Search

F1ows that have exceeded a given percentage of the last sliding window of N packets, denoted as heavy-hitter flows, require special handling, since they may disrupt the service of other flows or may be indicative of malicious traffic. However, even when equipped with a programmable switch, it is unclear how to detect heavy hitters on a per-packet basis, while obeying the stringent switch memory access rates. For instance, existing solutions, such as HashPipe, cannot detect heavy hitters without halving the line rate and do not support sliding windows. To the best of our knowledge, this paper is the first to present heavy-hitter detection solutions that provide per-packet granularity at line-rate performance. We realize this by introducing (1) Modulo sketching, a novel counting algorithm that reuses counters and limits the impact of smaller flows beyond early processing stages; and (2) Sequential Zeroing, a new approach to extending interval-based schemes to sliding window measurements. Our solutions are extensively evaluated, both via simulations and experiments on a Netronome SmartNIC, and demonstrate significant performance gains over the state-of-theart., Virtual/online event due to COVID-19 accepted author manuscipt, Embedded and Networked Systems

LoRaWAN has emerged as a popular IoT commu- nications technology. It comes with three classes of operation: A, B, and C. Although many IoT use-cases, like Firmware-over- the-Air updates, require multicast, Class A cannot be used for that purpose. Class C can, but consumes a lot of energy. This leaves Class B. In this paper, we investigate Class B multicast and its scalability properties. Issues like multicast member capacity, beacon blocking, and beacon collisions are highlighted, and several approaches to mitigate them are proposed: (1) “Ping-Slot Relaying,” to allow for more multicast members, (2) a scheduling approach indicating when to best send multicast packets, and (3) “Dynamic Region Formation” to coordinate the sending of beacons over multiple gateways. The proposed solutions do not require any modifications to the LoRaWAN protocol., Virtual/online event due to COVID-19 accepted author manuscript, Embedded and Networked Systems

F1ows that have exceeded a given percentage of the last sliding window of N packets, denoted as heavy-hitter flows, require special handling, since they may disrupt the service of other flows or may be indicative of malicious traffic. However, even when equipped with a programmable switch, it is unclear how to detect heavy hitters on a per-packet basis, while obeying the stringent switch memory access rates. For instance, existing solutions, such as HashPipe, cannot detect heavy hitters without halving the line rate and do not support sliding windows. To the best of our knowledge, this paper is the first to present heavy-hitter detection solutions that provide per-packet granularity at line-rate performance. We realize this by introducing (1) Modulo sketching, a novel counting algorithm that reuses counters and limits the impact of smaller flows beyond early processing stages; and (2) Sequential Zeroing, a new approach to extending interval-based schemes to sliding window measurements. Our solutions are extensively evaluated, both via simulations and experiments on a Netronome SmartNIC, and demonstrate significant performance gains over the state-of-theart., Virtual/online event due to COVID-19 accepted author manuscipt, Embedded and Networked Systems

LoRaWAN has emerged as a popular IoT commu- nications technology. It comes with three classes of operation: A, B, and C. Although many IoT use-cases, like Firmware-over- the-Air updates, require multicast, Class A cannot be used for that purpose. Class C can, but consumes a lot of energy. This leaves Class B. In this paper, we investigate Class B multicast and its scalability properties. Issues like multicast member capacity, beacon blocking, and beacon collisions are highlighted, and several approaches to mitigate them are proposed: (1) “Ping-Slot Relaying,” to allow for more multicast members, (2) a scheduling approach indicating when to best send multicast packets, and (3) “Dynamic Region Formation” to coordinate the sending of beacons over multiple gateways. The proposed solutions do not require any modifications to the LoRaWAN protocol., Virtual/online event due to COVID-19 accepted author manuscript, Embedded and Networked Systems

The blurring of boundaries between work life and private life, between Spredsheets and Facebook, between societal and private means that we need to reconsider learning both in the perspective of adoption of technological change and, as diffusion of continuous innovation. Both organizations and technologies are undergoing fundamental changes that transform and create new challenges in the ways we work and learn. Many organizations today require continuous development and effective learning processes to meet the challenges of globalization and digitalization. The questions that need to be raised are what new skills need to be recruited, and how can the capabilities and functionality be distributed among a mixture of both human and technical "workforce"? In this position paper we discuss arguments for a future research agenda where new digital phenomena's are viewed from a"learning at the digital boundaries" perspective, taking into account different waves of digitalization and infrastructural challenges at the boundaries of organizational setting and private life.

