Interconnect Demand Research Articles

Characterizing and Specifying Connectivity Strategies in and Across Internet Exchanges

Internet eXchanges (IXes) are developed to localize traffic, reduce connectivity costs, and reduce latency. Historically, transit savings were sufficiently substantive to justify investment in IX interconnection strategies, overshadowing other benefits enjoyed by IX participants. Recent work argues these basic IX effects facilitate regional economic development. In developed regions, as transit prices converge with comparable per volume price on Internet exchanges, a new question emerges: given this convergence, what additional value and benefits does the IX provide existing and would-be participants? Governance and organization studies of IXes provide some of the nuance necessary to answer this question. Based on interviews with IX managers and participants, there is value in a more fluid market of interconnection options that facilitate investment deferral and greater control over the repurposing of interconnection resources. Interconnection is presented in terms of platforms and how they implement interconnection options, highlighting how the mechanics of these implementation frame network actors’ decisions regarding how to maximize access to and investments in interconnection markets. Semi-structured interviews of a wide range of network operators and IX operators over the last year and a half imply decision heuristics for optimizing value and network resilience: selectively increasing redundancy, increasing unique interconnection partners, and reducing switching costs, are among the most common criteria. This work distills decision heuristics identified thus far into a partially parameterized model of interconnection decision making across platforms that serve as markets for interconnection options. Unpacking decision heuristics contributes to more precisely explaining the mix of interconnection options available to actors that both derive value from network- and application-level services but who also influence infrastructure investment and strategy decisions through the options they select. This work demonstrates how IX-platforms facilitate greater access to interconnection options and defer bargaining and measurement costs necessary for more specific asset investments. IX-enabled interconnection contributes to the “flattening” of the Internet by providing an investment path to more sophisticated bundles of (flatter) bilateral relations rather than participating solely in the transit hierarchy. The model developed here provides a clear formalism for comparing these. Two classes of parameters are missing: connectivity cost information and parameters representing actor-specific valuations of system properties such as redundancy or latency. A key contribution is to understand the options available by carefully specifying the parameter space based on empirical observation: what are the variables and how are they conceptually related? This work unpacks the mechanisms implementing two common types of interconnection option and concludes with working hypotheses that will be further refined and tested in ongoing work. The next stage of this work focuses on identifying appropriate operationalizations, data sources for cost, and strategies for comparing (validating) theoretical valuations based on empirical operationalizations of variables such as redundancy. A social science approach to indicator development is applied to frame heuristics as background concepts that are refined into specifications of incentives, relationships, and potential indicators. Variables specified here represent possible systematizations of background concepts elicited from interviews. Neither the specification nor the relationships are presented as being writ in stone — rather, they are a part of an ongoing project to elicit empirical trends in the interconnection industry. The variety of network actors that engage with IXes, their value propositions, and the means by which they connect to IX platforms provide insight into which interconnection options are appealing to which types of actors as well as common trade-offs. Qualitatively, this gives insight into which types of actors are willing to invest in which mixes of infrastructure types and topologies. By highlighting trade-off conditions in the demand for interconnection, this specification provides direction for further identification of interconnection costs and the constraints by those making investment decisions. The framework offers a number of contributions. Different interconnection modes are conceptualizes independently, but the framework subsequently highlights mutual dependence and benefits. Working hypotheses argue that different platforms structure and mediate interconnection option implementations (transport, colocation, IX), how they are interleaved with one another in practice, and offer a first pass at identifying and realizing critical paths through investment decisions as they grow and develop specific relationships (and asset investments) beyond simple transit. Building on the notion of a real option, flexibility in general purpose resources (IX-mediated interconnection) facilitates staged, dynamic investment decisions and learning effects.

