Matlab scripts for ``The Average Availability of Parallel Checkpointing Systems and Its Importance in Selecting Runtime Parameters''

  • James S. Plank
  • Michael G. Thomason
    See this paper for more detail on the checkpointing model, the Markov model, and this code.

    This material is based upon work supported by the National Science Foundation under Grant No. 9703390.

    These are scripts for determining the long-term availability of a distributed memory checkpointing system with the following parameters:

    There are five Matlab scripts that you'll need to have: Mike wrote these scripts, and I'm not much of a Matlab user, so I can't attest to exactly what each script does. However, if you fire up Matlab with these in the current directory, then you put the parameters into a file called PARAMS, which should have: N, a, C, L, R, lambda, theta and I in that order. Here is an example PARAMS file.

    Then run AVAIL, and it will spit out the availability. Here is a transcript using the example PARAMS file:

    UNIX> matlab
                                < M A T L A B (R) >
                    (c) Copyright 1984-98 The MathWorks, Inc.
                                All Rights Reserved
                                    Jan 17 1998
                    Your MATLAB license will expire in 23 days.
                    Please contact your system administrator or
                    The MathWorks to renew this license.
      To get started, type one of these: helpwin, helpdesk, or demo.
      For product information, type tour or visit
    >> AVAIL
    WARNING: The only check made on any values is I >= L. (Hit enter)
    File PARAMS is read for N,a,C,L,R,lam,theta,I in order. (Hit enter)
    PARAMS =
       1.0e+04 *
    Availability =
    >> quit
     40453714 flops.

    It will also create the matrices P, U and D, in case you want to look at them.