Performance of Slotted Store-and-Forward (sSnF)
Optical Circuit-Switched Networks -
A Simulation Study
Xiao Lin, Weiqiang Sun and Weisheng Hu
State Key Laboratory of Advanced Optical Communication Systems and Networks
Shanghai Jiao Tong University, Shanghai, China, 200240
Email: {linxiaocer, sunwq, wshu}@sjtu.edu.cn
Abstract—Increasing bulk data transfers have been
overwhelming the Internet. To overcome this, optical circuit-
switched (OCS) networks are equipped with assistive storage, so
that bulk data that are delay tolerant can be temporarily stored
at intermediate nodes and forwarded at later times. But, the use
of storage greatly complicates the routing problem, since data
storage must be incorporated into routing. This motivates us to
simplify this issue by applying slotted operations for the network.
Intuitively, the slotted network suffers from degraded network
performance due to the inefficient utilization incurred by the slot
constraint. However, our simulation shows that when the slot size
equals to half the mean duration, the blocking probability is
reduced from 0.076 to 8.5×10
-5
, and the number of network
reconfigurations is reduced by a factor of 5, compared to the
unslotted case. We reveal that in spite of the inefficient utilization,
the slotted operations mitigate bandwidth fragmentation. This
suggests in the slotted case, more bandwidth gaps on the links are
available for accommodating other requests, and they are aligned
with each other in time. Requests hence are delivered with less
store-and-forward (SnF) operations being performed. Thus,
when the number of SnF allowed for routing each request is
limited (in order to reduce the computational complexity of
routing), requests are more easily served in the slotted than in the
unslotted cases. Our research provides clue for designing scalable
slotted OCS networks with assistive storage.
Keywords—Bulk data transfer; optical circuit switching;
routing; storage; store-and-forward.
I. INTRODUCTION
Every day, many bandwidth-hungry applications, ranging
from commercial and scientific (e.g., data migration among
datacenters and grid computing) to consumer applications (e.g.,
IPTV and video on demand), are pouring multiple petabytes of
data into the Internet. The rapidly increasing bulk data transfers
incur unprecedented bandwidth contention and hence have
been overwhelming the Internet [1].
As a basic switching paradigm in optical networks, optical
circuit switching (OCS) provides guaranteed-bandwidth paths
and quality of service guarantees to users with minimum
control overhead. It appears to be a promising solution for bulk
data transfer. However, OCS requires reserving a fixed amount
of bandwidth over all links along the entire path prior to data
transfer, which may lead to inefficient utilization of bandwidth
[2]. In addition, large variation in the residual bandwidth under
dynamic traffic demands makes determining the bandwidth to
reserve over all links along the path more difficult [3].
To overcome this challenge, the nodes of the OCS network
are equipped with assistive storage. In this way, bulk data that
are delay tolerant can be temporarily stored at intermediate
nodes and forwarded at later times, i.e., performing Store-and-
Forward (SnF) using assistive storage inside the intermediate
nodes. This significantly mitigates bandwidth contention and
improves resource utilization. So far bulk data transfers have
benefited greatly from SnF OCS, in terms of network
performance, resource utilization and transfer cost [1-3].
However, the use of storage introduces additional
complexity into the conventional routing process, which makes
this difficult problem even more complex. For example, upon
receiving a request, the network must decide which the
intermediate nodes are used for storing, whether they have
sufficient storage capacity, for how much time the request
should be stored in each intermediate storage node, and at what
rate it should be transmitted to the next “hop” along the path.
Additionally, due to time-varying link capacities and storage
capacities, as well as additional network and storage
bottlenecks that affect the transfer completion time of each
request, those decisions are difficult to make, especially if an
optimization objective is involved [4,5].
Such complex routing issue naturally motivates us to
simplify it. One way is applying slotted network operations for
the SnF OCS network, where network operations, such as
resource reservation/release, only happen on the slot
boundaries. Intuitively, the slotted network suffers from
degraded network performance. Because the slot constraint
leads to inefficient utilization, e.g., reservation can only be
setup at the start of each time slot and reservation time can only
be an integer number of time slots. In other words, the slotted
operations offer the issue simplification at the cost of ineffici-
ent utilization as well as degraded network performance.
Interestingly, our studies herein illustrate that slotted operations
potentially provide better blocking and control benefits to the
SnF OCS network, compared with the unslotted case.
In this paper, we consider the slotted store-and
-fo
rward
optical circuit-switched network, i.e., the sSnF OCS network,
where network operations are slotted, and bulk data can be
temporarily stored at intermediate nodes and forwarded at later
times. Our contributions are summarized as follows:
1) We provide a comparison of network performance between
the slotted and the unslotted SnF OCS networks. Our
studies show that under certain conditions, the slotted
978-1-5090-2839-9/17/$31.00 ©2017 IEEE