101,992 research outputs found
Distribution-efficient networks
No full text\u3cp\u3ePublishing resources on a virtual network is a way to realize efficient data-distribution mechanisms. To this extent, each node needs to discover the other nodes, create neighborhoods and advertize its own resources. This chapter presents different techniques for making resources discoverable, considering two approaches dubbed as unstructured networks and structured networks. We discuss properties of different protocols in terms of signaling overheads and distribution efficiency.\u3c/p\u3
Six problems for the service provider
No full text\u3cp\u3eThe Internet is far from being perfect, and thus we are bound to see many remarkable changes in the near future. In examining what the Net can and cannot do today, what are the top six reasons to upgrade it? The Net is not sufficiently ubiquitous, reactive, proactive, information-driven, distribution-efficient and searchable. This chapter introduces these widely recognized issues, paving the way for the solutions presented in Chap. 4 and thereafter.\u3c/p\u3
The network, as we know it
No full text\u3cp\u3eIf you aren't a network guru and have no interest whatsoever in becoming one, though still wonder how the 'Net' works, this chapter will provide you with a number of precious answers. Together, we'll revisit the hectic journey of a data packet from the time it's conceived by an Internet server, until its destiny is accomplished in your computer. As with messengers, packets carry valuable information. Their purpose in life is simple: to find the best path to their addressee. However, in a network entangled with billions of links, how does your packet find its way through? How can streams of packets be delivered on time? You will appreciate the mechanisms that keep the network connected and stable. To those who are not network specialists, this chapter will provide all the elements required to tackle the more advanced networking concepts introduced in the rest of book. You will read about routers, packet switching, data buffering, message forwarding, the wonders of Dijkstra's algorithm and the tricks used to keep mobile terminals connected.\u3c/p\u3
Content-aware networks
No full text\u3cp\u3ePacket switching networks provide rudimentary means to move units of raw data around. The Net can transport, though it is unable to manipulate high-level content, video or audio sessions. Imagine what we could achieve with a network that is redesigned around what is the most precious thing in today's digital ecosystem: the content. This chapter introduces content-aware networks, ones that can re-route packets based on the content usage patterns and requirements. We look at peer-to-peer networks as a practical example in order to better understand how to build content-aware networks on top of ordinary packet switching networks.\u3c/p\u3
Discovering virtual resources
No full text\u3cp\u3eVirtual resources are considerably more numerous than the physical devices which host them. There are many more files than servers; more videos than people. Thus, the task of discovering relevant resources is certainly a daunting one. The search engines help to retrieve data from servers. Yet, discovering resources stored in a virtual environment requires deep searching techniques which must be able to explore not only the multitude of web servers, but also the ordinary computers and terminals. This chapter introduces different discovery techniques used in structured and unstructured P2P systems.\u3c/p\u3
Proactive networks
No full text\u3cp\u3eWhat is the secret of a fast-responding network? The ideal network will anticipate the communication needs of all nodes, building the necessary paths proactively. Unfortunately, this level of intelligence is not possible today. A brute-force approach, whereby the network continuously maintains all possible paths among all nodes, is also not a viable proposition because networks are far too vast and dynamic. This chapter explores strategies to reduce the impact of signaling in proactive routing. Through this exercise, we find that a fast network is one that can adaptively switch between proactive and reactive modes.\u3c/p\u3
Spontaneous networks
No full text\u3cp\u3eHow can we increase the capillarity of the Net without facing the daunting issues that come with large-scale infrastructures? Can we embed all necessary protocols into our terminals and then use the terminals themselves to relay packets? This chapter develops the vision of ubiquitous connectivity, pinpointing foundations and problems. Networks made without any dedicated hardware are possible, but require new protocols. Here, we discover how to build spontaneous, ad hoc networks starting from extremely simple mechanisms.\u3c/p\u3
A peek at the future internet
No full text\u3cp\u3eThe Internet connectivity machine is the generative engine of our modern digital society. It has been the launching pad of the Web (now the Web 2.0), truly the largest and most versatile information system ever built. While the Web phenomenon relentlessly continues, scientists worldwide are now living the dream of yet a more generative next-generation network. This chapter explores some prominent research directions, discussing the Internet of Things, context-aware networks, small world networks, scale-free networks, autonomic networks, dependable networks, the privacy vs. security dichotomy and the two facets of energy-efficient networks.\u3c/p\u3
Reactive networks
No full text\u3cp\u3eNetworks strive to keep all of their nodes connected. However, is this really necessary? Do we actually need to continuously maintain routes from and to any possible destination? This chapter looks at networks that can discern between active and non-active paths. The idea is to care for the nodes that are actively intercommunicating, leaving the rest of the network in standby mode. In this chapter, we will explore on-demand routing, one of the key ingredients that can make networks more reactive on a larger scale.\u3c/p\u3
On the Way to the Pervasive Web
No full textWeb applications bring about extraordinary breakthroughs regarding our digital ecosystem. Pretty much anything with a chip and a radio interface can connect to the Web. However, many advocate a complete overhaul of the Internet as the only means to sustain innovation and productivity. Nobody knows what the next-generation of the Internet will look like; though important clues are visible as years of research have already generated phenomenal ideas. Together, we'll bring a range of network mechanisms "out of the lab" that can make the Net more proactive, reactive, robust and, ultimately, more pervasive than it is today. Our journey starts by scrutinizing the inexorable transformation of Web Applications in order to unveil the intrinsic limitations of the Internet.</p
- …