The digitalization of the workplace and society at large brings new challenges for the field of Information Systems. A better understanding of how we learn with digital artifacts in our daily routines is needed. This paper is a literature review of how workplace learning has been addressed within the field, bringing together workplace learning and IS. The aim is to provide an alternative perspective to further IFIP 8.6 as a working group, where the suggestion is to re-image the group towards learning with digital artifacts. IS should be the leading field addressing digitalization of society. However, so far learning theorists has not been fully utilized within IS in fostering our understanding of how digitalization affects society. We argue that there is a need to explicitly talk about learning in IS and suggest IFIP 8.6 to be that place, which could contribute to advancements in our IS field in general.

The objective of this paper is to explain the importance of research in wastewater transportation (sewage systems) using new technologies such as robotics systems and information and communication technologies. ECHORD++ (European Coordination Hub for Open Robotics Development) is a very useful tool to foster this research and to meet needs and solutions. In this paper, authors explain the tool as well as the methodology to promote robotics research in urban environments, and the on-going experience will demonstrate that huge advances are made in this field., Peer Reviewed, Postprint (author's final draft)

Multipath routing is useful for networks to achieve load sharing among multiple routing paths. Multipath BGP (M-BGP) is a technique to realize inter-domain multipath routing by enabling a BGP router to install multiple equally-good routes to a destination prefix. Most of previous works did not distinguish between intra-domain and inter-domain multipath routing. In this paper, we present a measurement study on the deployment of M-BGP in a large Internet service provider (ISP) network. Our method combines control-plane BGP measurements using Looking Glasses (LG), and data-plane traceroute measurements using RIPE Atlas. We focus on Hurricane Electric (AS6939) because it is a global ISP that connects with hundreds of major exchange points and exchanges IP traffic with thousands of different networks. And more importantly, we find that this ISP has by far the largest number of M-BGP deployments among autonomous systems with LG servers. Specifically, Hurricane Electric has deployed M-BGP with 512 of its peering ASes at 58 PoPs around the world, including many top ASes and content providers. We also observe that most of its M-BGP deployments involve IXP interconnections. Our work provides insights into the latest deployment of M-BGP in a major ISP network and it highlights the characteristics and effectiveness of M-BGP as a means to realize load sharing.

Risk analysis methods help evaluate the costs of information security controls in relation to their benefits. Despite dramatic changes in the constellation of information security risks, the basic approach to these risk calculation methods remains unchanged. The fundamental mathematics underlying these methods is anchored to probability theory. Probability has the advantage of being widely known and conceptually simple. But it has a disadvantage in its grounding on expert estimates of frequency data because such data is often publicly unavailable. This paper proposes the use of possibility theory as an alternative grounding for information security risk calculations. Possibility theory assumes the data grounding will be estimations. The estimations include expert evaluations of both possibility and likelihood of risks. Using a design science research approach, we use possibility theory as the kernel theory in developing and evaluating a practical possibility-based risk estimation prototype. The results offer an expanded grounding to improve information security risk analysis through the use of a broader portfolio of distinct methodologies anchored to alternative mathematical theories of evidence.

Passive Optical access Networks (PONs) are currently the major contributor to the energy consumption budget of fixed optical networks. In PON, the largest part of the energy consumption is due to the equipments at the customer premises. This paper proposes a method for maximizing energy savings while providing services with delay guarantees (i.e., frame delivery time and frame delay variation). The method combines service-based variable sleep period and a queueing theory model to compute the optimal sleep time. Simulation results prove the effectiveness of the method for a Poisson frame arrival process., QC 20150707

What factors drive individuals to abuse information systems? Better understanding of the roots behind this individual decision could provide opportunities to reduce computer abuse by reducing the presence of these factors. In this paper, we examine the effects of both organizational and personal factors on an individual’s computer abuse behavior. We develop our theoretical model based on abuse opportunity structure theory and literature on emotion theory. Specifically, we identify the organization abuse structure as an organizational factor and three personal factors (goal conduciveness, abuse positive affect, and morality) as personal factors. We investigate their effects on the assessments and decisions that individuals make about computer abuse. The results of a controlled laboratory experiment showed that the organization abuse structure affects these decisions through the assessment of goal conduciveness and abuse positive affect. Morality, however, was found to directly and indirectly affect an individual’s abuse intent. These results imply that security practices are subject to individual appraisals that raise emotions that are in turn conditioned by morality in specific ways. A technical evaluation of the practices is incomplete when absent of the individual appraisal.

