BLOC: A Generic Resource Allocation Framework for Hybrid Packet/Circuit-Switched Networks

Zhangxiao Feng; Weiqiang Sun; Weisheng Hu

doi:10.1364/JOCN.8.000689

Journal of Optical Communications and Networking
Vol. 8,
Issue 9,
pp. 689-700
(2016)
•https://doi.org/10.1364/JOCN.8.000689

BLOC: A Generic Resource Allocation Framework for Hybrid Packet/Circuit-Switched Networks

Zhangxiao Feng, Weiqiang Sun, and Weisheng Hu

Not Accessible

Your library or personal account may give you access

Get PDF
Email
Share
Get Citation
Copy Citation Text
Zhangxiao Feng, Weiqiang Sun, and Weisheng Hu, "BLOC: A Generic Resource Allocation Framework for Hybrid Packet/Circuit-Switched Networks," J. Opt. Commun. Netw. 8, 689-700 (2016)

Export Citation
- BibTex
- Endnote (RIS)
- HTML
- Plain Text
Citation alert
Save article

Abstract

In the current Internet, large flows co-exist with small flows and often consume a significant proportion of the network bandwidth. This imposes a challenge on network resource management. Various hybrid switching architectures have been proposed in recent years, aiming to address this challenge and improve energy efficiency and resource utilization. However, there is still a need for means for the design and dimensioning of resources in multiple switching planes and understanding the system behavior under different traffic patterns. In this paper, we propose a framework that we call the blocking-loss curve (BLOC) to address the resource allocation problem in hybrid switching systems. BLOC can capture the behavior of hybrid switching systems under different resource allocation solutions considering packet loss and request blocking. This framework also allows us to identify the feasible region for resource allocation, which can satisfy both the packet loss and request blocking requirements. An optimal resource allocation strategy may be found based on this feasible region. We also study the application of BLOC to a specific hybrid switching system, where packet switching comprises TCP flows with active queue management, and circuit switching is single granular, e.g., lightpath switching. Numerical results demonstrate how the flow characteristics and system parameters may affect the feasible region and the optimal allocation of resources.

Full Article | PDF Article

More Like This

Resource Allocation in Electrical/Optical Hybrid Switching Data Center Networks

Zhangxiao Feng, Weiqiang Sun, Jie Zhu, Junyi Shao, and Weisheng Hu
J. Opt. Commun. Netw. 9(8) 648-657 (2017)

Dynamic Wavelength Allocation and Analytical Model for Flow Assignment in Optical Packet and Path Integrated Networks

Onur Alparslan, Shin’ichi Arakawa, and Masayuki Murata
J. Opt. Commun. Netw. 9(4) 304-318 (2017)

Time-Shifted Multilayer Graph: A Routing Framework for Bulk Data Transfer in Optical Circuit-Switched Networks With Assistive Storage

Xiao Lin, Weiqiang Sun, Malathi Veeraraghavan, and Weisheng Hu
J. Opt. Commun. Netw. 8(3) 162-174 (2016)

Previous Article Next Article

Cited By

You do not have subscription access to this journal. Cited by links are available to subscribers only. You may subscribe either as an Optica member, or as an authorized user of your institution.

Contact your librarian or system administrator
or
Login to access Optica Member Subscription

Figures (7)

You do not have subscription access to this journal. Figure files are available to subscribers only. You may subscribe either as an Optica member, or as an authorized user of your institution.

Contact your librarian or system administrator
or
Login to access Optica Member Subscription

Tables (5)

You do not have subscription access to this journal. Article tables are available to subscribers only. You may subscribe either as an Optica member, or as an authorized user of your institution.

Contact your librarian or system administrator
or
Login to access Optica Member Subscription

Equations (10)

You do not have subscription access to this journal. Equations are available to subscribers only. You may subscribe either as an Optica member, or as an authorized user of your institution.

Contact your librarian or system administrator
or
Login to access Optica Member Subscription

Scenario	Method	Model	Resource Allocation Strategy
Edge node	OBS/OCS [16]	Parallel	Long-lived and short-lived flows are transferred by OCS and OBS, respectively. Resources are allocated to OCS in preference to OBS.
Backbone	OBS/OCS [18]	Integrated	Circuits have non-preemptive priority over bursts. Circuits are allowed to queue and wait for an available wavelength.
	OPS/OCS [19]	Integrated	Some wavelengths are shared between OPS and OCS, and resource allocation is automatically adjusted according to the number of established lightpaths.
	PS/CS [20]	Integrated or parallel	Loaning backup resources from the circuit service to the packet service, and vice versa, but without sacrificing the survivability of both services.
Data center	EPS/OCS [14,15]	Parallel	Static resource allocation, where resources are not shared between the OCS and EPS planes. The OCS plane has higher priority for transmitting traffic.

Symbol	Definition
Variable:
$V_{PS} %$	Percentage of traffic volume handled by PS
$R_{PS} %$	Percentage of resources allocated to PS
Input:
$V$	Total traffic volume
$R$	Total resources
$P_{\max}$	Maximal allowed packet loss rate
$B_{\max}$	Maximal allowed request blocking probability
Output:
${(V_{PS} %, R_{PS} %)}$	Feasible combinations of traffic partitioning and resource allocation
$(V_{opt} %, R_{opt} %)$	The optimal resource allocation

Symbol	Definition
Variable:
$T_{f}$	Flow size threshold
$s$	The maximal number of circuits in CS
$B_{p}$	The bandwidth allocated to PS
$v$	The data volume of a flow (flow size)
$v_{p}$	The data volume of a flow transmitted by PS
$v_{c}$	The data volume of a flow transmitted by CS
Dependent Variable:
$λ_{p} (T_{f})$	Flow arrival rate in PS
$λ_{c} (T_{f})$	Flow arrival rate in CS
$g (v)$	Probability density function of flow size distribution
$G (v)$	Cumulative distribution function of flow size distribution
$g_{p} (v_{p}, T_{f})$	Probability density function of flow size distribution in PS
$g_{c} (v_{c}, T_{f})$	Probability density function of flow size distribution in CS
$a (T_{f})$	Traffic load in CS
$B (s, a)$	Request blocking probability in CS
$n (T_{f}, B_{p})$	The average number of flows
$p (n, B_{p})$	Packet loss rate in PS
Constant:
$C$	Total system capacity
$B_{0}$	The capacity of each circuit
$λ$	Flow arrival rate
$D$	The propagation delay
$L_{\min}$	The minimal queue length threshold
$L_{\max}$	The maximal queue length threshold
$D_{\max}$	The maximal dropping probability
$c$	A constant related to the TCP version
$p_{Size}$	Packet size

More elephants	↑	$V_{elep} ∶ V_{mice}$	$V_{PS} %$ ^a	$R_{PS} %$ ^b	↓	More mice
		$100 ∶ 1$	0	0
		$10 ∶ 1$	8.24	13.25
		$1 ∶ 1$	49.70	48.53
		$0.1 ∶ 1$	89.91	76.83
		$0.01 ∶ 1$	97.75	82.12

	Area of Feasible Region	$R_{PS} %$
Flow size distribution (proportion of mice flows)	Uncertain	Positive
Reconfiguration delay of circuits	Negative	Positive
$P_{\max}$ , $B_{\max}$	Positive	Negative
$L_{\min}$ , $L_{\max}$	Positive	Negative
Traffic total data volume	Negative	Positive
Load overall resources	Positive	Negative

Abstract

Cited By

Figures (7)

Tables (5)

Equations (10)

Journal of Optical Communications and Networking