The MASM Forum

General => The Workshop => Topic started by: Raistlin on February 13, 2017, 04:59:23 PM

Title: Distributed File System ideas and challenges
Post by: Raistlin on February 13, 2017, 04:59:23 PM
So, as some might know I've been working hard on a top secret application (exactly as to  'the what' is pasted all over the forum).

Part 1: The hardware enumerator is done  to the point of functionality for my intended use.
            (I still need to convert it to a DLL so all enthusiasts might reap the benefit of detailed
              system detection within their apps - I'll update the thread when done  ;)) (probably in a week)

Now for Part 2 and this might be a tall order - a distributed file system (master/slaves) sporting these characteristics:
-> high availability, close-to real time fail over, minimal bandwidth usage, dynamic replication node
      reallocation (ex. new nodes added, dead nodes removed), persistent storage across session runs

If anyone's interested in the subject, the seminal article by Levy 1990
discusses the ideas mentioned: http://www.cs.virginia.edu/~zaher/classes/CS656/levy.pdf

<EDIT:> Here's another more recent article probably more to the point of what I'am about to create:
https://www.microsoft.com/en-us/research/wp-content/uploads/2011/06/tidyfs.pdf

More to the point - if someone has done some work in these areas - I'd be happy (ecstatic !)  to hear your thoughts and advice.