Originally implemented by Google, QUIC gathers a growing interest by providing, on top of UDP, the same service as the classical TCP/TLS/HTTP/2 stack. The IETF will finalise the QUIC specification in 2019. A key feature of QUIC is that almost all its packets, including most of its headers, are fully encrypted. This prevents eavesdropping and interferences caused by middleboxes. Thanks to this feature and its clean design, QUIC is easier to extend than TCP. In this paper, we revisit the reliable transmission mechanisms that are included in QUIC. More specifically, we design, implement and evaluate Forward Erasure Correction (FEC) extensions to QUIC. These extensions are mainly intended for high-delays and lossy communications such as In-Flight Communications. Our design includes a generic FEC frame and our implementation supports the XOR, Reed-Solomon and Convolutional RLC error-correcting codes. We also conservatively avoid hindering the loss-based congestion signal by distinguishing the packets that have been received from the packets that have been recovered by the FEC. We evaluate its performance by applying an experimental design covering a wide range of delay and packet loss conditions with reproducible experiments. These confirm that our modular design allows the protocol to adapt to the network conditions. For long data transfers or when the loss rate and delay are small, the FEC overhead negatively impacts the download completion time. However, with high packet loss rates and long delays or smaller files, FEC allows drastically reducing the download completion time by avoiding costly retransmission timeouts. These results show that there is a need to use FEC adaptively to the network conditions.

IP mobility ensures network reachability and session continuity while IPv6 networks are on the move. In the Network Mobility (NEMO) model, the potential for NEMO Mobile Routers (MRs) to interconnect and extend Internet connectivity allows the formation Nested NEMO networks. With MANEMO, nested MRs can be efficiently interconnected in a tree-based structure with Internet access being maintained via a designated Gateway. However, this only supports single-homed Internet connectivity. With the span of wireless access technologies and the popularity of multi-interfaced devices, multihoming support in this scenario becomes critical. A Nested Mobile Network with heterogeneous available Internet access options would allow better overall network performance and optimal utilisation of available resources. In this paper, we present the Multihomed Mobile Network Architecture (MMNA), a comprehensive multihomed mobility solution. It provides a multihoming management mechanism for Gateway Discovery and Selection on top of a multihomed mobility model integrating different mobility and multihoming protocols. It enables a complex nested multihomed topology to be established with multiple gateways supporting heterogeneous Internet access. The results demonstrate that the proposed solution achieves better overall throughput, load sharing, and link failure recovery.

HTTP is a successful Internet technology on top of which a lot of the web resides. However, limitations with its current specification have encouraged some to look for the next generation of HTTP. In SPDY, Google has come up with such a proposal that has growing community acceptance, especially after being adopted by the IETF HTTPbis-WG as the basis for HTTP/2.0. SPDY has the potential to greatly improve web experience with little deployment overhead, but we still lack an understanding of its true potential in different environments. This paper offers a comprehensive evaluation of SPDY’s performance using extensive experiments. We identify the impact of network characteristics and website infrastructure on SPDY’s potential page loading benefits, finding that these factors are decisive for an optimal SPDY deployment strategy. Through exploring such key aspects that affect SPDY, and accordingly HTTP/2.0, we feed into the wider debate regarding the impact of future protocols.

Laminated composite sandwich panels have a layered structure, where individual layers have randomly varying stiffness and strength properties. The presence of multiple failure modes and load redistribution following partial failures are the reason for laminated composites to exhibit system behavior from reliability point of view. The present paper discusses the specifics of system reliability behavior of laminated composite sandwich panels, and solves an example system reliability problem for a glass fiber-reinforced composite sandwich structure subjected to in-plane compression.

