Jan-Erik Mångs

Resource allocation for distributed cloud: concepts and research challenges

In a cloud computing environment, dynamic resource allocation and reallocation are keys for accommodating unpredictable demands and, ultimately, contribute to investment return. This article discusses this process in the context of distributed clouds, which are seen as systems where application developers can selectively lease geographically distributed resources. This article highlights and categorizes the main challenges inherent to the resource allocation process particular to distributed clouds, offering a stepwise view of this process that covers the initial modeling phase through to the optimization phase.

Real-Time Measurement of End-to-End Available Bandwidth using Kalman Filtering

This paper presents a new method, BART (bandwidth available in real-time), for estimating the end-to-end available bandwidth over a network path. It estimates bandwidth quasi-continuously, in real-time. The method has also been implemented as a tool. It relies on self-induced congestion, and repeatedly samples the available bandwidth of the network path with sequences of probe packet pairs, sent at randomized rates. BART requires little computation in each iteration, is lightweight with respect to memory requirements, and adds only a small amount of probe traffic. The BART method uses Kalman filtering, which enables real-time estimation (a.k.a. tracking). It maintains a current estimate, which is incrementally improved with each new measurement of the inter-packet time separations in a sequence of probe packet pairs. The measurement model has a strong non-linearity, and would not at first sight be considered suitable for Kalman filtering, but we show how this non-linearity can be handled. BART may be tuned according to the specific needs of the measurement application, such as agility vs. stability of the estimate. We have tested an implementation of BART in a physical test network with carefully controlled cross traffic, with good accuracy and agreement. Test measurements have also been performed over the Internet. We compare the performance of BART with that of pathChirp, a state-of-the-art tool for measuring end-to-end available bandwidth in real-time

Open Source Cloud Computing Platforms

With the popularization of cloud computing, several enterprises and open-source communities have developed their own cloud solutions. A number of factors weigh on user selection, as each one has peculiar characteristics and may target different usage scenarios. Considering such challenge, this paper focuses on giving the reader an understanding of some major existing open cloud computing solutions – XCP, Eucalyptus and Open Nebula. Hopefully, a deep comparison of such solutions can leverage the cloud computing research area providing a good starting point to research groups and interested readers.

Fog computing: Data analytics and cloud distributed processing on the network edges

The term fog computing was coined in 2012. However, the concept of pushing data and application logic to the network edges is not a novelty. Similar proposals were observed with edge computing, from the early 2000s, and cloudlets, from 2009. In fact, the cloudlet concept is a subset of edge computing applied to mobile networks and the fog concept is a subset of edge computing applied to Internet of Things (IoT). This paper demystifies these concepts and provides a comprehensive survey of references from academia and industry. It analyzes the terminology and dimensions of performance, security, and governance, based on a taxonomy proposed and presented in the paper. In addition we provide a thorough analysis of related topics, identifying the main research areas correlated to edge computing. Finally, we draw conclusions regarding the state of the art and the future of edge computing.

The Use of Distributed Processing and Cloud Computing in Agricultural Decision-Making Support Systems

One of the main challenges in agriculture is to sustainably meet the demand for food while preserving natural resources for future productions. Information Technology can assist producers to make better decisions by providing them with data and tools that enhance decision-making process, consequently allowing better management of the natural resources. Cloud-computing platforms and the extraction of data available on public weather related data sets allow the development of web applications that can assist producers with their investing and planning decisions. This paper describes the Big Weather solution, an agricultural decision-making support system that utilizes a cloud-computing platform, distributed processing technologies, and a big data framework. This paper also presents Big Weather architecture and an example of metric calculations (average temperature and humidity) and discusses the performance of the solution when tested in different virtual machine scenario configurations. The novelty is the transparency of the framework, which allows farmers to make better decisions based on data available on the cloud.

VIBox - Virtualized Internets-in-a-box: A tool for network planning experimentation

The simplistic design of the original Internet is being slowly replaced by the presence of advanced middleboxes that are introduced in the network. While some of these middleboxes are transparent and passively monitor the network, others can directly affect the data traffic (e.g., content accelerators). Forecasting user and middlebox traffic patterns are usually done through oversimplified models and sometimes unrealistic assumptions. In this context, network planning and dimensioning has become a complicated and sophisticated task. Even though virtualization technology today allows one to build experimental networks that can aid the task of network planning, this is hindered by lack of adequate functions in existing virtualization platforms. Amongst those functions are the support for refined network configurations and modification of software and operating system configurations running inside a virtual machine. In this paper we present VIBox, a tool for network planning experimentation that aims at creating virtual networks in a simple and user-friendly way.

Building an operator CDN the virtual way

Virtualization has opened an exciting and powerful way to experiment and evaluate complex distributed systems. In this paper we describe the experiences and lessons learned from building a distributed operator CDN consisting of 260+ virtual machine nodes in a six blade server unit. This provided us with a flexible platform to analyze a caching architecture we were developing. We also discuss the impact of virtual networking and virtual software routers on hardware resources.

Towards Bandwidth Optimization in Fog Computing using FACE Framework.

The continuous growth of data created by Internet-connected devices has been posing a challenge for mobile operators. The increase in the network traffic has exceeded the network capacity to efficiently provide services, specially for applications that require low latency. Edge computing is a concept that allows lowering the network traffic by using cloud-computing resources closer to the devices that either consume or generate data. Based on this concept, we designed an architecture that offers a mechanism to reduce bandwidth consumption. The proposed solution is capable of intercepting the data, redirecting it to a processing node that is allocated between the end device and the server, in order to apply features that reduce the amount of data on the network. The architecture has been validated through a prototype using video surveillance. This area of application was selected due to the high bandwidth requirement to transfer video data. Results show that in the best scenario is possible to obtain about 97% of bandwidth gain, which can improve the quality of services by offering better

How Advanced Cloud Services Can Improve Gaming Performance

This paper presents the idea of applying distributed processing to gaming. Trade Wind (TW), a cloud deployment and management solution, offers a distributed processing feature that can be applied to real-time games to improve performance and user experience. The main objective of this paper is to demonstrate how TW can optimize online gaming. A hypothesis was created based on the current features of TW and it is explained on this paper.