Bespoke servers for every workload
Scan provides a comprehensive range of 3XS server solutions, designed from the ground up and optimised for various workloads including GPU-accelerated HPC and deep learning. Using our intuitive server configurators you are able to customise every aspect of your server build enabling you to get exactly the right product for your particular application or challenge.
The 3XS Server Philosophy
Specialised servers for bespoke workloads are our bread and butter, rather than the niche of many mainstream server brands. A vast choice of enterprise-grade components allow you to configure the ideal solution.
We provide a one-stop-shop for all your server needs - offering seamless support during build, after-sales and with our professional services offering.
We are proud to say a large proportion of our server customers come back to us for upgrades and new builds.
Here we will highlight some of the latest and innovative server technologies that can be incorporated into your 3XS server, but to learn more about all our configurable components please read our comprehensive server buyers guide.
AMD EPYC CPUs
The AMD EPYC 7003 Series of processors set the new standard for the modern datacentre. Featuring leadership architecture, performance, and security, the third generation AMD EPYC processors help customers turbocharge their application performance, transform their datacentre operations, and help secure their critical data. Based on AMD Zen 3 Architecture, they feature 7nm hybrid-multi-die design and offering up to 64-cores, PCIe Gen4 and 32MB of L3 cache per core. The AMD EPYC 7002 Series continues to offer many of these features, but using the Zen 2 architecture to deliver more cost effective solutions.FIND OUT MORE
Intel Xeon Scalable CPUs
The latest 3rd Gen Intel Xeon Scalable Platform processor architecture is designed to optimise interconnectivity with a focus on speed, without compromising data security. Features such as up to 40-cores, 64 lanes of PCIe 4.0 and Intel Advanced Vector Extension 512 (Intel AVX-512) improve with workload-optimised performance and throughput increases for advanced analytics, HPC applications, and data compression, whilst Deep Learning Boost is designed to improve AI workloads such as image recognition and object detection.FIND OUT MORE
Depending on the type of workload the server is intended for, a large amount of system memory may have less or more impact, especially if used in combination with GPU accelerators. NVIDIA recommends at least double the amount of system RAM as GPU RAM, so high-end systems may scale into the TBs. Additionally Intel Xeon based servers can make use of a combination of traditional DIMMs and Intel Persistent Optane Memory DIMMs, allowing a flexible solution addressing performance, fast caching and extra storage capacity.
GPU Accelerators for Servers
Acceleration by NVIDIA
Many of our customisable server builds are based on chassis that can take anywhere between two and ten acceleration PCIe add-in cards such as GPUs. This allows parallel computing to be undertaken a significant scale using any of a wide range of NVIDIA GPU cards - from the cutting-edge Ampere-based A100 and A40 models optimised for high-end compute tasks, to the RTX A6000 and other cards designed for visualisation workloads. You can learn more about individual GPUs and their suitabilities for various workloads by reading our professional GPU buyers guide.FIND OUT MORE
Virtual GPU Solutions
Alternatively, a bespoke server can be configured to deliver NVIDIA virtualisation solutions for graphics visualisation or compute. This range of virtual GPU (vGPU) solutions brings the power of NVIDIA GPUs to virtual desktops, apps, and workstations accelerating graphics and compute to make virtualised workspaces accessible to creative and technical professionals working from home offices, remote sites or anywhere. There is also a virtual compute server option for data science applications. NVIDIA vGPU solutions deliver the ultimate user experience with the ability to support both compute and graphics workloads in hypervisor-based virtualisation environments.FIND OUT MORE