Krzysztof Rzadca

Decentralized Online Social Networks

Current Online social networks (OSN) are web services run on logically centralized infrastructure. Large OSN sites use content distribution networks and thus distribute some of the load by caching for performance reasons, nevertheless there is a central repository for user and application data. This centralized nature of OSNs has several drawbacks including scalability, privacy, dependence on a provider, need for being online for every transaction, and a lack of locality. There have thus been several efforts toward decentralizing OSNs while retaining the functionalities offered by centralized OSNs. A decentralized online social network (DOSN) is a distributed system for social networking with no or limited dependency on any dedicated central infrastructure. In this chapter we explore the various motivations of a decentralized approach to online social networking, discuss several concrete proposals and types of DOSN as well as challenges and opportunities associated with decentralization.

Interdisciplinary Matchmaking: Choosing Collaborators by Skill, Acquaintance and Trust

Social networks are commonly used to enhance recommender systems. Most of such systems recommend a single resource or a person. However, complex problems or projects usually require a team of experts that must work together on a solution. Team recommendation is much more challenging, mostly because of the complex interpersonal relations between members. This chapter presents fundamental concepts on how to score a team based on members’ social context and their suitability for a particular project. We represent the social context of an individual as a three-dimensional social network (3DSN) composed of a knowledge dimension expressing skills, a trust dimension and an acquaintance dimension. Dimensions of a 3DSN are used to mathematically formalize the criteria for prediction of the team’s performance. We use these criteria to formulate the team recommendation problem as a multi-criteria optimization problem. We demonstrate our approach on empirical data crawled from two web2.0 sites: onephoto.net and a social networking site. We construct 3DSNs and analyze properties of team’s performance criteria.

On the placement of reservations into job schedules

We present a new method for determining placements of flexible reservation requests into a schedule. For each considered placement the what-if method inserts a placeholder into the schedule and simulates the processing of batch jobs currently known to the system. Each placement is evaluated wrt. well-known scheduling metrics. This information may be used by a Grid reservation service to choose the most likely successful placement of a reservation. According to the results of extensive simulations, the what-if method grants more reservations and improves the performance of local jobs compared to our previously used load method.

Artificial immune systems applied to multiprocessor scheduling

We propose an efficient method of extracting knowledge when scheduling parallel programs onto processors using an artificial immune system (AIS). We consider programs defined by Directed Acyclic Graphs (DAGs). Our approach reorders the nodes of the program according to the optimal execution order on one processor. The system works in either learning or production mode. In the learning mode we use an immune system to optimize the allocation of the tasks to individual processors. Best allocations are stored in the knowledge base. In the production mode the optimization module is not invoked, only the stored allocations are used. This approach gives similar results to the optimization by a genetic algorithm (GA) but requires only a fraction of function evaluations.

Heterogeneous multiprocessor scheduling with differential evolution

The problem of scheduling a parallel program given by a directed acyclic graph (DAG) of tasks is a well-studied area. We present a new approach which employs differential evolution to numerically optimize the priorities of tasks. Our algorithm starts with a number of acceptable solutions, results of different heuristics, and merges them to achieve better one in a small number of function evaluations. The algorithm outperforms both a number of greedy heuristics and a classical genetic algorithm on the most of the program graphs considered in our experiments.

Promoting cooperation in selfish computational grids

In distributed computing, the recent paradigm shift from centrally-owned clusters to organizationally distributed computational grids introduces a number of new challenges in resource management and scheduling. In this work, we study the problem of Selfish Load Balancing which extends the well-known load balancing (LB) problem to scenarios in which each processor is concerned only with the performance of its local jobs. We propose a simple mathematical model for such systems and a novel function for computing the cost of the execution of foreign jobs. Then, we use the game-theoretic framework to analyze the model in order to compute the expected result of LB performed in a grid formed by two clusters. We show that, firstly, LB is a socially-optimal strategy, and secondly, for similarly loaded clusters, it is sufficient to collaborate during longer time periods in order to make LB the dominant strategy for each cluster. However, we show that if we allow clusters to make decisions depending on their current queue length, LB will never be performed. Then, we propose a LB algorithm which balances the load more equitably, even in the presence of overloaded clusters. Our algorithms do not use any external forms of compensation (such as money). The load is balanced only by considering the parameters of execution of jobs. This analysis is assessed experimentally by simulation, involving scenarios with multiple clusters and heterogeneous load.

