For my Operating Systems theory class, we were required to write a term paper on a topic of our choice. I chose file systems, one to be a little different than the vast majority of people who were choosing the recommended topic, and two, because I see it as an area that will need a huge revamp in the coming years. The area of file systems has been a relatively quiet one as of late, the only newest advancement in public use being EXT4 for Linux operating systems. Many devices are going portable, and by effect are going flash based. I don’t blame them, in fact flash is a fantastic technology, it’s temperature tolerent, shock resistant and has lower power consumption. The problem with the technology is it wears out, and it wears out much faster than traditional media unless the issue of wear leveling is addressed in the design of the driver.
The paper details research that other individuals have done into the sustainability of file systems in various environments, and I relate them to their importance to flash memory. File systems are going to need to undergo a drastic redesign in the next few years if we are to provide faster technology that can handle higher bandwidth media either locally or through network storage. I also included a short section on DIY information for flash storage solutions out of simple USB enabled flash memory that would allow an individual to obtain speeds faster than traditional hard drives. The scope of the material is limited, but I will be posting a tutorial in the near future in relation to how to mount yourself a flash RAID out of your USB keys as well as recommended file systems for the fastest access times and wear leveling. For more information, I have posted the paper in my repository under ‘Papers’ and it is free for all to read, use and critique. If you have any comments, feel free to leave them below. Also, a follow up on my Web 2.0 concepts soon to come.