1,721,099 research outputs found
Recommended from our members
Efficient Resource Allocation and Management in Cloud Data Centers
No full textWith the rapid development of cloud computing and big data techniques, data centers have become a key infrastructure for most institutions and companies to offload and run their IT-related services and applications. As a result, data centers have presented significant challenges in terms of network load balancing and queue management. Inside a data center, numerous applications generate heterogeneous flows, often with conflicting requirements, to be transferred frequently between servers across the network. However, current load balancing schemes like equal-cost multip-path (ECMP) routing protocol [63] cannot achieve ideal load balancing for data center traffic. Switches in current data centers are manufactured with up to 8 classes of service queues per port. However, current ECN-based transport protocols fall short when applied to switches with multi-queue ports. Thus it is essential for data center operators to develop and use efficient resource allocation and management techniques that satisfy the stringent performance needs of different applications.
In this thesis, we develop new complementary schemes that overcome these challenges. More specifically, firstly, we review and classify recent studies that investigate and measure the traffic behavior of operational DC networks. Then propose CAFT, an in-network congestion-aware, fault-tolerant load balancing protocol for 3-tier Clos data center networks. CAFT overcomes the limitations of the load balancing schemes, designed for 2-tier Clos networks when applied in 3-tier typologies. It leverages the TCP’s 3-way handshaking process to exchange failure and congestion information among switches in real time to make efficient load balancing decisions. The protocol avoids the bottleneck links in asymmetric topologies by allowing aggregation switches to exchange link failure information.
Secondly, we propose ML-ECN, a multi-level ECN marking mechanism that provides fair ECN marking among flows while delivering high throughput and low latency simulta-neously. It classifies queues into small, medium, and large service queues to allow specific ECN marking on each queue class. ML-ECN also introduces a probabilistic marking, at different ECN threshold levels, on different flows based on their sizes to insure fairness among flows in the same queue.
Finally, we propose AQ-CAR, a framework that combines active queue management and congestion-aware load balancing protocol to deliver low latency for short flows while maintaining high throughput for long flows. The framework classifies the queues in each switch port into small, medium, and large classes, each serving a specific flow type. Then it performs adaptive queueing where a flow initially is enqueued at the small service queue and then gets migrated adaptively to the other queues based on the number of bytes it sent. To have congestion and traffic-aware routing and rerouting algorithms, the load balancer first detects the flow type and then piggybacks the congestion information of the queues that serve this flow type. Then it routes/reroutes the flow to the path with minimum congestion
Going Beyond Counting First Authors in Author Co-citation Analysis
Get PDFThe present study examines one of the fundamental aspects of author co-citation analysis (ACA) - the way co-citation
counts are defined. Co-citation counting provides the data on which all subsequent statistical analyses and mappings
are based, and we compare ACA results based on two different types of co-citation counting - the traditional type that
only counts the first one among a cited work's authors on the one hand and a non-traditional type that takes into
account the first 5 authors of a cited work on the other hand. Results indicate that the picture produced through this non-traditional author co-citation counting contains more coherent author groups and is therefore considerably clearer. However, this picture represents fewer specialties in the research field being studied than that produced through the traditional first-author co-citation counting when the same number of top-ranked authors is selected and analyzed. Reasons for these effects are discussed
Recommended from our members
Fair packet enqueueing and marking in multi-queue datacenter networks
Get PDFRecently, Explicit Congestion Notification (ECN) has been leveraged by most Datacenter Network (DCN) protocols for congestion control to achieve high throughput and low latency. However, the majority of these approaches assume that each switch port has one queue while current industry trends towards having multiple queues per switch port. To this end, we propose ML-ECN, a fairness-aware packet enqueueing and multi-level probabilistic ECN marking scheme for DCNs enabled with multiple-service, multiple-queue switch ports. The main design of ML-ECN relies on the separation between small, medium, and large flows by dedicating multiple queues for each flow class to ensure fair enqueueing. ML-ECN employs one ECN marking threshold for the small queue class and multiple thresholds with a probabilistic marking for the medium and large queue classes to achieve low latency for mice (small) and high throughput for elephant (large) flows. In addition, ML-ECN performs fairness-aware ECN marking that ensures that packets of short flows are not getting marked due to buffer buildups caused by longer flows. Large-scale ns-2 simulations show that ML-ECN outperforms existing approaches at different performance metrics
Variations on the Author
Get PDF“Variations on the Author” discusses two of Eduardo Coutinho’s recent films (Um Dia na Vida, from 2010, and Últimas Conversas, posthumously released in 2015) and their contribution to the general question of documentary authorship. The director’s filmography is characterized by a consistent yet self-effacing form of authorial self-inscription: Coutinho often features as an interviewer that rather than express opinions propels discourses; an interviewer that is good at listening. This mode of self-inscription characterizes him as an author who is not expressive but who is nonetheless markedly present on the screen. In Um Dia na Vida, however, Coutinho is completely absent form the image, while Últimas Conversas, on the contrary, includes a confessional prologue that moves the director from the margins to the center of his films. This article examines the ways in which these works stand out in the filmography of a director who offers new insights into the notion of cinematic authorship
