Many distributed computing applications need transfer large files between distributed locations as fast as possible. A
dynamic routing algorithm for optical network is designed to modify existing transfers and spare network resources for
new request to satisfy both old and new transfers' requirements. In data intensive application on circuit-switch optical
network, light-path resources are scarce and there should be concurrent file transfers competing for the same fibers. In
static routing optical network, if new coming file transfer cannot acquire light-path with enough bandwidth, it could only
wait for the releasing of current used resources. Due to the waiting, the delay time will be large. So we use our dynamic
routing algorithm to schedule and modify existing light-paths, to spare a light-path with enough bandwidth for new
coming file. Our optimized target is to make every file finish transferring in less time, so we propose two objectives
defined in the paper: one is to make maximal delay time of all tasks less and the other is to make average delay time less.
The algorithm proposed has two mainly steps: 1) Routing process; 2) Dynamic routing process. In routing step, when
task of file arrives we firstly get k random paths, then use Least Congestion Algorithm (LCA) (or Shortest Path
Algorithm (SPA)) to get the primary path P1 of maximal residual bandwidth (RB) from k paths and the alternate path P2
of the second maximal RB. If the bandwidth of P1 is enough for this task, transfer the file in P1 path. If not, we go to the
dynamic routing process. In the second process, get all the links of P1 then we change the existing light paths of tasks in
the P1 path one by one to their alternative paths until we can get enough bandwidth of P1. In the dynamic routing
process, we design two different queuing strategies. The first strategy is First Arrive First Modified (FAFM) strategy,
namely we schedule the first arrival task firstly. The other is Larger Bandwidth First Modified (LBFM) and the file with
larger bandwidth is scheduled firstly.
By comparison of simulation results, we can prove that our two kinds of dynamic routing algorithms can get better
results for both decreasing maximal delay time and average delay time than LCA and SPA routing algorithms. In the two
queuing strategies, LBFM can get better results than FAFM strategy. The receivers in the destinations can get better
results by using our dynamic routing algorithm.
Optical Grid that provides a way to integrate the geographical distributed resources for large-scale grid applications. This
paper mainly focuses on the task scheduling issue in Optical Grid and proposes two task scheduling algorithms to minimize
the total time or scheduling length required for a given task. A simulated optical grid has been addressed to demonstrate
the effectiveness of the scheduling approach. The performance tests and analysis of the two algorithms are investigated
to show the effect of the routing policy on the scheduling length. A more flexible policy is able to further improve
scheduling length.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
INSTITUTIONAL Select your institution to access the SPIE Digital Library.
PERSONAL Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.