Bare Machine Computing

Bare Machine Computing (BMC):
Over a long period of time, operating systems (OS) or kernels (micro, mini, nano, tiny, etc..) provided higher layers of hardware abstractions to offer a convenient programming platform for the application developers. During this time period, hardware became smaller and smarter and the software became huge and complex. The complexity in OS coupled with proliferation of platforms and comptuer applications created a myriad of products for the same application domain resulting in tremendous waste and rapid obsolescence. It would be nice to have a single email application that can run on a variety of computing devices such as a Mainframe, Desktop, Laptop, PDA, Cell phone, or any other network device! We would be spending time and energy in enhancing and improving such applications instead of writing, re-writing, porting, and re-porting the same application many times over. This simple example illustrates the need for a computing paradigm that is independent of a computing platform or environments. We propose a Bare Machine Computing (BMC) paradigm that accomplishes the above objectives. When OS or any centralized controller is avoided, then a computer application directly communicates to hardware that is bare and does not have any valuable resources. Such computer application is referred to as an application object (AO). The AO runs on a bare machine which consists of a CPU, Memory, and I/O devices. When OS layer is tumbled then the computing platform becomes flat and it has only two entities: hardware and software (AO). When hardware architecture (e.g. X86) is made standard, then any AO can communicate to this hardware thus making the AO stable over a long period of time. This will reduce the speed of obsolescence and impede the rapid changes in computer applications. It will also polarize the computing application-centric instead of being susceptable to ever changing computing environments. Our research focuses on explorations of the BMC paradigm by studying complex computer applications that run on a bare machine. We have made tremendous progress in developing reasonably complex systems such as Web server, Email server, Webmail server, VoIP soft-phone, TLS, IP-Sec, and so on. This work demonstrates that the BMC applications are much simpler, smaller in code size, achieve an order of improvement in performance over their conventional OS-based systems, and they seem inherently secure.
When the BMC becomes a reality, it changes the information technology world in many ways. The hardware vendors have to extend their hardware interfaces rather than discard and rebuild them. The software vendors will be transformed into AO builders and they will be polarized on computer applications rather than computing environments. The AO builders also have to extend their AOs instead of discard and rebuild them. There will be universal AOs that will serve all users in the world as a single source for global objects. The Internet service providers can manage and distribute AOs and charge fee as a single source of service for users. When AOs become global objects, one can build pervasive AOs using the same API for hardware and the same processor architecture in pervasive devices. As the focus will turn into building AOs, thus eventually all necessary AOs will be captured constituting global objects to do global computing. As the AOs are stable over a long period of time, or for ever, there is no need to dump hardware and software frequently, thus saving tremendous product cost, dumping cost, time for retraining, and preserving people's skills. That is, the BMC paradigm will help reduce speed of obsolescence and result in a revolutionary paradigm for the future computing.
The simplest view of BMC can be postulated as follows. You can carry on your own AO in a flash drive and run on any bare machine available anywhere in the world, which is connected to the Internet. In such computing environment, the hardware has no protected resources and readily available everywhere and the AOs are owned by people and can be leased from AO vendors, who can customize them for individual needs. The bare machine vendors should keep their interfaces fairly stable and extend their interfaces as needed for enhanced AOs. Similarly, the AO vendors should also extend their AOs resulting in new AOs while preserving the old AOs. The technology evolution will now focus on development of new and enhanced applications instead of new and unstable environments. The incredible impact of this approach will span across many dimensions including: less disposal of hardware and software, preserve people skills, reduce capital investments, and encourage object-oriented philosophy in every aspect of computing development.
