HPC On-Demand Pricing

Full Service HPC Offering for a Consistent and Predictable Low Price

Unlike other cloud service offerings, POD provides customers with a clear and transparent pricing model that allows them to plan for future budgets and requisition requests. POD provides immense value as it includes industry-leading support from our team of HPC experts, and there are no hidden charges for network transfers or other basic maintenance. Rather than bill per the hour, we price our services every three seconds, giving our customers timely, real-time billing that accurately reflects usage. POD includes no minimum charges and does not require a basic fee to open or maintain an account.

The POD reporting dashboard provides highly-detailed live reporting yet is intuitive and easy to read and share billing information. You can also review billing history or get a snapshot of billing status, compartmentalized by users, groups or projects.

Comparing Costs

Contact sales for EDU, GOV and volume pricing.

Number of Cores ($0.08 or $0.10)

Job Length (minutes or hours)


0 cores x 0 minutes = $0

Compute Cores

Unlike many competitors, we do not round up to the nearest hour; you pay for exactly what you use, measured down to the second.

  Description $/core/hr
Free Compute 24 Cores for 5 Min Maximum Jobs FREE
Intel® X5600 Series (Westmere) or
Intel® E5-2600 Series (Sandy Bridge) Compute
Bare-metal, QDR Infiniband interconnect, 10GigE data network, and minimum 4GB RAM per core. $0.08
Intel® E5-2600v3 Series (Haswell) Compute Bare-metal, QDR Infiniband interconnect, 10GigE data network, and minimum 6.4GB RAM per core. $0.10

GPU Compute

  Description $/GPU/hr
NVIDIA® Tesla® K40 GPU Compute Bare-metal, two NVIDIA® Tesla® K40 GPUs per server, QDR Infiniband interconnect, 10GigE data network, and 64GB RAM per node. $2.32

Storage

Storage is charged per GB, per month, for allocated storage, tracked daily.

Storage1GB FREE
Additional Storage$0.10/GB/month

Login Nodes

Each user gets one free login node for access to the compute cluster. Larger, more powerful login nodes can also be launched on-demand and charged by the second.

pod.free 1 core, 256MB RAM FREE
pod.1x2 1 core, 2 GB RAM $0.09/hr
pod.2x4 2 core, 4 GB RAM $0.18/hr
pod.4x8 4 core, 8 GB RAM $0.36/hr