abstract.tex

\chapter*{Abstract}

The Personal Cloud model is a mainstream service that meets the growing demand of millions of users for reliable off-site storage.
However, very little is known about the quality of service (QoS) of Personal Clouds.

In this Thesis, first we present a \textit{measurement study} of three major Personal Clouds:
Dropbox, Box and SugarSync. Actively accessing to free accounts through their REST APIs, we analyzed important aspects to characterize their QoS, such as transfer speed and variability. Our measurement, conducted during two months, is the first to deeply analyze many facets of these popular services and reveals new insights, such as important performance differences among providers or the existence of transfer speed daily patterns.

Second, we focus on the synchronization protocol of various Personal Clouds. We perform some tests in order to
compare traffic overhead produced by desktop clients. In these tests we compare StackSync among other private and commercial solutions. StackSync, which has been developed by us during the last two years, is an open-source Personal Cloud that
provides scalable file synchronization and sharing.

We believe that the present analysis of Personal Clouds is of interest to researchers and developers with diverse concerns about Personal Clouds, since our observations can help them to understand and characterize the nature of these services.

%Unfortunately, commercial Personal Cloud solutions, e.g. Dropbox, Box, SkyDrive and the likes, are
%closed and proprietary, which supposes a serious impediment to progress research, also forcing
%people to be stranded into locked systems. In this context, we
%argue about the necessity for an open-source Personal Cloud framework that provides scalable file synchronization
%and sharing, and that allows anyone to easily	implement and evaluate new ideas.
%This framework, called StackSync, is modular and extensible, and contains all the software pieces to run
%a basic Personal Cloud, namely support for metadata management, efficient notification, deduplication, and data
%storage, among others. Our reference implementation in Java is a push-based architecture built around an asynchronous 
%high-performance message broker (RabbitMQ). We demonstrate the feasibility of our approach using
%our standard benchmarking test suite.

%\keywords{Personal Cloud, synchronization, storage, messaging middleware}