Laminated composite sandwich panels have a layered structure, where individual layers have randomly varying stiffness and strength properties. The presence of multiple failure modes and load redistribution following partial failures are the reason for laminated composites to exhibit system behavior from reliability point of view. The present paper discusses the specifics of system reliability behavior of laminated composite sandwich panels, and solves an example system reliability problem for a glass fiber-reinforced composite sandwich structure subjected to in-plane compression.

Laminated composite sandwich panels have a layered structure, where individual layers have randomly varying stiffness and strength properties. The presence of multiple failure modes and load redistribution following partial failures are the reason for laminated composites to exhibit system behavior from reliability point of view. The present paper discusses the specifics of system reliability behavior of laminated composite sandwich panels, and solves an example system reliability problem for a glass fiber-reinforced composite sandwich structure subjected to in-plane compression.

The integration of diverse but complementary cellular and wireless technologies in the next generation of wireless communication systems requires the design of intelligent vertical handoff decision algorithms to enable mobile users to seamlessly switch network access and experience uninterrupted service continuity anywhere and anytime. This paper provides an adaptive multiple attribute vertical handoff decision algorithm that enables wireless access network selection at a mobile terminal using fuzzy logic concepts and a genetic algorithm. A performance study using the integration of wireless wide area networks (WWANs) and wireless metropolitan area networks (WMANs) as an example shows that our proposed vertical handoff decision algorithm is able to determine when a handoff is required, and selects the best access network that is optimized to network conditions, quality of service requirements, mobile terminal conditions, user preferences, and service cost.

The integration of diverse but complementary cellular and wireless technologies in the next generation of wireless communication systems requires the design of intelligent vertical handoff decision algorithms to enable mobile users to seamlessly switch network access and experience uninterrupted service continuity anywhere and anytime. This paper provides an adaptive multiple attribute vertical handoff decision algorithm that enables wireless access network selection at a mobile terminal using fuzzy logic concepts and a genetic algorithm. A performance study using the integration of wireless wide area networks (WWANs) and wireless metropolitan area networks (WMANs) as an example shows that our proposed vertical handoff decision algorithm is able to determine when a handoff is required, and selects the best access network that is optimized to network conditions, quality of service requirements, mobile terminal conditions, user preferences, and service cost.

We explore a new stage in the evolution of digital trust, trusting strangers with your funds. We address the trust issues when giving money to others and relying on them to forward it. For fraud identification, we leverage our deployed blockchain which gradually builds trust between interacting strangers. Our blockchain fabric, called TrustChain, records interactions between entities in a scalable manner. This work represents a small step towards a generic infrastructure for trust, moving beyond proven single vendor platforms like eBay, Uber and Airbnb. Expanding upon established trust relations, we designed, implemented and evaluated an overlay network: Internet-of-Money. Internet-of-Money routes money to different banks through individuals, so-called money routers. This removes the need for central banks, to handle a payment. Our network reduces the duration of traditional inter-bank payments from up to a day and even a few days during weekends, to mere seconds. Internet-of-Money is fully decentralized, scalable and privacy-preserving. With real-world experimentations, we prove that Internet-of-Money enables fast money forwarding. We show that the overlay network is capable of discovering a majority of available money routers within a minute. Finally, we demonstrate how profit of cheating routers is limited and that misbehaviour is punished., Green Open Access added to TU Delft Institutional Repository ‘You share, we take care!’ – Taverne project https://www.openaccess.nl/en/you-share-we-take-care Otherwise as indicated in the copyright section: the publisher is the copyright holder of this work and the author uses the Dutch legislation to make this work public., Distributed Systems

We explore a new stage in the evolution of digital trust, trusting strangers with your funds. We address the trust issues when giving money to others and relying on them to forward it. For fraud identification, we leverage our deployed blockchain which gradually builds trust between interacting strangers. Our blockchain fabric, called TrustChain, records interactions between entities in a scalable manner. This work represents a small step towards a generic infrastructure for trust, moving beyond proven single vendor platforms like eBay, Uber and Airbnb. Expanding upon established trust relations, we designed, implemented and evaluated an overlay network: Internet-of-Money. Internet-of-Money routes money to different banks through individuals, so-called money routers. This removes the need for central banks, to handle a payment. Our network reduces the duration of traditional inter-bank payments from up to a day and even a few days during weekends, to mere seconds. Internet-of-Money is fully decentralized, scalable and privacy-preserving. With real-world experimentations, we prove that Internet-of-Money enables fast money forwarding. We show that the overlay network is capable of discovering a majority of available money routers within a minute. Finally, we demonstrate how profit of cheating routers is limited and that misbehaviour is punished., Green Open Access added to TU Delft Institutional Repository ‘You share, we take care!’ – Taverne project https://www.openaccess.nl/en/you-share-we-take-care Otherwise as indicated in the copyright section: the publisher is the copyright holder of this work and the author uses the Dutch legislation to make this work public., Distributed Systems

