Speed time to insight with elastic HPC as a service

HPE GreenLake for HPC makes it easier for enterprises of any size to get the benefits of HPC without the deployment challenges. It features the world’s most-proven, market-leading HPC systems based on HPE Apollo hardware, plus software, storage, networking and services – all specifically configured for running modelling and simulation workloads. Choose from small, medium or large configurations, receive them quickly and gain a fully managed service from HPE.


Streamline HPC administration with HPC

HPE GreenLake Central offers administrators exceptional control and visibility: configuring and monitoring scheduling queues through a Slurm-based scheduler, managing user access and quotas, accessing historical job monitoring data, managing licences and much more – all through a single interface.

Benefit from purpose-built HPC systems

HPE Apollo systems are designed to support HPC workloads without the need for a supercomputer. These high-performance systems are optimised for HPC workloads including manufacturing, automotive, biosciences and more.

Offload monitoring and management tasks with HPC

Enjoy a fully managed and operated HPC platform, with support from HPE experts in HPC, AI and app services. Free your HPC teams from the burden of infrastructure administration so that they can focus on the business challenges they need to solve.

Stay competitive with the latest technology

HPE GreenLake edge-to-cloud platform engagements include a technology refresh feature, allowing you to benefit from the latest technology in the market as it becomes available. HPE can even help you retire and recover value from your existing aging infrastructure.

TAP IMAGE TO ZOOM IN

HPE GreenLake for HPC

HPE GreenLake for HPC runs on premises – at your edge, in a colocation or in your data centre – so you benefit from the security and control that an on-premise infrastructure provides. Pick from small, medium or large configurations built on an enterprise-grade hardware/software stack optimised for HPC and purpose-built on innovative cloud modules, delivering GPU support for the HPC cloud service. Featuring consumption-based billing, this service provides:

  • Transparent, pay-as-you-go pricing model that delivers an on-premises service as an operational expense
  • Elasticity to scale up or down based on workload demands
  • Reserved capacity and a usage-based consumption model provides cost predictability while supporting variable demand as data volumes expand.

See how HPE GreenLake for HPC works with a free trial

The HPE GreenLake edge-to-cloud platform is delivering results for enterprises around the world, and we want you to experience it for yourself. When you request a free trial of HPE GreenLake for HPC, we’ll install the instance in a co-located HPE data centre and set up your account on the HPE GreenLake Central platform – providing management capabilities, usage insights and detailed consumption reporting.

  • A 10-day trial is standard and can be extended, if needed.
  • A 21-day trial can also be requested for complex use cases and can also be extended, if needed.
  • A trial engineer is assigned and available throughout the trial period to help you navigate through the use case scenarios and to answer your questions.
  • Service is delivered through a co-located HPE data centre, so no equipment is needed on your premises.
  • Your trial includes a dedicated HPC cluster with up to 1,280 cores so that you can provision your apps and load data to see how it would work in your environment and get a taste of the HPE GreenLake platform experience.
  • There’s no cost to you.
TAP IMAGE TO ZOOM IN

Get the details

Small configuration

Medium configuration

Large configuration

Who is this recommended for?

  • Number of racks: 1
  • Total physical cores: 1,280
  • Compute: 20 HPE Apollo 2000 XL225n (2 CPUs, 64 cores, 512 GB)
  • Scratch storage: Local NVME
  • Shared storage: 8 HPE DL325 storage nodes with 232 TB raw capacity
  • Network: High-performance 100 GbE Aruba
  • GPU (Optional): NVIDIA A100 GPU with HPE Apollo 6500 servers
  • Number of racks: 3
  • Total physical cores: 4,352
  • Compute: 68 HPE Apollo 2000 XL225n (2 CPUs, 64 cores, 512 GB)
  • Scratch storage: Local NVME
  • Shared storage: 16 HPE DL325 storage nodes with 464 TB raw capacity
  • Network: High-performance 100 GbE Aruba
  • GPU (Optional): NVIDIA A100 GPU with HPE Apollo 6500 servers
  • Number of racks: 6
  • Total physical cores: 8,960
  • Compute: 140 HPE Apollo 2000 XL225n (2 CPUs, 64 cores, 512 GB)
  • Scratch storage: Local NVME
  • Shared storage: 28 HPE DL325 storage nodes with 812 TB raw capacity
  • Network: High-performance 100 GbE Aruba
  • GPU (Optional): NVIDIA A100 GPU with HPE Apollo 6500 servers

Software stack

  • Singularity Open Source (container runtime)
  • Singularity PRO (optional)
  • Slurm
  • SLES 15 SP2 (HPC-ready golden image)
  • HPE Ezmeral Data Fabric (shared NFS)
  • HPE Performance Cluster Manager
  • Singularity Open Source (container runtime)
  • Singularity PRO (optional)
  • Slurm
  • SLES 15 SP2 (HPC-ready golden image)
  • HPE Ezmeral Data Fabric (shared NFS)
  • HPE Performance Cluster Manager
  • Singularity Open Source (container runtime)
  • Singularity PRO (optional)
  • Slurm
  • SLES 15 SP2 (HPC-ready golden image)
  • HPE Ezmeral Data Fabric (shared NFS)
  • HPE Performance Cluster Manager

Deployment options

  • Customer data centre
  • Colocation facility
  • Customer data centre
  • Colocation facility
  • Customer data centre
  • Colocation facility

Control plane

HPE GreenLake Central orchestration with HPC-specific module to monitor and manage the HPC environment and jobs.

HPE GreenLake Central orchestration with HPC-specific module to monitor and manage the HPC environment and jobs.

HPE GreenLake Central orchestration with HPC-specific module to monitor and manage the HPC environment and jobs.

What is metered

Usage is metered based on the compute (per hour) and storage (per GB) used by the nodes in a cluster.

There are two (2) meters used to calculate usage over the reserved capacity:

  • Compute nodes on/off
  • Raw capacity used on shared storage

Usage is metered based on the compute (per hour) and storage (per GB) used by the nodes in a cluster.

There are two (2) meters used to calculate usage over the reserved capacity:

  • Compute nodes on/off
  • Raw capacity used on shared storage

Usage is metered based on the compute (per hour) and storage (per GB) used by the nodes in a cluster.

There are two (2) meters used to calculate usage over the reserved capacity:

  • Compute nodes on/off
  • Raw capacity used on shared storage

Included services

HPE GreenLake Management Services

  • IT service management
  • System operations and administration
  • Support services
  • HPC-specific support services

HPE GreenLake Management Services

  • IT service management
  • System operations and administration
  • Support services
  • HPC-specific support services

HPE GreenLake Management Services

  • IT service management
  • System operations and administration
  • Support services
  • HPC-specific support services

Optional add-on services

Advisory & Professional Services

  • Container adoption services (Container Factory)
  • Virtual desktop infrastructure (VDI)
  • Customer portal integration
  • Storage and data management

Advisory & Professional Services

  • Container adoption services (Container Factory)
  • Virtual desktop infrastructure (VDI)
  • Customer portal integration
  • Storage and data management

Advisory & Professional Services

  • Container adoption services (Container Factory)
  • Virtual desktop infrastructure (VDI)
  • Customer portal integration
  • Storage and data management