Network Delay-Aware Load Balancing in Selfish and Cooperative Distributed Systems

We consider a geographically distributed request processing system composed of various organizations and their servers connected by the Internet. The latency a user observes is a sum of communication delays and the time needed to handle the request on a server. The handling time depends on the server congestion, i.e. the total number of requests a server must handle. We analyze the problem of balancing the load in a network of servers in order to minimize the total observed latency. We consider both cooperative and selfish organizations (each organization aiming to minimize the latency of the locally-produced requests). The problem can be generalized to the task scheduling in a distributed cloud; or to content delivery in an organizationally-distributed CDNs. In a cooperative network, we show that the problem is polynomially solvable. We also present a distributed algorithm iteratively balancing the load. We show how to estimate the distance between the current solution and the optimum based on the amount of load exchanged by the algorithm. During the experimental evaluation, we show that the distributed algorithm is efficient, therefore it can be used in networks with dynamically changing loads. In a network of selfish organizations, we prove that the price of anarchy (the worst-case loss of performance due to selfishness) is low when the network is homogeneous and the servers are loaded (the request handling time is high compared to the communication delay). After relaxing these assumptions, we assess the loss of performance caused by the selfishness experimentally, showing that it remains low. Our results indicate that a set of servers handling requests, connected in a heterogeneous network, can be efficiently managed by a distributed algorithm. Additionally, even if the network is organizationally distributed, with individual organizations optimizing performance of their requests, the network remains efficient.

SharedMind: A tool for collaborative mind-mapping

Current collaborative software usually have no or limited support for ad-hoc collaboration. SharedMind supports synchronous collaboration, i.e. real-time collaboration, and asynchronous collaboration, i.e. the merging of local instances of a document modified by different users after dis- and reconnects to a group of collaborators. SharedMind is completely decentralized and supports ad-hoc collaboration for interconnected (sub)groups. It demonstrates the confluence of social media and tools for computer supported collaborative works.

Game-Theoretic Mechanisms to Increase Data Availability in Decentralized Storage Systems

In a decentralized storage system, agents replicate each other’s data to increase availability. Compared to organizationally centralized solutions, such as cloud storage, a decentralized storage system requires less trust in the provider and may result in smaller monetary costs. Our system is based on reciprocal storage contracts that allow the agents to adopt to changes in their replication partners’ availability (by dropping inefficient contracts and forming new contracts with other partners). The data availability provided by the system is a function of the participating agents’ availability. However, a straightforward system in which agents’ matching is decentralized uses the given agent availability inefficiently. As agents are autonomous, the highly available agents form cliques replicating data between each other, which makes the system too hostile for the weakly available newcomers. In contrast, a centralized, equitable matching is not incentive compatible: it does not reward users for keeping their software running. We solve this dilemma by a mixed solution: an “adoption” mechanism in which highly available agents donate some replication space, which in turn is used to help the worst-off agents. We show that the adoption motivates agents to increase their availability (is incentive-compatible), but also that it is sufficient for acceptable data availability for weakly-available agents.

Fair Share Is Not Enough: Measuring Fairness in Scheduling with Cooperative Game Theory

We consider the problem of fair scheduling in a multi-organizational system in which organizations contribute their own resources to the global pool and the jobs to be processed on the common resources. We consider on-line, non-clairvoyant scheduling of sequential jobs without preemption. To ensure that the organizations are willing to cooperate the scheduling algorithm must be fair.