Guaranteed performance and availability SLAs
  • Sustainable and ethical computing options
  • Specialized configurations for AI/ML and scientific computing

Best For:

  • Organizations wanting managed HPC services
  • Specialized workloads requiring expert optimization
  • Sustainable computing initiatives
  • Rapid deployment without infrastructure management overhead

Reference Documentation Structure​

Each compute provider reference includes:

  • Technical specifications and supported configurations
  • Integration methods and setup procedures
  • Performance optimization guidelines
  • Security and compliance considerations
  • Cost management strategies
  • Monitoring and troubleshooting guidance
  • Best practices for production deployments

Getting Started​

Choose the appropriate reference documentation based on your compute requirements:

For implementation guides and step-by-step instructions, see the Compute Providers section of the documentation.

Multi-Provider Strategies​

Vantage enables hybrid and multi-cloud strategies that combine different provider types:

  • Hybrid Deployments: Combine on-premises and cloud resources
  • Burst Computing: Scale from on-premises to cloud during peak demand
  • Cost Optimization: Leverage different pricing models across providers
  • Workload Distribution: Route workloads to optimal compute environments

See Also​