Managed HPC Services are a suite of professional upgrade and maintenance services we provide to our clients. Our highly trained, certified engineering team will help you manage, update and service your systems whenever and wherever you need us.
HPC Managed Services by Aspen Systems are the suite of professional upgrade and maintenance services Aspen Systems provides to our customers. These HPC services allow our customers to leverage Aspen Systems’ trained engineer pool and expertise to help manage their systems in many different circumstances.
If you can allow remote access to your system, we can provide additional software, configuration, and model tuning help, stand in for your system administrator or even fully manage your system for you, based on the options you select for your contract or arrange for on a case by case basis.
Our HPC Managed Services are implemented via an encrypted secure network connection to your cluster. You may choose one of several different connection methodologies, depending on your organizational security requirements.
We can configure your cluster with a VPN client that “calls home” to a dedicated Aspen Systems secure support network. Clients connected to our secure support network cannot view or access any other client systems. Only our engineers can access the secure support network in order to communicate with your system for routine monitoring, maintenance and upgrades. The Aspen Systems VPN requires no inbound ports to be opened on your organizational firewalls, and more than one system in your cluster can be configured with an Aspen Systems VPN client for redundancy. Most clients prefer this approach as little or no network configuration is needed at your site.
We can configure your cluster with several monitoring tools to help you and your support team get the most value out of your technology investment. Nearly all aspects of your cluster can be monitored, including performance/utilization, network saturation, power consumption and temperature monitoring can be added to your cluster a la carte. All of the monitoring tools can be made available to both your team and our support team in order to help you troubleshoot or optimize system performance.
Secure Shell, or SSH, is an encryption and network protocol that is widely used to provide secure communications between networked hosts. Aspen Systems can perform most remote management with SSH access. This configuration requires that port 22 be allowed and forwarded to the outside interface of your clusters.
In some cases, your security rules may require that Aspen Systems utilize your VPN solution to access your cluster. Aspen Systems supports most major VPN clients, but will require that you provide such things as cryptographic devices and client licenses.
Aspen Systems HPC Managed Services are offered in several packages. We can also customize a plan to meet your specific requirements. Features can be added to Standard and Flex packages by request.
| Features | Standard Package | Flex Package | Total Package |
|---|---|---|---|
| Length of Coverage | 1-5 Years | Annual | Annual |
| Distribution Updates | Minor Updates Included | Included | Included |
| Utilities Updates | Minor Updates included | Included | Included |
| Applications Updates | Minor Updates Included | Included | Included |
| Hardware Support | Advanced Cross Ship or Optional Onsite Warranty | Advanced Cross Ship or Optional Onsite Warranty | Advanced Cross Ship or Optional Onsite Warranty |
| Free Two-Way Shipping | Standard/Expedited | Standard/Expedited | Expedited |
| Performance Tuning | Included | Included | Included |
| Hardware Troubleshooting | Included | Included | Included |
| Case Tracking | Included | Included | Included |
| System Administration Assistance | X | X | Included |
| Disaster Recovery | X | X | Included |
| Onsite Visit/Check-up | X | Annually Upon Request | Semi-Annually Upon Request |
| See More Details | Learn More | Learn More | Learn More |
*Included with all HPC Clusters
Learn MoreIn addition to our Standard Service, Flex Service, and Total Service offerings, we offer many different customized services that can be mixed and matched or used as a stand-alone solution to meet your specific requirements and situation.
Hardware component failure is inevitable in today’s high performance computing environment and we’ve got you covered. Our highly trained professional network of technicians are available to be dispatched on a 24×7 or 8×5 on-site as little in a 4 hour response time locally or worldwide.
We cover the shipping costs of your system parts both ways, for the life of your warranty. For time-sensitive maintenance tasks we’ll select priority shipping to ensure the rapid repair of your systems, regardless if it’s for entire nodes or just peripherals and components.
When planning for the lowest possible downtime in repairs and maintenance, on-site spares are critical. We offer you additional support by providing advance replacements to complement your on-site spares from our distribution warehouses with priority shipping.
Catastrophic system events can happen to any organization, regardless of any preventive measures taken in the planning stages. We understand the true cost of down time from disasters and the immediate need to recover to full operations and offer complete support in helping you recover quickly from these extreme situations.
Many organizations have key system administrative and support staff that have a well-deserved vacation or time-off planned. Our on-demand expert engineering team can handle your complex workload on behalf of your personnel during their absence for a seamless shift in your workload balance.
When resources are stretched beyond normal capacity and immediate assistance is required, we can join the fight. By providing instant support relief with remote administration and on-site engineers, we can ensure that the quickest resources can be deployed for the best possible resolution.
The most common plug/receptacle types you will encounter in North America are:
CoolingTo figure out which size unit is best for your cooling needs: