High Performance Computing

High Performance Computing (HPC) has evolved from High Performance Technical Computing (HPTC) with the influx of industry standard servers, Linux and Microsoft.

HPTC, which was focused on research, scientific and highly specialized parallel engineering and analytical tasks, has now been broadened to encompass parallel tasks in finance and insurance ( risk analysis, funds modeling, econometrics), digitial content creation (rendering, CGI, animation sequencing), as well as a range of computing models starting with distributed computing to grid to cloud computing.

HRG’s observation of this marketplace is also changing. Once a bastion of R&D labs and governmental research centers, large corporate computing research centers,  and roll your own problem solving programs, the R&D roles and traditional business datacenter roles are blurring. The introduction of dual and quad core processors and SMP architectures with 2 and 4 processor complexes, the traditional parallel cluster of machines for small HPTC can now be housed in one machine with superior graphics based on either Nvdia or attached FPGA co processors or GPU’s. This emergence of low cost high performance hardware coupled with Linux and or Microsoft and a huge influx of Independent Software Vendor applications ported to Linux and Microsoft has turned this market segment into a major IT battleground.  

Over the years, R&D managers have built in processes and  features  to provide the highest levels of performance to maximize every FLOP (FloatingPointOperation) provided. Simplistically, these rules have now changed. New services have been developed to handle better management of distributed, parallel computing and/or GRID environments which allow sharing of data and application processes across one or more geographic sites. Storage has also become a major management concern in that the explosion of data is overwhelming these computing center sites and management of this iterative and potentially sensitive data is a corporate asset that requires high availability and data security features throughout the computing center. In addition, networking and application costs should also be a consideration.

Layered on top of the increasingly complex processes, base software needed to actually meet computational needs are the expectations of the business leaders who look at these operations as a utility - flip the switch and it’s on.

HRG finds that responses to these issues have grown into a patch-work quilt rather than a well designed and coherent mosaic.  Data center staff doesn’t, and shouldn’t, have the range of expertise and experience needed to redesign and re-engineer the computing center to ensure the needed computational availability.

HRG has a history of success in delivering results, offering unparalleled advice and assistance in developing long-range strategies, as well as hands-on implementation of short-term tactical solutions.  HRG’s team of experts has been in the industry for over 30 years across multiple companies either developing or operating with HPTC and HPC products and services.

HRG offers services designed to meet the needs for extending your current applications and data.  These include tailored best practices for application and virtualization management, planning services for computational levels to meet your current and future needs as well as auditing services for your computing center operations.

HRG has an on-site service suite covering not only the physical plant such as power and cooling applications and data, but HRG can also analyze your system architecture and make recommendations to insure that your performance levels and data management needs are secure and always available.

HRG’s philosophy is simple: HRG values integrity, confidentiality and results.  HRG’s clients are successful organizations requiring specific, accurate assessment of technology trends, and advice to differentiate them from the competition and achieve their financial goals.