Telecoms Forensics Equipment Ltd

                         Rid your space of all internet vermin, get a Subpico CAT! 



Are you tired of waiting weeks or even months for scheduled use of a supercomputer?
 Now  you  can  quickly  log  in  and  run  your  programme,  and  access  your  results  easily.



Supercomputers are required for demanding applications and data sets in all areas of industry and research.

High Performance Computing (HPC)

Today High Performance Computing (HPC) platforms are supercomputing clusters. HPC systems are called 'loosely coupled' because basically CPUs on different nodes don't share a common hardware bus and have to communicate through a slow protocol stack and cables. Typically they don't share memory very well, if at all. By contrast a tightly coupled SMP system puts components in the same bus and space, making things simpler and faster to program and run.

HPC clusters are economical, and are popular with online hosting companies and other service providers because of its low cost. However to achieve the performance that HPC clusters promise, applications must be tailored for the architecture, by a resident MPI programmer, IT consultant or engineer, which is where the cost escalates.

Libraries of cluster-ready HPC applications for specific communities have and are being developed. But there are many important end-user applications left unaddressed. Entire fields of scientific endeavour that could benefit from high performance computing have not, due primarily to the programming complexity of HPC clusters. Emerging areas of science, engineering and industry lose ground and time-to-market because of the overheads in programming and scheduling.

Most industry experts: scientists, researchers, engineers and analysts like to focus on their areas of specialty, get their computations executed quickly, and avoid becoming entangled in the computer science complexities that supercomputing clusters demand.

Symmetric Multi-Processing (SMP)

All Applications from gaming, web, to science and engineering are typically developed on laptops, desktops, workstations etc, and may even use servers. These computers are Symmetric Multi-Processing (SMP) systems and align more closely with your normal user computer skills enabling the user to focus on their areas of specialization and interest. 

Programs can be written compiled and run easily on SMP systems. Anyone wanting a boost of compute power would rather not re-write their applications for a supercomputing cluster and would prefer to use SMP supercomputers. 

However SMP supercomputers are expensive to buy and run. They have been out of reach economically for many due to their reliance on costly proprietary hardware and services, and operational maintenance. 


Software-based SMP systems are becoming more available and economical. They can be made from the same lower cost, non-proprietary, off-the-shelf servers that have always made non-SMP HPC supercomputer clusters attractive. The new software lets the HPC cluster mimic an SMP system. Unfortunately it relies on hypervisor implementations -- a middleware layer of software that executes above the operating system.

This hypervisor layer of software adversely affects performance thru-put and compute capability, limiting its High Performance Computing applicability.


TFE Supercomputing

TFE Subpico Supercomputer AMX-CNC.1.002 a new generation of SMP Supercomputer.

The next generation is a hybrid implementation at the operating system level, firmware and in hardware

 inherently faster than a hypervisor and completely transparent to the user applications.

In the Subpico AMX-CNC.1.002 we implement our ground breaking power saving and security technologies making our system 100% secure and environmentally responsible. Power, security and reliability are our core elements.

Also TFE Supercomputing exceeds the performance of HPC supercomputing clusters with the simplicity and ease of use of a desktop or workstation. Our technology brings the benefits of supercomputing i.e. (compute power) to unreached areas of academia, research, development and analysis but making it affordable removing the queues and complexity.

  Affordable Supercomputing
  Apply compute power to more projects and enhance your output.
  Multiple Simultaneous Processing
  Faster project completion. No more delays waiting for scheduled HPC time.
  Large Single Shared Memory
  Ideal for large memory applications.
  Multilevel Security
  Data Integrity, reliability and protection. 
  Optimized Transport 
  Efficient inter/intra process communications leading to quicker results.
  Power Efficient
  Saves money and runs cooler, speeding up compute time and making a greener footprint.
  Single Software Image
  Simple and scalable multi-threaded distributed programming. No complicated cluster tailoring.