IP address counts are typically used as a surrogate metric for the number of hosts in a network, as in the case of ISP rankings based on botnet infected addresses. However, due to effects of dynamic IP address allocation, such counts tend to overestimate the number of hosts, sometimes by an order of magnitude. In the literature, the rate at which hosts change IP addresses is referred to as DHCP churn. Churn rates vary significantly within and among ISP networks, and such variation poses a challenge to any research that relies upon IP addresses as a metric. We present the first attempt towards estimating ISP and Internet-wide DHCP churn rates, in order to better understand the relation between IP addresses and hosts, as well as allow us to correct data relying on IP addresses as a surrogate metric. We propose an scalable active measurement methodology and then validate it using ground truth data from a medium-sized ISP. Next, we build a statistical model to estimate DHCP churn rates and validate against the ground truth data of the same ISP, estimating correctly 72.3% of DHCP churn rates. Finally, we apply our measurement methodology to four major ISPs, triangulate the results to another Internet census, and discuss the next steps to more precisely estimate DHCP churn rates., Multi Actor Systems, Technology, Policy and Management

We propose Spectrum Aware Virtual Coordinate (SAViC) for multi hop cognitive radio network (CRN) to facilitate geographic routing. The proposed virtual coordinates (VC) of any two secondary users reflect both geographic distance and opportunistic spectrum availability between them. As a result, geographic routing is able to detour the area affected by licensed users or cut through the area with more available spectrum. According to different spectrum occupation patterns of primary user, two versions of SAViC are designed based on the channel utility and primary user’s sojourning time respectively. Simulation shows the proposed virtual coordinate facilitates geographic routing to achieve high success rate of path construction. When duty cycle on the licensed channel is heterogeneous in the network, channel utility based virtual coordinate supports geographic routing to outperform a state-of-the-art geographic routing protocol by 40% on packet delivery ratio. When the channel utility is identical on each secondary node, and the sojourning time of primary users for secondary users are different from each other, SAViC based on primary user’s sojourning time achieves significantly shorter delay than other virtual coordinates., Qc 20150618

We propose Spectrum Aware Virtual Coordinate (SAViC) for multi hop cognitive radio network (CRN) to facilitate geographic routing. The proposed virtual coordinates (VC) of any two secondary users reflect both geographic distance and opportunistic spectrum availability between them. As a result, geographic routing is able to detour the area affected by licensed users or cut through the area with more available spectrum. According to different spectrum occupation patterns of primary user, two versions of SAViC are designed based on the channel utility and primary user’s sojourning time respectively. Simulation shows the proposed virtual coordinate facilitates geographic routing to achieve high success rate of path construction. When duty cycle on the licensed channel is heterogeneous in the network, channel utility based virtual coordinate supports geographic routing to outperform a state-of-the-art geographic routing protocol by 40% on packet delivery ratio. When the channel utility is identical on each secondary node, and the sojourning time of primary users for secondary users are different from each other, SAViC based on primary user’s sojourning time achieves significantly shorter delay than other virtual coordinates., Qc 20150618

We propose Spectrum Aware Virtual Coordinate (SAViC) for multi hop cognitive radio network (CRN) to facilitate geographic routing. The proposed virtual coordinates (VC) of any two secondary users reflect both geographic distance and opportunistic spectrum availability between them. As a result, geographic routing is able to detour the area affected by licensed users or cut through the area with more available spectrum. According to different spectrum occupation patterns of primary user, two versions of SAViC are designed based on the channel utility and primary user’s sojourning time respectively. Simulation shows the proposed virtual coordinate facilitates geographic routing to achieve high success rate of path construction. When duty cycle on the licensed channel is heterogeneous in the network, channel utility based virtual coordinate supports geographic routing to outperform a state-of-the-art geographic routing protocol by 40% on packet delivery ratio. When the channel utility is identical on each secondary node, and the sojourning time of primary users for secondary users are different from each other, SAViC based on primary user’s sojourning time achieves significantly shorter delay than other virtual coordinates., Qc 20150618