Assessment of the interconnection demand of the river-lake system in the Lake Taihu basin

水系连通需求分析是河湖水系连通战略研究的重要内容.水系连通需求指人类社会经济发展对水资源调配、防洪排涝、改善生态环境等方面的需求及对河湖水系健康的维护,它源于水系能够提供各种水生态系统服务功能的生态学特性.通过分析水系连通需求、水生态系统服务功能和水系连通工程之间的内在关系,认为水系连通需求评价可转化为对水生态系统服务功能的评价.运用水当量评价方法,建立了河湖水系连通需求评价的方法体系,并以河网地区的典型代表——太湖为例,定量评价了太湖在水资源调配、调蓄洪水、水环境净化和维持生境等方面的连通需求.结果表明,太湖水系年均连通需求最大的是净化入湖废污水需水量,其次为水资源调配需水量,而太湖调蓄洪水的需求减少;湖水自身净化需水量较大,且为一次性需水.水环境净化的需求需要通过降低污染物入湖量,进行湖泊生态修复等主要措施以及引清水增加环境容量这一辅助连通措施共同完成.水系连通需求的定量评价可为水系连通战略及工程的规划设计提供理论基础.;River-lake system interconnection demand is one of the key elements among the research of river-lake system interconnection strategy.It reflects the demand of the social-economic development on water diversion,flood regulation,eco-environment improvement and the maintenance of the health of river-lake system.It is based on the inherent ecological properties that a drainage system can provide various water ecosystem services to a river-lake system.Since the river system,river system interconnection demand, the water ecosystem services,and river system interconnection engineering are closely interconnected,the methods which have been adequately studied for assessing the water ecosystem services can be applied to the assessment of river system interconnection demand.So methodologies for evaluation of river system interconnection demand are established by applying the water equivalent method.With a long historical practices of river system interconnection in Lake Taihu watershed,which is a typical and emergent floodplain river network area in eastern China,demanding on Lake Taihu river system interconnection is assessed in this study.The main demand includes water resource diversion,flood regulation,water purification,and habitat sustainment.Purification water demand per year of retained pollutants in the lake is 2.85×10¹⁰ m³/a,ranked the first place.Water resource diversion demand is 3.0×10⁹-4.2×10⁹ m³/a,ranked the second place,while the demand of flood regulation of the lake is decreased by 7.2×10⁸-10.7×10⁸ m³/a.Purification demand of the water from the lake is 8.14×10⁹ m³,and is consumed once only;the habitat maintenance water demand is 3.58×10⁹ m³.The results show that water environment improvement demand is tremendous.To ease the stress,countermeasures such as control of the pollutant entering into the lake,desilting and dredging the lake,and ecological restoration of the lake water need to be applied firstly,then the diversion of clean water as an assisting way.There are different corresponding interconnection ways for each demand,and it is necessary to confirm the ratio of each interconnection project.

Earth Jurisprudence and the Murray Darling: The Future of a River

This article critiques the draft Murray-Darling Basin Plan from the perspective of an emerging theory of law termed Earth Jurisprudence. While the Draft Plan purports to be concerned with ecological management, this paper argues that it actually legitimizes the continued exploitation of the Murray-Darling. Next, the paper considers an alternative approach to ecosystem management that is based on the scientific concept of ‘Ecological Integrity’. This approach views the Murray-Darling as an interconnected hole and demands that its ecological needs receive the highest priority.

A QoS-Enabled On-Die Interconnect Fabric for Kilo-Node Chips

To meet rapidly growing performance demands and energy constraints, future chips will likely feature thousands of on-die resources. Existing network-on-chip solutions weren't designed for scalability and will be unable to meet future interconnect demands. A hybrid network-on-chip architecture called Kilo-NoC co-optimizes topology, flow control, and quality of service to achieve significant gains in efficiency.

제주 HVDC#3 영향 분석

According to the 5th national plan for electric power demand and supply, the 3rd HVDC link will be installed between KEPCO and Jeju Island system and go into operation in June, 2016. However, if it adopt LCC HVDC, Jeju power system could be in voltage instable condition. In order to solve the problem, we propose the application of VSC HVDC to the 3rd HVDC project. In the past, VSC HVDC has many disadvantages such as, inefficiency in conversion, size limit, and expensiveness, compared to LCC HVDC. Recently, the new technologies and the demand for renewable energy interconnection are leading a large growth in VSC HVDC market. In this paper, we will introduce the technical and market trends of VSC HVDC and power system stability analysis results for Jeju power system having LCC and VSC HVDCs.

Modeling formalisms in Systems Biology.

