Vortrag: Prof. Barton Miller, 01.12.1998

[AURORA Project]

                             UNIVERSITAET WIEN 
              INSTITUT FUER SOFTWARETECHNIK UND PARALLELE SYSTEME
                               gemeinsam mit 
                                 VCPC 
                EUROPEAN CENTRE FOR PARALLEL COMPUTING AT VIENNA 
                        
                       FWF-Projekt SFB F011 "AURORA"


      EINLADUNG ZU EINEM VORTRAG IM RAHMEN DES INSTITUTS-KOLLOQUIUMS
                        
                
         
                                                               
                          Adaptive Operating Systems: 
                    An Architecture for Evolving Systems

                   
 			   
                              Barton P. Miller
                         Computer Sciences Department
                            University of Wisconsin
                          {tamches,bart}@cs.wisc.edu
             
             
                 ZEIT: Dienstag, 1. 12. 1998, 17.00 Uhr c.t.
          ORT: Institut fuer Softwaretechnik und Parallele Systeme
                   1090 Wien, Liechtensteinstrasse 22, 
                          Seminarraum, Mezzanin


Abstract


Operating systems used to be viewed as static entities, changing almost as
slowly as the underlying hardware.  Recent commercial systems have provided
for a small amount of run-time change by allowing device drivers to be
installed while the system is running.  We are developing "adaptive operating
systems" whose code can change and evolve while the system is running.  This
adaptation can be used to instrument the code for profiling or debugging
purposes, or to modify and extend the operating system to adapt to changing
work loads, application demands, and configuration.

Our work differs from other efforts in this area in two ways.  First, we
can instrument and modify a stock, commercial operating system (Solaris) as
it is delivered to the customer.  We operate directly on the executable code
while it is running.  Second, we can modify the operating system at almost
any point in its code.  We are not constrained to system call or procedure
call replacements.

Our research is embodied in a facility called KernInst.  We will describe
the basic KernInst mechanism and several uses of KernInst, including
performance profiling and dynamically modifying and customizing the operating
system in response to its work load.  We will also present a case study
using KernInst to profile the Solaris kernel under a Web proxy server (Squid)
workload.

A paper on kerninst is available from:
    ftp://grilled.cs.wisc.edu/technical_papers/kerninst.ps
    ftp://grilled.cs.wisc.edu/technical_papers/kerninst.ps.Z
    ftp://grilled.cs.wisc.edu/technical_papers/kerninst.ps.gz
    ftp://grilled.cs.wisc.edu/technical_papers/kerninst.pdf