We propose Spectrum Aware Virtual Coordinate (SAViC) for multi hop cognitive radio network (CRN) to facilitate geographic routing. The proposed virtual coordinates (VC) of any two secondary users reflect both geographic distance and opportunistic spectrum availability between them. As a result, geographic routing is able to detour the area affected by licensed users or cut through the area with more available spectrum. According to different spectrum occupation patterns of primary user, two versions of SAViC are designed based on the channel utility and primary user’s sojourning time respectively. Simulation shows the proposed virtual coordinate facilitates geographic routing to achieve high success rate of path construction. When duty cycle on the licensed channel is heterogeneous in the network, channel utility based virtual coordinate supports geographic routing to outperform a state-of-the-art geographic routing protocol by 40% on packet delivery ratio. When the channel utility is identical on each secondary node, and the sojourning time of primary users for secondary users are different from each other, SAViC based on primary user’s sojourning time achieves significantly shorter delay than other virtual coordinates., Qc 20150618

We propose Spectrum Aware Virtual Coordinate (SAViC) for multi hop cognitive radio network (CRN) to facilitate geographic routing. The proposed virtual coordinates (VC) of any two secondary users reflect both geographic distance and opportunistic spectrum availability between them. As a result, geographic routing is able to detour the area affected by licensed users or cut through the area with more available spectrum. According to different spectrum occupation patterns of primary user, two versions of SAViC are designed based on the channel utility and primary user’s sojourning time respectively. Simulation shows the proposed virtual coordinate facilitates geographic routing to achieve high success rate of path construction. When duty cycle on the licensed channel is heterogeneous in the network, channel utility based virtual coordinate supports geographic routing to outperform a state-of-the-art geographic routing protocol by 40% on packet delivery ratio. When the channel utility is identical on each secondary node, and the sojourning time of primary users for secondary users are different from each other, SAViC based on primary user’s sojourning time achieves significantly shorter delay than other virtual coordinates., Qc 20150618

IP address counts are typically used as a surrogate metric for the number of hosts in a network, as in the case of ISP rankings based on botnet infected addresses. However, due to effects of dynamic IP address allocation, such counts tend to overestimate the number of hosts, sometimes by an order of magnitude. In the literature, the rate at which hosts change IP addresses is referred to as DHCP churn. Churn rates vary significantly within and among ISP networks, and such variation poses a challenge to any research that relies upon IP addresses as a metric. We present the first attempt towards estimating ISP and Internet-wide DHCP churn rates, in order to better understand the relation between IP addresses and hosts, as well as allow us to correct data relying on IP addresses as a surrogate metric. We propose an scalable active measurement methodology and then validate it using ground truth data from a medium-sized ISP. Next, we build a statistical model to estimate DHCP churn rates and validate against the ground truth data of the same ISP, estimating correctly 72.3% of DHCP churn rates. Finally, we apply our measurement methodology to four major ISPs, triangulate the results to another Internet census, and discuss the next steps to more precisely estimate DHCP churn rates., Multi Actor Systems, Technology, Policy and Management

Data traffic in cellular networks increased tremendously over the past few years and this growth is predicted to continue over the next few years. Due to differences in access technology and user behavior, the characteristics of cellular traffic can differ from existing results for wireline traffic. In this study we focus on understanding the flow rates and on the relationship between the rates and other flow properties by analyzing packet level traces collected in a large cellular network. To understand the limiting factors of the flow rates, we further analyze the underlying causes behind the observed rates, e.g.,network congestion, access link or end host configuration. Our study extends other related work by conducting the analysis from a unique dimension, the comparison with traffic in wired networks, to reveal the unique properties of cellular traffic. We find that they differ in variability and in the dominant rate limiting factors.