Systems Biology has taken advantage of computational tools and high-throughput experimental data to model several biological processes. These include signaling, gene regulatory, and metabolic networks. However, most of these models are specific to each kind of network. Their interconnection demands a whole-cell modeling framework for a complete understanding of cellular systems. We describe the features required by an integrated framework for modeling, analyzing and simulating biological processes, and review several modeling formalisms that have been used in Systems Biology including Boolean networks, Bayesian networks, Petri nets, process algebras, constraint-based models, differential equations, rule-based models, interacting state machines, cellular automata, and agent-based models. We compare the features provided by different formalisms, and discuss recent approaches in the integration of these formalisms, as well as possible directions for the future.

An Overview of Molecular Computing Approaches (Part II)

Abstract This is the second part of an article on molecular electronics. The first part, published in the August 2004 issue of EDFA, discussed the development of molecular devices including nanowires, rectifiers, switches, and transistors. Here, the author describes nontraditional molecular computing architectures based on crosspoint arrays, randomized nanocells, and cellular automata. Challenges associated with interconnect demand, lithography alternatives, and defect tolerance are also discussed.

Development of a general purpose configurable architecture for smart-pixel research

The ever-increasing demand for communication bandwidth and system interconnectivity has been a motivating factor behind the integration of optoelectronics devices and conventional data processing circuitry. Based on the smart-pixel architectures first developed in the last decade, the architecture presented here monolithically integrates optical sensors with silicon CMOS-based circuitry to produce a generically programmable smart-pixel array. Two generations of the architecture are described and compared. We have proposed a reconfigurable photonic information-processing chip based on photonic VLSI device technology. Integrating detectors into a SIMD array removes the bottleneck associated with fetching slices of data. By fabricating the detectors along with logic circuits in a bulk CMOS process, the cost is minimized. The modular nature of the array organization facilitates replication of the configurable architecture for smart-pixel research (CASPR) concept into large arrays without a significant increase in design overhead. Thus, the CASPR architecture can provide the maximum flexibility associated with a reconfigurable smart-pixel array. Finally, we implement two design iterations of the CASPR architecture and show how the architecture might be used in a page-oriented optical data processing application.

Heterogeneous architecture models for interconnect-motivated system design

On-chip interconnect demand is becoming the dominant factor in modern processor performance and must be estimated early in the design process. This paper presents a set of heterogeneous architectural models that combines architecture description and Rent's rule-based wiring models. These architecture models allow flexible heterogeneous system specifications, enabling investigations of prospective designs in different technology scenarios. Comparisons against actual data demonstrate the models' effectiveness for architecture explorations with highly accurate estimations of local and global wiring demand, as well as chip area and cycle time. Simulation of two candidate system designs reveal trends in interconnect delay with increasing architectural complexity, and confirm the need for high computational locality and short global wires for future architectures.

LGA Connectors: An Automated Design Technique for a Shrinking Design Space

The ever-increasing demand for higher-density interconnection between a multi-chip module and the printed circuit board has resulted in the emergence of Land-Grid Array (LGA) connectors as an alternative to the traditional pin and socket area-array connectors. The design of high-density land-grid array connectors involves trade-off between conflicting performance requirements on the normal force, wipe, bulk resistance, contact resistance, stress, contact z-dimensional thickness, and z-compression. These stringent design requirements have significantly shrunk the space of viable designs and have necessitated automated search procedures for finding designs that satisfy the design requirements. In this paper, such an automated design procedure based on nonlinear optimization techniques is presented. The design procedure includes a general shape representation scheme based on B-spline curves and a set of programs for carrying out automated nonlinear elastic-plastic-contact finite element analysis (for a given shape) using a commercial finite element code. This automated analysis procedure is coupled with a nonlinear optimization code to carryout optimal design of LGA connectors. The design of LGA connectors is mathematically formulated as an optimization problem and nine different design cases (with representative dimensions and material) are solved to determine the influence of initial design and optimization problem formulation. It is shown that better solutions (with less stress) result if both the width and the thickness of the contacts are allowed to vary. In general, the choice of initial design strongly influences the optimal solution. A triply-curved symmetric contact shape is shown to produce the least stress of three possible common LGA shape designs. [S1043-7398(00)00403-5]

