HPC

Our experience

HPC Specialists

Do IT Systems boasts extensive experience in HPC projects in many fields and covers a range that goes from small-size systems to Top 500 supercomputers. In addition to use advanced technologies, we constantly monitor the current market needs and developments and closely collaborate with the sector’s technological leaders with the aim to test the innovations in an initial phase in order to promptly offer them in our products and solutions.

We offer the best solutions related to the type of use on CPU and GPU architectures, advanced server technologies and high-performance networks in order to achieve utmost scalability and performances for our customers’ applications.
Our engineers are always available to help minimizing the complexity of the hardware and software, optimizing the entire system according to the customer’s needs.

Thanks to our partnerships with market leaders, we are able to supply on-premise and cloud solutions. Our solutions will improve problem solving and will help you to calculate to compete.

The potentials are endless. The limits are set by the design, the sizes and quality of the supercomputer: and these limits can be overcome with the right product and skills.

PARALLEL CALCULATION

Parallel calculation is the simultaneous use of more calculation resources to solve a calculation problem. In the parallel calculation, a problem is divided in a certain number of sections, and each section is further divided into a series of instructions; the instructions of each section are performed simultaneously on various processors and a general coordination and control mechanism is used.

HIGH PERFORMANCE

The new generation of graphic cards (GPU, GPGPU and co-processors) further boost the processing abilities of the HPC system. For supported applications, a GPGPU card can remarkably speed up calculations.

SCALABILITY

The parallel nature of HPC systems allows them to be expanded extremely easily and effortlessly. A well-designed HPC cluster can be re-sized according to calculation needs: hence users have the possibility to start with a small size and build their system as required.

STABILITY

For their nature, the clusters allow to always have an active system: in case of fault of one of the nodes of the HPC system, the activity can be continued without compromising the cluster. The use of Linux combined with a reliable hardware, redundant discs and power supplies allows stability and minimizes downtimes.

ADDITIONAL INFORMATION

HPC, WHO IS IT FOR?

HPC (High Performance Computing) allows processing large quantities of data more quickly by using the aggregated calculation power of more servers.
This way, it is possible to process high-intensity data calculations and calculations that are too large for standard computers or that would require long processing times. Specialized HPC systems, when implemented on large scale, are sometimes defined “supercomputer”.
Desktop computers boast processors, memory, disc and operating system as supercomputers. Nonetheless, the latter consist of thousands of processors and act as a single machine.
HPC systems use high-speed network inter-connections to perform parallel calculations, meaning dividing a calculation into smaller problems.
The most common users of HPC systems are scientific researchers, engineers, bioinformaticians and data scientists. The manufacturing industry, government agencies, schools, weather centres rely on HPC for complex applications.
Recently, HPC was used by researchers in social media, semantics, geology, archaeology, materials, urban planning, graphic design, genomics, brain imaging, economics, game design and even music. The list will continue to expand because more and more people are being introduced to the possibility to use HPC systems, defining how it can be employed in their fields.

At the moment, the sectors that implement HPC are mainly the following:

Researchers
Schools
Engineers
Government agencies