We model and evaluate the performance of a distributed key-value storage system that is part of the Spotify backend. Spotify is an on-demand music streaming service, offering low-latency access to a library of over 16 million tracks and serving over 10 million users currently. We first present a simplified model of the Spotify storage architecture, in order to make its analysis feasible. We then introduce an analytical model for the distribution of the response time, a key metric in the Spotify service. We parameterize and validate the model using measurements from two different testbed configurations and from the operational Spotify infrastructure. We find that the model is accurate - measurements are within 11% of predictions - within the range of normal load patterns. We apply the model to what-if scenarios that are essential to capacity planning and robustness engineering. The main difference between our work and related research in storage system performance is that our model provides distributions of key system metrics, while related research generally gives only expectations, which is not sufficient in our case., QC 20130129

We report on design, implementation and evaluation of a resource management system that builds upon OpenStack, an open-source cloud platform for private and public clouds. Our implementation supports an Infrastructure-as-a-Service (IaaS) cloud and currently provides allocation for computational resources in support of both interactive and computationally intensive applications. The design supports an extensible set of management objectives between which the system can switch at runtime. We demonstrate through examples how management objectives related to load-balancing and energy efficiency can be mapped onto the controllers of the resource allocation subsystem, which attempts to achieve an activated management objective at all times. The design is extensible in the sense that additional objectives can be introduced by providing instantiations for generic functions in the controllers. Our implementation monitors the fulfillment of the relevant management metrics in real time. Testbed evaluation demonstrates the effectiveness of our approach in a dynamic environment. It further illustrates the trade-off between closely meeting a specific management objective and the associated cost of VM live-migration., QC 20130129

We model and evaluate the performance of a distributed key-value storage system that is part of the Spotify backend. Spotify is an on-demand music streaming service, offering low-latency access to a library of over 16 million tracks and serving over 10 million users currently. We first present a simplified model of the Spotify storage architecture, in order to make its analysis feasible. We then introduce an analytical model for the distribution of the response time, a key metric in the Spotify service. We parameterize and validate the model using measurements from two different testbed configurations and from the operational Spotify infrastructure. We find that the model is accurate - measurements are within 11% of predictions - within the range of normal load patterns. We apply the model to what-if scenarios that are essential to capacity planning and robustness engineering. The main difference between our work and related research in storage system performance is that our model provides distributions of key system metrics, while related research generally gives only expectations, which is not sufficient in our case., QC 20130129

We model and evaluate the performance of a distributed key-value storage system that is part of the Spotify backend. Spotify is an on-demand music streaming service, offering low-latency access to a library of over 16 million tracks and serving over 10 million users currently. We first present a simplified model of the Spotify storage architecture, in order to make its analysis feasible. We then introduce an analytical model for the distribution of the response time, a key metric in the Spotify service. We parameterize and validate the model using measurements from two different testbed configurations and from the operational Spotify infrastructure. We find that the model is accurate - measurements are within 11% of predictions - within the range of normal load patterns. We apply the model to what-if scenarios that are essential to capacity planning and robustness engineering. The main difference between our work and related research in storage system performance is that our model provides distributions of key system metrics, while related research generally gives only expectations, which is not sufficient in our case., QC 20130129

We model and evaluate the performance of a distributed key-value storage system that is part of the Spotify backend. Spotify is an on-demand music streaming service, offering low-latency access to a library of over 16 million tracks and serving over 10 million users currently. We first present a simplified model of the Spotify storage architecture, in order to make its analysis feasible. We then introduce an analytical model for the distribution of the response time, a key metric in the Spotify service. We parameterize and validate the model using measurements from two different testbed configurations and from the operational Spotify infrastructure. We find that the model is accurate - measurements are within 11% of predictions - within the range of normal load patterns. We apply the model to what-if scenarios that are essential to capacity planning and robustness engineering. The main difference between our work and related research in storage system performance is that our model provides distributions of key system metrics, while related research generally gives only expectations, which is not sufficient in our case., QC 20130129

We model and evaluate the performance of a distributed key-value storage system that is part of the Spotify backend. Spotify is an on-demand music streaming service, offering low-latency access to a library of over 16 million tracks and serving over 10 million users currently. We first present a simplified model of the Spotify storage architecture, in order to make its analysis feasible. We then introduce an analytical model for the distribution of the response time, a key metric in the Spotify service. We parameterize and validate the model using measurements from two different testbed configurations and from the operational Spotify infrastructure. We find that the model is accurate - measurements are within 11% of predictions - within the range of normal load patterns. We apply the model to what-if scenarios that are essential to capacity planning and robustness engineering. The main difference between our work and related research in storage system performance is that our model provides distributions of key system metrics, while related research generally gives only expectations, which is not sufficient in our case., QC 20130129