Performance of integrated interlan telephony service over an ATM network

The demand for a seamless interconnection between the end subnetworks using ATM technology is steadily arising in the customer premises. A great demand also exists in enterprise networks for obtaining interLAN telephony service integrated with asynchronous data services. We determine the end-to-end performance of internetworked integrated services ATM LANs by using a simulation based approach. We further demonstrate the feasibility of providing interLAN telephony service integrated with asynchronous data services in internetworked ATM LANs. This entails the implementation of a crucial technique, which we have developed in the interworking units so as to segregate the voice data from the asynchronous data. The end subnetworks considered in case of the integrated services homogenous ATM LANs are IEEE 802.5 LANs, whereas the end subnetworks considered in case of integrated services heterogenous ATM LANs are IEEE 802.5 and the standardised 802.9 (isoEthernet) LANs. A significant issue addressed in this case is the internetworking of the packet based IEEE 802.5 LAN to the 802.9 LAN.

Packaging alternatives to large silicon chips: tiled silicon on MCM and PWB substrates

Recent advances in area array chip bonding combined with the availability of high density substrates facilitate novel approaches to partitioning future systems. We examine one such new paradigm here: tiled silicon, in which system integration is achieved by tiling a set of chips together using area bonding on high density substrates rather than by pursuing single chip integration. We simulate the partitioning of large silicon/complementary metal-oxide-semiconductor (Si/CMOS) chips into tiled arrays of silicon chips, including in the analysis wiring lengths, electrical interconnect issues, I/O requirements, including drivers and electrostatic discharge (ESD) protection, wiring capacity, floorplans, wiring demand, escape, manufacturing yield, cost, and other electrical and thermal issues. Partitions are assumed to be interconnected via random logic, bus or memory type net topologies. Our results clearly show that it is possible to effectively tile silicon chips, when they are connected by reduced Rent exponent random logic, buses, or memory type net topologies. Systems with high interconnect demand, and thus little or no functional integration, cannot be tiled because of problems with larger chip real estate for drivers for off-chip lines and off-chip wiring capacity.

Modernization of a Burner Control System in a Thermoelectric Power Plant using Logic Programmable Controllers

The increasing need of maintenance, the lack of spare parts and the extra requirements of a very complex interconnected electrical system demands a new strategy which can support the reliability of a particular power plant. The present maturity of microprocessor technology makes feasible the development of a Burner Management System based on programmable controllers (PLC's). This paper reviews the retrofit of such system utilizing local PLC's developed by the Institute of Electrical Research (IIE). The problem is described in detail including its main constraints. The general characteristics (hardware and software) of the new system are also mentioned.

Fast Transient Security Assessment

The growth of large interconnected power systems demands a high degree of security for normal operation. This requirement emerges from the fact that the possible disturbances in large power systems could have catastrophic results. Therefore it is the duty of power system control center to retain the operating point within secure boundaries. In order to control the security of system, one has to monitor and analyze the level of system security and enhance it in case it does not fulfill the system security requirement. The pattern recognition technique which is proposed in this paper is a method suitable for fast, on-line security assessment. Most of the literature in this area concerns the use of steady state variables for identification pruposes; the use of transient measurements, however, has been found to be better suited for the fast classification of operating points with minimal error. One of the salient conclusions of this paper is that the pattern recognition technique is a serious candidate for on-line applications. This favorable conclusion is based on the inclusion of transient state variables in the feature vector. The main objective of the pattern recognition method in transient security assessment is to reduce the computational requirement to a minimum. This is done at the expense of elaborate off-line computations. The classical methodology of pattern recognition consists of defining a pattern vector x whose components consist of all significant variables of the system.

Internetworking analysis

In recent years the demand for network interconnection has increased. Since the size of a computer system can be increased at reasonable cost only within certain limits the need for internetworking arises. The aim of the paper is to analyse the problems of internetworking by discussing both protocol and interface aspects in datagram and virtual-call environments and to suggest reliable and efficient methods of interconnection. The use of microcomputers in low-cost modular and flexible gateways is also discussed.