GPU Grid Computing

Go Beyond Cluster Computing!

Our NovaServ GPU systems are exceptional for not only cluster computing, but also grid computing. Grid computing is very similar to cluster computing, but typically larger in scale, involving the clustering of clusters. GPU solutions, which allow for a degree of scalability not present in CPU-only systems, offer significant performance enhancement in any cluster or grid environment.

With preconfigured turnkey development and cluster management tools, ICC's NovaServ solutions are exceptionally deployable solutions in a variety of environments. They are the perfect choice to meet your supercomputing needs.

If you think ICC's NovaServ systems can meet your cluster and grid computing needs, feel free to reach out to an ICC GPU System Sales Engineer and learn how we can take your supercomputing to the next level.

About Grid Computing


GPU Grid Computing 1

Grids vs Clusters

Grid computing and cluster computing are often confused, and for good reason as they are very similar in nature. Both involve the combination (in some form) of different resources into a "single" whole that works towards a particular purpose (if you are unfamiliar with cluster computing read our overview to HPC first.) Some say that clusters are strictly homogenous computing, but that is not quite accurate since clusters can run multiple operating systems across different nodes.

The biggest difference with grid computing is simply the scope and sacle. Grid represents a bigger framework and architecture, and focuses on the broader scope or objective. It incorporates many of the key characteristics of clusters, and clusters often become one of the many components of grids. By enabling the sharing, selection, and aggregation of a wide variety of geographically distributed resources including supercomputers, storage systems, data sources, and specialized devices, grid computing allows large organizations and separate organizations to collaborate internally and amongst themselves.

Put in one sentence, clusters can fit into grid architecture and grid computing for the ultimate sharing of resources at a higher level of aggregation.


GPU Grid Computing 2

GPU Grid Computing in the Real World

GPU grid computing is utilized in a variety of fields, but it remains prominent in the fields which GPU supercomputing is perhaps best suited for: engineering and scientific computing, as well as computationally-intensive projects. One of the biggest open movements is GPUGRID, a volunteer and distributed computing initiative for biomedicine. GPUGRID's infrastructure is comprised of many NVIDIA graphics cards joined together to deliver high-performance all-atom biomolecular simulations. The molecular simulations performed by the volunteers are some of the most common types performed by scientists in the field, but they are also some of the most computationally demanding, typically requiring a supercomputer.

GPU will also play a major role in FutureGrid a project led by Indiana University and funded by the National Science Foundation (NSF) to develop a high-performance grid test bed that will allow scientists to collaboratively develop and test innovative approaches to parallel, grid, and cloud computing. The ultimate goal is to build a system that helps researchers identify cyberinfrastructure that best suits their scientific needs.