Types of clusters

High Performance Computing Clusters

HPC (High Performance Computing) clusters are our main focus at Advanced Clustering Technologies. HPC clusters currently make up more than 80% of the 500 fastest computers in the world. HPC clusters pass on a massive cost savings benefit over SMP and MPP-based computers by leveraging the hardware made for consumer and general business usage.

High performance computing clusters increase computer performance exponentially by sharing the workload, spreading any given task among different cluster nodes.

The two main types of jobs that are run on HPC clusters are serial and parallel. Parallel jobs break the problem up into parts and use a communication layer — such as MPI or PVM — to spread the workload among many machines and processors. All of the machines that are part of the run will communicate and work together. A serial job runs on one machine with no communication to other nodes.

The type of job you are running will greatly influence the type of hardware and software setup of your cluster. Advanced Clustering Technologies has an expert sales team that closely works with our engineers to help you make the right choices in your cluster planning and purchase. We also have a test cluster for you to try! Test some options on our demo cluster and have confidence that your new cluster fits your needs.

Visualization Clusters

A visualization cluster is an HPC cluster with the addition of powerful graphics cards, normally designed to work in sync with each other to tackle high-resolution and real-time simulations. Our visualization clusters add high-powered 3D-accelerated graphics cards to the compute nodes.

altAt Advanced Clustering, we have designed and delivered visualization clusters for universities, military, government and corporate clients.

Space requirements are often a concern when it comes to visualization clusters, and heat generation and power demands are also considerations with video cards, which require more cooling than some processors. Advanced Clustering Technologies has developed an entire line of systems with rack-density and efficient cooling in mind. Check out our GPU servers section for more information about these products.

Visualization clusters now include an ever-growing group of sub-categories. A visualization cluster is often used for:

  • Weather forecasting
  • DNA sequence analysis
  • Protein folding simulations
  • WMD simulations
  • Multi-user immersive virtual reality
  • Climate and environmental modeling
  • Molecular nanotechnology
  • 3D modeling

Grid Computing

A grid allows an organization to pool multiple computing resources: HPC clusters, visualization clusters, workstation and instruments. The resources can be pulled as needed and access can be granted to a wide range of users.

Grid computing is an exciting use of computing resources. A grid ties many types of resources together in differing locations. These resources can be entire clusters, unused workstations or even large specialized instruments, such as telescopes or particle accelerators. A grid is not a cluster but it can — and often does — contain clusters. A cluster is generally a uniform type of computer located in the same physical space.

To maximize the departmental resources of an organization, clusters may be interconnected to form a grid, which creates more computing power for each department and puts unused resources to work, thus increasing the ROI of each system.

 

Customer testimonial

"By the way, I probably have said this before, but Billy is awesome. RMA's can be such a pain with some companies, but he makes it painless." -- Damon G. at NASA Goddard Space Flight Center.

Find out even more reasons to purchase from us

 
Contact Info
Toll-free: 866-802-8222 International: 913-643-0300 Email: info, sales, support