Sifat Ferdousi

Disaster-aware datacenter placement and dynamic content management in cloud networks

Recent targeted attacks and natural disasters have made disaster-resilient cloud network design an important issue. Network operators are investigating proactive and reactive measures to prevent huge data loss and service disruptions in case of a disaster. We present novel techniques for disaster-aware datacenter placement and content management in cloud networks that can mitigate such loss by avoiding placement in given disaster-vulnerable locations. We first solve a static disaster-aware datacenter and content placement problem by adopting an integer linear program with the objective to minimize risk, defined as expected loss of content. It is a measure of how much, in terms of cost or penalty, a network operator may lose probabilistically due to possible disasters in a cloud network. We also show how a service providers budget constraint can affect disaster-aware placement design. Since disaster scenarios, content popularity, and/or importance are always changing in time, content placement should rapidly adapt to these changes. We propose a disaster-aware dynamic content-management algorithm that can adjust the existing placement based on dynamic settings. Besides reducing the overall risk and making the network disaster-aware, reducing network resource usage and satisfying quality-of-service requirements can also be achieved in this approach. We also provide a cost analysis of employing a dynamic disaster-aware placement design in the network based on real-world cloud pricing.

Disaster-aware data-center and content placement in cloud networks

Recent occurrences of disasters and targeted attacks have made disaster-resilient data-center network design an important issue. Network operators are investigating proactive and reactive measures to avoid huge data loss and service disruptions in case of a disaster. Disaster-aware data-center and content placement can mitigate the effects of such failures by avoiding placement in disaster-vulnerable locations. Compared to disaster-unaware approach, this approach can significantly reduce the risk, i.e., expected loss of content due to a disaster, in a cloud network.

mSIGMA: An Efficient Handoff Scheme for Multiclass Networks

Wireless networks are becoming more heterogeneous; different classes of networks coexist and users want to connect to any available network, anytime. So, it is important to have a mobility management scheme that can manage handoff for both interclass and intraclass mobility so that the users can connect to and roam between any network. We propose an end-to-end mobility management scheme, Multiclass SIGMA (mSIGMA), that performs soft handoff for interclass and intraclass mobility in wireless network and can perform low-loss location management. Our analysis shows mSIGMA can perform seamless handoff across networks with low delay and packet loss with efficient location management. We have also shown though experimental analysis that mSIGMA is implementable with existing networking technologies and can perform handoff efficiently.

Mixed-line-rate optical network design with wavebanding

Abstract To cope with ever increasing and more heterogeneous traffic demands, today’s optical backbone networks are expected to support mixed line rates (MLR) over different wavelength channels. MLR networks can be designed to provide flexible rate assignments to low-bit-rate services and high-bit-rate services in a cost-effective manner. But with increasing number of wavelengths in the network, aggregating wavelengths into wavebands can further reduce the network cost. In this study, we incorporate the idea of waveband switching in MLR network design. Wavebanding or grouping of optical paths reduces the optical switch size at the optical cross-connects (OXCs). When several lightpaths share several common links, they can be grouped together and routed as a single waveband. For optical bypass at a transit node, only two optical ports are required for each waveband, hence reducing the port cost. It can be a challenge for an MLR network to waveband wavelengths of different line rates that have different transmission reaches. In our design, we present a suitable switching architecture and propose an efficient and cost-effective approach for wavebanding in an MLR network. The design problem is formulated as a mixed integer linear program (MILP) where the objective is to minimize transponder cost and port cost. A heuristic algorithm for wavebanding in MLR networks is provided. To further optimize our solution, we also present a Simulated Annealing algorithm for wavebanding. Our results show a significant improvement in cost savings compared to single-line-rate (SLR) networks with wavebanding and an MLR network employing only wavelength switching.

Rapid data evacuation for large-scale disasters in optical cloud networks

In case of large-scale disasters, cloud networks may be vulnerable to data loss. We propose a rapid data-evacuation strategy to move maximum amount of data from disaster regions using survived resources under strict time constraints.

