メインコンテンツに移動

Why Supermicro for On-Prem Data Centers

Data centers have become the primary repository for corporate data. While the location of data and the associated storage and computing capabilities may vary, many organizations, for various reasons, are implementing a cloud-based data center within their “four walls.” Supermicro servers and storage systems are ideal for creating an on-premise cloud.

First to Deploy

With Supermicro’s tight relationship with CPU and GPU suppliers and a design methodology using Server Building Block Solutions ®, new technology can be installed quickly in on-prem data centers once a product is released. Many workloads depend on the latest servers that incorporate Intel or AMD CPUs and/or Intel, AMD, or NVIDIA GPUs to perform as expected. Many organizations that require the performance or new features that are part of a new generation of microprocessors do not have the patience to wait for a public cloud entity to make these servers available. Acquiring new servers for an on-prem installation may result in the earliest access and allow organizations to meet their goals earlier.

Supermicro is known for being a first-to-market IT systems provider. Over the past few years, the timeliness and variety of products available when new technologies become available from Supermicro have been unmatched. As a result, organizations get the needed configurations early in the technology cycle.

Application and Workload Optimized

The variety of form factors, CPU speeds, memory, GPUs, and networking options can be enormous. Modern servers are created to accommodate a number of different components that are designed to work together. The exact combination of CPU performance, memory capacity and density, GPU availability and type, direct storage capacity and performance, and networking speeds affect modern applications. Specific configurations consisting of the list of options that a workload requires may only be available in an on-prem data center.

Supermicro’s range of servers and storage systems spans many workloads. Whether tuned for enterprise applications with significant amounts of memory, HPC with high-frequency CPUs, or systems designed to support many GPUs, the Supermicro portfolio is designed for many different workloads. The energy-saving design reduces power usage and thus OPEX, reducing the TCO. In addition, a number of different servers can be cooled with liquid cooling, whether direct to chip or with immersion tanks.

On-Prem data centers allow IT administrators to reduce the TCO and increase infrastructure performance. Customized configurations can be installed and optimized for specific workloads. Full IT Solutions can be staged and completely tested before installation.

Service Level Agreements

When an SLA is not met for uptime, performance, or capacity, diagnostics can be run on the on-prem infrastructure to determine the root cause and take corrective action. With the Supermicro SuperCloud Composer, the entire lifecycle of servers, storage, networking, and software can be managed from a central console. The on-prem location of the components of a modern IT infrastructure can be monitored and fixed as needed without having to search and contact a distant data center.

Entire Solutions

Whether a single node running a specific application or a cluster of hundreds to thousands of systems, Supermicro has the experience and knowledge to deliver complete IT solutions. With L12 assembly and testing capacities for some of the largest clusters ever built, the Supermicro manufacturing facilities can cover whatever an IT department requires. Supermicro can offer various servers, storage, and networking hardware racks, install software and test the entire cluster in conjunction with the customer to ensure performance and power targets. Supermicro has the manufacturing space and infrastructure to design and certify complex IT requirements for data center environments that require liquid cooling.