Use of Network Latency Profiling and Redundancy for Cloud Server Selection

Kwon, Minseok; Dou, Zuochao; Heinzelman, Wendi; Soyata, Tolga; Ba, He; Shi, Jiye

doi:10.1109/cloud.2014.114

Cited by 46 publications

(24 citation statements)

References 21 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The work in [18] and our experiments show that network instabilities (e.g., network congestion due to server workload increment [19], intermediate node dynamic wake-up time scheduling, and/or reprogramming [20,21]) may cause network communications latency to vary over time. CLAS addresses this issue through preprocessing of RTL measurements and using multiple temporal profiling.…”

Section: Introductionmentioning

confidence: 99%

“…In [18], the authors perform extensive experiments to measure network communications latency and conclude that it approximately follows a Gaussian distribution with the mean and standard deviation used to characterize different cloud servers. We partially leverage this observation to develop our authentication scheme.…”

Section: Related Workmentioning

confidence: 99%

“…In this context, RTL is defined as the time elapsed between sending out a packet and receiving its acknowledgment. It has been observed that RTL between two communicating parties (e.g., client and server) approximately follows a Gaussian distribution regardless of the Internet access technology used (e.g., WiFi, wireline, and 3G/4G) [18]. The RTL changes when any of the communicating parties changes the location of its Internet access.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

CLAS: A Novel Communications Latency Based Authentication Scheme

Dou

Khalil²,

Khreishah

2017

Security and Communication Networks

Self Cite

View full text Add to dashboard Cite

We design and implement a novel communications latency based authentication scheme, dubbed CLAS, that strengthens the security of state-of-the-art web authentication approaches by leveraging the round trip network communications latency (RTL) between clients and authenticators. In addition to the traditional credentials, CLAS profiles RTL values of clients and uses them to defend against password compromise. The key challenges are (i) to prevent RTL manipulation, (ii) to alleviate network instabilities, and (iii) to address mobile clients. CLAS addresses the first challenge by introducing a novel network architecture, which makes it extremely difficult for attackers to simulate legitimate RTL values. The second challenge is addressed by outlier removal and multiple temporal profiling, while the last challenge is addressed by augmenting CLAS with out-of-band-channels or other authentication schemes. CLAS restricts login to profiled locations while demanding additional information for nonprofiled ones, which highly reduces the attack surface even when the legitimate credentials are compromised. Additionally, unlike many state-ofthe-art authentication mechanisms, CLAS is resilient to phishing, pharming, man-in-the-middle, and social engineering attacks. Furthermore, CLAS is transparent to users and incurs negligible overhead. The experimental results show that CLAS can achieve very low false positive and false negative rates.

show abstract

Section: Introductionmentioning

confidence: 99%

Section: Related Workmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

CLAS: A Novel Communications Latency Based Authentication Scheme

Dou

Khalil²,

Khreishah

2017

Security and Communication Networks

Self Cite

View full text Add to dashboard Cite

show abstract

“…Results for application classification using supervised learning [37] report F1 scores up to 0.98. There has also been work [21] on QoS profiling using linear models for the purposes of predicting bulk data transfer latency between data centres via an ISP backbone, with error rates up to 3%. Our work goes beyond the state of the art on IP traffic classification of application type and looks at the more difficult task of classifying periods of application packet loss and latency, ultimately providing an indirect way to quantify the severity of these periods.…”

Section: Related Workmentioning

confidence: 99%

Scalable classification of QoS for real-time interactive applications from IP traffic measurements

Middleton

Modafferi

2016

Computer Networks

View full text Add to dashboard Cite

Measurement of network Quality of Service (QoS) has attracted considerable research effort over the last two decades. The recent trend towards Internet Service Providers (ISP's) offering application-specific QoS is creating possibilities for more sophisticated QoS metrics to be offered by ISP's in service level agreements. This in turn could be used for the purposes of improved network optimization and billing according to application specific QoS guarantees. We report a scalable near real-time approach using passively logging IP traffic data for classification of application latency and packet loss across a range of real-time interactive applications. We run six experiments involving Minecraft, Quake 3 Urban Terror, VLC video streaming and the commercial Wirofon VOIP application. We use a mixture of laboratory and realworld deployments, with run times ranging from hours to days, and observe a combination of real and simulated ISP latency and packet loss events. Our binary classification (i.e. classes 'OK' or 'lag') 10-fold cross validation F1 scores are between 0.80 and 0.93 depending on application type. Our multi-class classification (i.e. classes representing discrete packet loss or latency ranges) 10-fold cross validation F1 scores for Minecraft are 0.89 for latency and 0.90 for packet loss. With new business models between ISP's and application developers being actively considered this work represents a significant contribution to the debate by providing scientific evidence relating to a novel approach to scalable QoS measurement

show abstract

“…This network should have transparent platform for reliable delivery of services to distributed users and applications. Thus we are presenting a flow chart of distributed service broker policy (DSBP) algorithm in which discrete recourses are allocated for applications rather than completely isolated systems [12].…”

Section: Network Latency Profiling and Redundancymentioning

confidence: 99%

A New Algorithm for Improving Latency in Distributed Data Center for Logistics Information System over Cloud

Dubey¹,

Jain²

2015

IJCA

View full text Add to dashboard Cite

The cloud based distributed data center uses virtualization technology to share the resources to the outside world through a virtual machine. Cloud administrator selects the data center to access virtual machines by using administrative and dynamic policies. Every data center has multiple virtual machines. Selection of data center is an important task which affects on performance as well as cost effectiveness of the data center. This problem can be solved by centralized as well as distributed data center. In Logistics Company, Centralized data center faces bottleneck in operations like virtual machine migration, creation, deletion, and needs to contact central administrator which can increase the negligible amount of network traffic. The paper presents comparison of distributed and centralized data center and strategies of distributed data center for reducing the latency and cost of selection of data center over the cloud by proposing an algorithm distributed service broker policy (DSBP) for logistics information system.

show abstract

Use of Network Latency Profiling and Redundancy for Cloud Server Selection

Cited by 46 publications

References 21 publications

CLAS: A Novel Communications Latency Based Authentication Scheme

CLAS: A Novel Communications Latency Based Authentication Scheme

Scalable classification of QoS for real-time interactive applications from IP traffic measurements

A New Algorithm for Improving Latency in Distributed Data Center for Logistics Information System over Cloud

Contact Info

Product

Resources

About