Custom Rack Scale Solutions

Hyperscale Your Data Center the Aspen Systems Way

Capacity demand for data center resources is increasing by 40% year over year, and Aspen Systems Inc. is your strategic partner for meeting that demand. Our team of veteran engineers and production experts is ready to design, build, and deploy your very own, fully custom rack scale solution from the ground up, and at any scale. Together, we can help you break out of the traditional IT data center infrastructure mold and achieve hyper scalability.

Get the racks you need and the support you always wanted from the experts at Aspen Systems Inc.

Why Aspen Systems?

Aspen Systems can help you get a competitive edge over your competition by creating the right Rack Scale solution based on Intel Rack Scale Design (RSD) without breaking the bank.

The current advances in this world are demanding massive compute power and data storage in the data center. We will help you design the right base system which is easily managed, and expanded over time. With the Aspen Systems Rack Scale solution, you can upgrade by adding only the right components you need, when you need it. For management, we offer software which can be used to manage your whole data center from one screen.

What is Rack Scale Design?

Intel® Rack Scale Design (Intel® RSD) is a reference architecture for building disaggregated, interoperable, and composable IT infrastructure that scales quick and lean while reducing ongoing capex and opex. Aspen Systems Inc. leverages this framework to help our customers hyperscale their data centers to meet exploding demand.

Disaggregated

Buy only what you need when you need it

With disaggregated architecture, various data center compute, storage, and accelerator hardware resources are installed individually within a rack and packaged as blades, sleds, chassis, drawers. This makes for an unprecedented degree of modularity and serviceability, allowing you to invest in individual resources as the need arises.

Interoperable

Choose the best hardware without vendor lock-in

By utilizing a unique single pane-of-glass harware management that utilizes open standard APIs, rack scale architecture allows you to maintain vendor independence. Utilize RSD-compliant racks procured from different vendors without any effect on overall system performance.

Composable

Compose on the fly for specific workloads

By disaggregating resources into individual resource pools and then connecting them via a high-speed interconnect fabric, Rack Scale design reference code and API’s then allow us to compose operating environments in seconds for a wide range of workloads.

Rack Scale Architecture Overview

Hardware Resources

Being a disaggregated architecture, RSD means that data center hardware resources (compute, storage, acceleration, and networking) can be installed as individual modules within a rack. These can be packaged as blades, sleds, chassis, drawers or larger physical configurations as needed.

Pooled System Management Engine (PSME)

At the lowest level, each individual resource module will have a software controller component called the Pooled System Resource Manager (PSME), which can be installed at the blade, module, or rack level. The PSME reveals what resources are available in the module and exposes APIs to higher levels via the Ethernet management network.

Rack Manager Module (RMM)

The Rack Manager Module (RMM) communicates with the PSMEs and takes responsibility for a variety of rack-oriented functions, including distributing security certificates to PSMEs, managing rack power and cooling, reporting environmental and health status, and designating and reporting the physical location of each asset with the rack.

Pod Manager (PODM)

The Pod Manager (PODM) software manages functions within and across racks for a single pod. The southbound PODM APIs interact with the PSMEs and the RMMs. The northbound PODM APIs are exposed to the user’s choice of orchestration software such as OpenStack*, Vmware, or other commercial, open source or custom DCIM software.

Data Network and Management Network

Resource modules within the RSD architecture will have at least two networking connections, the high-bandwidth data connection, and a separate out-of-band Ethernet link to a dedicated management network. These two networks typically connect to top-of-rack switches that can be connected to other racks, making up a management domain called a “pod.”

Benefits of Rack Scale Design

Increase Agility

RSD allows data center managers to quickly react to a rapidly changing data landscape by selectively purchasing a mix of hardware modules to match their current needs. Additionally, the improved standardization of RSD makes rack installs and refreshes faster and easier.

Reduce CAPEX

By allowing data center managers to selectively invest in resource, this increases resource utilization and reduces the cost per unit of capacity. Additionally, since resource are pooled and better utilized, fewer resources (such as NVMe drives) are needed.

Reduce OPEX

RSD has been shown to reduce technician time by up to 77% due to far fewer handoffs, increased standardization, and simplifying the steps required to refresh and scale-out racks. With fully disaggregated servers, you can refresh just the CPU and DRAM modules, saving 44 percent over a full-acquisition refresh.

Why Now?

Data is Growing

We live in a world that is more connected than ever, millions of new devices come online each year. The Internet of Things, big data analytics, and machine learning are driving an annual increase of 30 to 40 percent in compute power and data storage capacity – and tradition data center infrastructure absolutely cannot keep up.

RSD not only provides a roadmap to building the IT infrastructure needed to serve the growing demand of the information age, but provides solutions to reduce cost and improve serviceability, ensuring that we continue to meet demand sustainably.

Future-Proof The Data Center

Housing thousands of server racks with their own processors, memory, storage, and other resources is wasteful and expensive. When one server breaks or needs upgrading, it’s standard to replace the entire server. With shortages and long delivery times continuing to ravage our industry, this is not only costly, but irresponsible.

RSD not only prevents needless waste by allowing you to flexibly upgrade resources without needing to throw out entire servers, but it also keeps your up-front capex low – buy only what you need, when you need it.

Get a Quote for a Custom Rack Scale Solution

System Architecture

Power and Cooling

Storage

Networking

What network technology would you prefer?

Accelerators

TPM

Do you need a TPM?

Budget

Submit Inquiry

Complete the brief form below to submit your inquiry.












    *Aspen Systems Inc. will never share or sell your personal information for any reason. Read our policy on data privacy for more information.