The Final Research Project
I am finally done planning my research project, now simply titled "Distributed Operating Systems". The programming project will remain the exact same. I will design and implement:
automatic process migration
paging over the network (including memory mapped files)
distributed file operations (files are addressed as :)
some form of IPC, without which real evaluation would be impossible
many different schemes of speeding up these actions
algorithms on how to route to process to maximize performance
This will take quite a while and should give me a lot of experience in the trade offs involved in distributed operating systems. It will also allow me to a lot of further research on this and related topics in the future.
It should be noted that I will most likely not yet address security, reliability (I will assume nodes and connections never go down), trading of processor time/cycles, or reliable distributed file storage that I would like to address some time in the future.
In addition to the programming I will read a great number of books on Operating Systems, Networking, Compilers, Distributed Systems and Computer Architecture to help me with understanding the applicability of my research on real operating systems, and the trade offs and problems associated with those.
Finally, if all goes well, I will attend the
3rd USENIX symposium on Virtual Machine Research and Technology (VM '04), May 6-7. This mostly depends on whether I can get funding for my project approved quickly enough, and I really hope that this will work out, because both
Professor Rosenblum and
Miguel de Icaza will be keynoting at the event.
I am extremely happy to have finished planning the project, and hope that everything goes well.
Posted by csar at April 20, 2004 12:23 AM
to category Computer Science
| TrackBack