Relevant Publications:
N. Kazemi, A. L. Wijesinha, and R. Karne. Design and Implementation of IPsec on a Bare PC, 2nd International Conference on Computer Science and its Applications (CSA), 2009.
A. Alexander, A. L. Wijesinha, and R. Karne. An Evaluation of Secure Real-Time Protocol (SRTP) Performance for VoIP, 3rd International Conference on Network and System Security (NSS), 2009.
G. H. Khaksari, A. L. Wijesinha, and R. Karne. Secure VoIP using a Bare PC 3rd International Conference on New Technologies, Mobility and Security (NTMS), 2009.
Ford,G.H., Karne, R.K., Wijesinha, A.L., and Appiah-Kubi, P. The Performance of a Bare Machine Email Server, 21st International Symposium on Computer Architecture and High Performance Computing (SBAC-PAD 2009), IEEE / ACM Publications, 28-31 October 2009, Sï¿½o Paulo, SP, Brazil, pp. 143-150.
Ford,G.H., Karne, R.K., Wijesinha, A.L., and Appiah-Kubi, P. The Design and Implementation of a Bare PC Email Server,33rd Annual IEEE International Computer Software and Applications Conference (COMPSAC 2009), Seattle, Washington, July 2009, p480-485.
A. Emdadi, R. K. Karne, and A. L. Wijesinha. Implementing the TLS Protocol on a Bare PC, ICCRD2010, The 2nd International Conference on Computer Research and Development, Kaula Lumpur, Malaysia, May 2010.
He, L., Karne, R.K., Wijesinha, A.L., and Emdadi, A. A Study of Bare PC Web Server Performance for Workloads with Dynamic and Static Content,The 11th IEEE International Conference on High Performance Computing and Communications (HPCC-09), Seoul, Korea, June 2009, p494-499.
Yasinovskyy, R., Wijesinha, A.L., Karne, R.K., and Khaksari, G. A Comparison of VoIP Performance on IPv6 and IPv4 Networks, The 7th ACS/IEEE International Conference on Computer Systems and Applications (AICCSA-2009), Rabat, Morocco. May 10-13, 2009.
He, L., Karne, R.K., Wijesinha, A.L., and Emdadi, A. A Study of Bare PC Web Server Performance for Workloads with Dynamic and Static Content,The 11th IEEE International Conference on High Performance Computing and Communications (HPCC-09), Seoul, Korea, June 2009, p494-499.
Karne, R.K, Wijesinha, A.L, and Ford, G. Opinion-- Stay on course with an evolution or choose a revolution in computing, ACM SIGARCH Computer Architecture News, Volume 36, Number 4, September 2008, p1-6.
He, L., Karne, R. K., Wijesinha, A.L., and Emdadi, A. Design and Performance of a Bare PC Web Server, International Journal of Computers and Their Applications, June 2008.
Khaksari, G., Wijesinha, A.L., Karne, R.K., Qi, Y., and Parikh, K. A VoIP Softphone on a Bare PC, ESA'07-The 2007 International Conference on Embedded Systems and Applications, Las Vegas, Nevada, USA (June 25-28, 2007).
Khaksari, G., Wijesinha, A.L., Karne, R.K., He, L., and Girumala, S. A Peer-to-Peer Bare PC VoIP Application, IEEE Consumer Communications and Networking Conference, Seamless Consumer Connectivity, CCNC 2007, Los Vegas, Nevada, January 2007.
He, L., Karne, R.K, Wijesinha, A., Girumala, S., Khaksari, G. Design Issues in a Bare PC Web Server, Seventh ACIS International Conference on Software Engineering, Artificial Intelligence, Networking, and Parallel/Distributed Computing (SNPD'06) pp.165- 170
Karne, R.K, Venkatasamy, K, Ahmed, T. Dispersed Operating System Computing (DOSC), Onward Track, OOPSLA 2005, San Diego, CA, October 2005.
Karne, R.K, Venkatasamy, K, Ahmed, T. How to run C++ applications on a bare PC, 6th ACIS International Conference on Software Engineering, Artificial Intelligence, Networking, and Parallel / Distributed Computing (SNPD), Towson, Maryland, May 2005
Karne, R.K. Application-oriented Object Architecture: A Revolutionary Approach, 6th International Conference, HPC Asia 2002, December 2002.
Karne, R.K., Gattu, R., Dandu, R., and Zhang, Z., ï¿½Application-oriented Object Architecture: Concepts and Approach,ï¿½ 26th Annual International Computer Software and Applications Conference, IASTED International Conference, Tsukuba, Japan, NPDPA 2002, October 2002.
Karne, R.K., ï¿½Object-oriented Computer Architectures for New Generation of Applications,ï¿½ Computer Architecture News, December 1995, Vol. 23, No. 5, pp. 8-19.

Main Menu

See Also

Main Menu