Rapid data evacuation for large-scale disasters in optical cloud networks [Invited]

Cloud services based on content sharing and distribution are most prevalent in todays world, and datacenter (DC) networks supporting such services manage a huge volume of data transfers, in the range of terabytes to zettabytes. Usually, contents are replicated in geographically distributed DCs for high availability and reliability, but in the case of high-impact, large-scale disasters, like weapons of mass destruction attacks, DC networks can suffer massive service disruptions and data loss. To save critical data under such circumstances, contents could be evacuated in response to an upcoming disaster alert from a likely disaster region to a safe location before the disaster occurs and causes serious data damage. Depending on the forecasted disaster scenario, content evacuation can be greatly constrained by limited available network resources and strict deadlines (evacuation times). We propose a rapid-data-evacuation (RDE) heuristic that selects the least-delay paths (considering propagation delays, network bandwidth, and congestion) through an anycast network model, and schedules critical and vulnerable contents for evacuation such that the maximum amount of contents can be evacuated within the evacuation deadline or equivalently, a given amount of contents can be evacuated in minimum time. We compare our RDE heuristic with a nearest-evacuation approach, which evacuates data only to the nearest DC using the shortest path. Our results show that, for typical scenarios considered in this study, compared to nearest evacuation, our rapid-evacuation approach provides about 64% time savings, or equivalently, about 97% more volume of evacuated contents for a given deadline. Since our algorithm is based on a greedy approach, we also present an enhanced RDE algorithm based on the simulated annealing (SA) metaheuristic as a benchmark. We show that our proposed RDE heuristic performs very close to the SA based algorithm and is much faster in computation time. Hence, it is suitable and efficient for rapid evacuation.

Disaster-Aware Dynamic Content Placement in Optical Cloud Networks

Content placement in cloud networks should be resilient to data loss due to disaster-driven failures. We propose a disaster-aware dynamic content placement scheme to reduce the expected content loss while satisfying resource constraints and QoS.

Green Data Center Placement in Optical Cloud Networks

The rapid growth of cloud services has raised concerns on the environmental sustainability of cloud computing, as data centers (DCs) consume a huge amount of brown energy, i.e., energy derived from polluting sources such as coal, oil, natural gas, etc. One way to decrease carbon emissions of DCs is to replace brown energy with green energy, i.e., energy produced by renewable sources, such as wind farms, solar panels, hydroelectric dams, etc. But, to maximize utilization of green energy, effective decisions need to be taken both at the “DC placement” stage (i.e., when placing DCs) and the “DC addition” stage (i.e., when adding new DCs to accommodate traffic growth). The resulting green DC-placement problem is characterized by a tradeoff between brown energy reduction and cost reduction. On one hand, due to geo-diverse locations of renewable energy sources, and due to the need for low latency and high availability for users, a large number of DCs should be placed. On the other hand, capital and operational expenditure would significantly increase if a large number of DC is deployed. In this paper, we propose two solution methods, based on multiobjective optimization, to address the tradeoff between brown energy consumption and cost in cloud networks in both DC placement and DC addition scenarios. We show via simulations how to choose the optimal number of DCs and their locations over two study cases based on NSFNET and USNET topologies.

mSIGMA: An Efficient Handoff Scheme for Multi-Class Networks

Wireless networks are becoming more heterogeneous; different classes of networks co-exist and users want to connect to any available network, anytime. So, it is important to have a mobility management scheme that can manage handoff for both inter-class and intra-class mobility so that the users can connect to and roam between any network. We propose an end-to- end mobility management scheme, Multi-class SIGMA (mSIGMA), that performs soft handoff for inter-class and intra-class mobility in wireless network. Our analysis shows mSIGMA performs seamless handoff across networks with low delay and packet loss. We have also shown though experimental analysis that mSIGMA is implementable with existing networking technologies and can perform handoff efficiently.

Progressive datacenter recovery over optical core networks after a large-scale disaster

Todays cloud system are composed of geographically distributed datacenter interconnected by high-speed optical networks. Disaster failures can severely affect both the communication network as well as datacenters infrastructure and prevent users from accessing cloud services. After large-scale disasters, recovery efforts on both network and datacenters may take days, and, in some cases, weeks or months. Traditionally, the repair of the communication network has been treated as a separate problem from the repair of datacenters. While past research has mostly focused on network recovery, how to efficiently recover a cloud system jointly considering the limited computing and networking resources has been an important and open research problem. In this work, we investigate the problem of progressive datacenter recovery after a large-scale disaster failure, given that a network-recovery plan is made. An efficient recovery plan is explored to determine which datacenters should be recovered at each recovery stage to maximize cumulative content reachability from any source considering limited available network resources. We devise an Integer Linear Program (ILP) formulation to model the associated optimization problem. Our numerical examples using the ILP show that an efficient progressive datacenter-recovery plan can significantly help to increase reachability of contents during the network recovery phase. We succeeded in increasing the number of important contents in the early stages of recovery compared to a random-recovery strategy with a slight increase in resource consumption.