Recommended from our members
Failure-resilient congestion-aware load balancing protocol for three-tier clos data centers
Get PDFClos-based network topologies have been deployed in production data center networks to provide multiple path alternatives between the pairs of network hosts. Production data centers operate under varying traffic dynamics and topological asymmetry. Therefore, a good load balancing scheme must adapt to network conditions and dynamics in real-time and intelligently distribute traffic among all possible paths to avoid traffic bottlenecks and to overcome link congestion to be caused by link failures. Yet today's prevalent load balancing scheme in data center networks, equal-cost multi-path (ECMP), is congestion agnostic and performs poorly in asymmetric topologies. In this paper, we propose CAFT, a distributed, congestion-aware, fault-tolerant load balancing protocol for 3-tier data center networks. CAFT first collects, in real-time, link congestion information of two subsets from the set of all possible paths between pairs of hosts. Then, information about the least congested path from each subset is carried across the switches, during TCP's connection establishment process, to make path selection decisions. In the case of topological asymmetry, CAFT avoids bottleneck links by allowing aggregation switches to exchange link failure information. Large-scale ns-3 simulations show that, compared to Expeditus, CAFT achieves slightly better performance in normal cases and significantly better performance in asymmetric cases
Domain-Adaptive Device Fingerprints for Network Access Authentication Through Multifractal Dimension Representation
Get PDFRF data-driven device fingerprinting through the use of deep learning has
recently surfaced as a potential solution for automated network access
authentication. Traditional approaches are commonly susceptible to the domain
adaptation problem where a model trained on data from one domain performs badly
when tested on data from a different domain. Some examples of a domain change
include varying the device location or environment and varying the time or day
of data collection. In this work, we propose using multifractal analysis and
the variance fractal dimension trajectory (VFDT) as a data representation input
to the deep neural network to extract device fingerprints that are domain
generalizable. We analyze the effectiveness of the proposed VFDT representation
in detecting device-specific signatures from hardware-impaired IQ signals, and
evaluate its robustness in real-world settings, using an experimental testbed
of 30 WiFi-enabled Pycom devices under different locations and at different
scales. Our results show that the VFDT representation improves the scalability,
robustness and generalizability of the deep learning models significantly
compared to when using raw IQ data
Recommended from our members
Uplink Performance Characterization and Analysis of Two-Tier Femtocell Networks
No full textThis paper provides a cross-layer analysis of uplink performance in femtocell networks. It characterizes the uplink physical
interference in femtocell networks and studies its impact on the delay and data loss rate of constant-bit-rate (CBR) traffic, as well
as on the maximum achievable femto-user throughput. Our work derives data-link layer QoS performances as a function of physical
layer parameters, thereby establishing key cross-layer relationships that can be useful for designing efficient resource allocation
techniques for FC networks.This is the author's peer-reviewed final manuscript, as accepted by the publisher. The published article is copyrighted by IEEE-Institute of Electrical and Electronics Engineers and can be found at: http://ieeexplore.ieee.org/xpl/RecentIssue.jsp?punumber=25. ©2013 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other users, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works for resale or redistribution to servers or lists, or reuse of any copyrighted components of this work in other works.Keywords: Link Delay and Capacity, Outage Probability, Femtocell Networks, Uplink Interferenc
Recommended from our members
No blind spots : on the resiliency of device fingerprints to hardware warm-up through sequential transfer learning
No full textDeep Learning-based RF fingerprinting has emerged as a game-changer for offering robust network device authentication and identification solutions. However, it struggles in cross-time scenarios, particularly during hardware warm-up phases. This often-overlooked vulnerability jeopardizes the reliability of these solutions. In response to this critical gap, we dive deep into the anatomy of RF fingerprints, revealing insights into temporal variations in DL-based RF fingerprinting during and post hardware stabilization. Introducing HEEDFUL, a novel framework harnessing sequential transfer learning and targeted impairment estimation, we address these challenges with remarkable consistency, eliminating blind spots even during challenging warm-up phases. Our extensive evaluation showcases HEEDFUL's efficacy, achieving remarkable classification accuracies of up to 96% during the initial intervals of device operation-far surpassing traditional models. Cross-domain assessments confirm HEEDFUL's superiority, achieving a steady 87% classification accuracy across warm-up intervals on the Day 2 dataset. Additionally, we release a WiFi RF fingerprinting dataset that, for the first time, incorporates both the time-domain representation and real hardware impairments of the frames. This inclusion underscores the importance of leveraging actual hardware impairment data, enabling a deeper understanding of fingerprints and facilitating the development of more resilient solutions
A network-layer soft handoff approach for mobile wireless IP-based systems
No full textThis material is presented to ensure timely dissemination of scholarly and technical work. Copyright and all rights therein are retained by authors or by other copyright holders. All persons copying this information are expected to adhere to the terms and constraints invoked by each author's copyright. In most cases, these works may not be reposted without the explicit permission of the copyright holder
- …