We address the problem of resource allocation in a large-scale cloud environment, which we formalize as that of dynamically optimizing a cloud configuration for green computing objectives under CPU and memory constraints. We propose a generic gossip protocol for resource allocation, which can be instantiated for specific objectives. We develop an instantiation of this generic protocol which aims at minimizing power consumption through server consolidation, while satisfying a changing load pattern. This protocol, called GRMP-Q, provides an efficient heuristic solution that performs well in most cases—in special cases it is optimal. Under overload, the protocol gives a fair allocation of CPU resources to clients. Simulation results suggest that key performance metrics do not change with increasing system size, making the resource allocation process scalable to well above 100,000 servers. Generally, the effectiveness of the protocol in achieving its objective increases with increasing memory capacity in the servers., QC 20110818

We address the problem of resource allocation in a large-scale cloud environment, which we formalize as that of dynamically optimizing a cloud configuration for green computing objectives under CPU and memory constraints. We propose a generic gossip protocol for resource allocation, which can be instantiated for specific objectives. We develop an instantiation of this generic protocol which aims at minimizing power consumption through server consolidation, while satisfying a changing load pattern. This protocol, called GRMP-Q, provides an efficient heuristic solution that performs well in most cases—in special cases it is optimal. Under overload, the protocol gives a fair allocation of CPU resources to clients. Simulation results suggest that key performance metrics do not change with increasing system size, making the resource allocation process scalable to well above 100,000 servers. Generally, the effectiveness of the protocol in achieving its objective increases with increasing memory capacity in the servers., QC 20110818

We address the problem of resource allocation in a large-scale cloud environment, which we formalize as that of dynamically optimizing a cloud configuration for green computing objectives under CPU and memory constraints. We propose a generic gossip protocol for resource allocation, which can be instantiated for specific objectives. We develop an instantiation of this generic protocol which aims at minimizing power consumption through server consolidation, while satisfying a changing load pattern. This protocol, called GRMP-Q, provides an efficient heuristic solution that performs well in most cases—in special cases it is optimal. Under overload, the protocol gives a fair allocation of CPU resources to clients. Simulation results suggest that key performance metrics do not change with increasing system size, making the resource allocation process scalable to well above 100,000 servers. Generally, the effectiveness of the protocol in achieving its objective increases with increasing memory capacity in the servers., QC 20110818

We address the problem of resource allocation in a large-scale cloud environment, which we formalize as that of dynamically optimizing a cloud configuration for green computing objectives under CPU and memory constraints. We propose a generic gossip protocol for resource allocation, which can be instantiated for specific objectives. We develop an instantiation of this generic protocol which aims at minimizing power consumption through server consolidation, while satisfying a changing load pattern. This protocol, called GRMP-Q, provides an efficient heuristic solution that performs well in most cases—in special cases it is optimal. Under overload, the protocol gives a fair allocation of CPU resources to clients. Simulation results suggest that key performance metrics do not change with increasing system size, making the resource allocation process scalable to well above 100,000 servers. Generally, the effectiveness of the protocol in achieving its objective increases with increasing memory capacity in the servers., QC 20110818

We address the problem of resource allocation in a large-scale cloud environment, which we formalize as that of dynamically optimizing a cloud configuration for green computing objectives under CPU and memory constraints. We propose a generic gossip protocol for resource allocation, which can be instantiated for specific objectives. We develop an instantiation of this generic protocol which aims at minimizing power consumption through server consolidation, while satisfying a changing load pattern. This protocol, called GRMP-Q, provides an efficient heuristic solution that performs well in most cases—in special cases it is optimal. Under overload, the protocol gives a fair allocation of CPU resources to clients. Simulation results suggest that key performance metrics do not change with increasing system size, making the resource allocation process scalable to well above 100,000 servers. Generally, the effectiveness of the protocol in achieving its objective increases with increasing memory capacity in the servers., QC 20110818

QCR 20170322