Jon Crowcroft

Bio: Jon Crowcroft is an academic researcher from University of Cambridge. The author has contributed to research in topics: The Internet & Multicast. The author has an hindex of 87, co-authored 672 publications receiving 38848 citations. Previous affiliations of Jon Crowcroft include Memorial University of Newfoundland & Information Technology University.

Network Time Protocol (NTP) over the OSI Remote Operations Service

Abstract: This memo suggests an Experimental Protocol for the OSI and Internet communities. Hosts in either community, and in particular those on both are encouraged to experiment with this mechanism.

LEDBAT performance in sub-packet regimes

Abstract: Wireless Community Networks in the developing world satisfy the basic needs of remote users to information access. However, community networks in developing regions usually rely on low-bandwidth backhaul links that are shared amongst a large user base, driving these links to sub-packet regimes where the per-flow throughput is less than one packet per RTT. TCP performance significantly degrades in such conditions, resulting in severe unfairness and high packet loss rates. In this paper, we investigate the performance of scavenger transport methods, namely LEDBAT and its fair modification fLEDBAT, in the sub-packet regime of shared backhaul links in developing regions. Our intention is to explore the feasibility of using such scavenger transport methods for uploading content over bandwidth constrained backhauls. Our findings show that LEDBAT achieves higher link efficiency and fairness compared to TCP in a variety of sub-packet regime scenarios. When TCP and LEDBAT flows share the same link in the sub-packet regime, LEDBAT flows are more aggressive, consuming more resources than TCP. Therefore, we conclude that a more conservative strategy after consecutive timeouts and shared bottleneck detection mechanisms need to be incorporated into the core LEDBAT algorithm, in order to correctly adjust its congestion window in the sub-packet regime.

The Case for Context-Aware Resources Management in Mobile Operating Systems

Abstract: Efficient management of mobile resources from an energy perspective in modern smart-phones is paramount nowadays. Today’s mobile phones are equipped with a wide range of sensing, computational, storage and communication resources. The diverse range of sensors such as microphones, cameras, accelerometers, gyroscopes, GPS, digital compass and proximity sensors allow mobile apps to be context-aware whereas the ability to have connectivity almost everywhere has bootstrapped the birth of rich and interactive mobile applications and the integration of cloud services. However, the intense use of those resources can easily be translated into power-hungry applications. The way users interact with their mobile handsets and the availability of mobile resources is context dependent. Consequently, understanding how users interact with their applications and integrating context-aware resources management techniques in the core features of a mobile operating system can provide benefits such as energy savings and usability. This chapter describes how context drives the way users interact with their handsets and how it determines the availability and state of hardware resources in order to explain different context-aware resources management systems and the different attempts to incorporate this feature in mobile operating systems.

I and i

Abstract: There is, I think, something ethereal about i —the square root of minus one. I remember first hearing about it at school. It seemed an odd beast at that time—an intruder hovering on the edge of reality. Usually familiarity dulls this sense of the bizarre, but in the case of i it was the reverse: over the years the sense of its surreal nature intensified. It seemed that it was impossible to write mathematics that described the real world in …

Community detection in graphs

Abstract: The modern science of networks has brought significant advances to our understanding of complex systems. One of the most relevant features of graphs representing real systems is community structure, or clustering, i. e. the organization of vertices in clusters, with many edges joining vertices of the same cluster and comparatively few edges joining vertices of different clusters. Such clusters, or communities, can be considered as fairly independent compartments of a graph, playing a similar role like, e. g., the tissues or the organs in the human body. Detecting communities is of great importance in sociology, biology and computer science, disciplines where systems are often represented as graphs. This problem is very hard and not yet satisfactorily solved, despite the huge effort of a large interdisciplinary community of scientists working on it over the past few years. We will attempt a thorough exposition of the topic, from the definition of the main elements of the problem, to the presentation of most methods developed, with a special focus on techniques designed by statistical physicists, from the discussion of crucial issues like the significance of clustering and how methods should be tested and compared against each other, to the description of applications to real networks.