Salvatore Tucci

A Walk through Content Delivery Networks

Content Delivery Networks (CDN) aim at overcoming the inherent limitations of the Internet. The main concept at the basis of this technology is the delivery at edge points of the network, in proximity to the request areas, to improve the users perceived performance while limiting the costs. This paper focuses on the main research areas in the field of CDN, pointing out the motivations, and analyzing the existing strategies for replica placement and management, server measurement, best fit replica selection and request redirection.

Federated Agent-based Modeling and Simulation Approach to Study Interdependencies in IT Critical Infrastructures

Agent-based modeling and simulation (ABMS) is one of the more promising simulation techniques to study the interdependencies in critical infrastructures. Moreover, federated simulation has two relevant properties, simulation models reuse and expertise sharing, that could be exploited in a multi-sectorial field, such as critical infrastructure protection. In this paper we propose a new methodology which exploits the benefit of both ABMS and Federated simulation, to study interdependencies in critical infrastructures. First of all we discus advantages of federated agent-based modeling and difficulties in implementing a Federated ABMS framework. To demonstrate the relevance of our solution we propose an example driven approach that poses the attention on critical information infrastructure. We have also implemented a Federated ABMS framework, which federate Repast, an agent-based simulation engine and OMNeT++ an IT systems and communication networks modeling and simulation environment. A selection of simulation results shown how Federated ABMS could shed light on system interdependencies and how it helps in quantifying them.

Static and dynamic scheduling algorithms for scalable Web server farm

Multiprocessor-based servers are often used for building popular Web sites which have to guarantee an acceptable Quality of Web Service. In common multi-node systems, namely Web server farms, a Web switch (say, Dispatcher) routes client requests among the server nodes. This architecture resembles a traditional cluster in which a global scheduler dispatches parallel applications among the server nodes. The main difference is that the load reaching Web server farms tends to occur in waves with intervals of heavy peaks. These heavy-tailed characteristics have motivated the use of policies based on dynamic state information for global scheduling in Web server farms. This paper presents an accurate comparison between static and dynamic policies for different classes of Web sites. The goal is to identify main features of architectures and load management algorithms that guarantee scalable Web services. We verify that a Web farm with a Dispatcher with full control on client connections is a very robust architecture. Indeed, we demonstrate that if the Web sire provides only HTML pages or simple database searches, the Dispatcher does not need to use sophisticated scheduling algorithms even if the load occurs in heavy bursts. Dynamic scheduling policies appears to be necessaly for scalability only when most requests are for Web services of three or more orders of magnitude higher than providing HTML pages with some embedded objects.

Mechanisms for quality of service in Web clusters

The new generation of Web systems provides more complex services than those related to Web publishing sites. Users are increasingly reliant on the Web for up-to-date personal and business information and services. Web architectures able to guarantee the quality of service (QoS) that rules the relationship between users and Web service providers require a large investment in new algorithms and systems for dispatching, load balancing, and information consistency. In this paper, we consider Web cluster architectures composed of multiple back-end server nodes and one front-end dispatcher and we analyze how to provide differentiated service levels to various classes of users. We demonstrate through simulation experiments under realistic workload models that the proposed mechanisms are able to satisfy QoS requirements of the most valuable users classes, without impacting too negatively on the other users.

Performance analysis of circuit-switching interconnection networks with deterministic and adaptive routing

This paper compares three link conflict resolution strategies applied to multicomputers with symmetric topologies and circuit-switching interconnection networks. Several performance parameters are evaluated through an approximate analytical model based on the flow analysis. The main peculiarity of this method with respect to previous studies is the capacity to take into account actual network delays and all feedback effects among probability of link conflict, routing controller overhead, and message latency. An extensive simulation analysis has been carried out to validate the analytical models. The results show that our approach is quite accurate for a wide range of message traffic loads, independently of the link conflict resolution strategy and message length distribution.

Use of hybrid recursive CSR/COO data structures in sparse matrix-vector multiplication

Recently, we have introduced an approach to basic sparse matrix computations on multicore cache based machines using recursive partitioning. Here, the memory representation of a sparse matrix consists of a set of submatrices, which are used as leaves of a quad-tree structure. In this paper, we evaluate the performance impact, on the Sparse Matrix-Vector Multiplication (SpMV), of a modification to our Recursive CSR implementation, allowing the use of multiple data structures in leaf matrices (CSR/COO, with either 16/32 bit indices).

On the Usage of 16 Bit Indices in Recursively Stored Sparse Matrices

In our earlier work, we have investigated the feasibility of utilization of recursive partitioning in basic (BLAS oriented) sparse matrix computations, on multi-core cache-based computers. Following encouraging experimental results obtained for SPMV and SPSV operations, here we proceed to tune the storage format. To limit the memory bandwidth overhead we introduce usage of shorter (16 bit) indices in leaf sub matrices (at the end of the recursion). Experimental results obtained for the proposed approach on 8-core machines illustrate speed improvements, when performing sparse matrix-vector multiplication.

A hierarchical approach for bounding the completion time distribution of stochastic task graphs

The analytical evaluation of the completion time distribution of a general directed acyclic graph (DAG) is known to be an NP-complete problem. In this paper we present a new algorithm, named Tree Bound , for the evaluation of bounds on the completion time of stochastic graphs assuming ideal conditions for shared resources and independent random variables as task execution times. The Tree Bound method uses a hierarchical approach that first gives a tree-like representation of the graph, and then evaluates lower and upper bounds through a single visit of the tree. As lower bound the method takes the distribution of an embedded series‐parallel graph which is evaluated by means of a simple recursion. The upper bound is based on a hierarchical application of other bounding techniques. In this paper, we use the Shogan algorithm because its determinism allows us to demonstrate some interesting properties of the Tree Bound method. Indeed, through stochastic ordering and stochastic comparison techniques, we demonstrate analytically that our approach provides tighter bounds than Shogan’s and Yazici-Pekergin’s bounds. On the other hand, we cannot compare formally the Tree Bound accuracy to that of other important methods, such as Kleinoder and Dodin, because of their non-determinism. Various empiric comparisons show that the Tree Bound algorithm provides analogous or superior results than heuristics derived from main non-deterministic methods. Moreover, the Tree Bound algorithm keeps linear complexity and avoids non-determinism. Finally, it represents a useful basis for the combination of different bounding techniques which seems the only way to achieve even tighter bounds on the completion time distribution of stochastic graphs.

On BLAS Operations with Recursively Stored Sparse Matrices

Recently, we have proposed a recursive partitioning based layout for multi-core computations on sparse matrices. Based on positive results of our initial experiments with matrix-vector multiplication, we discuss how this storage format can be utilized across a range of BLAS-style matrix operations.

Modeling and Simulation of Complex Interdependent Systems: A Federated Agent-Based Approach

Critical Interdependent Infrastructures are complex interdependent systems, that if damaged or disrupted can seriously compromise the welfare of our society. This research, part of the CRESCO project, faces the problem of interdependent critical infrastructures modeling and simulation proposing an agent-based solution. The approach we put forward, named Federated ABMS, relies on discrete agent-based modeling and simulation and federated simulation. Federated ABMS provides a formalism to model compound complex systems, composed of interacting systems, as federation of interacting agents and sector specific simulation models. This paper describes the formal model as well it outlines the steps that characterize the Federated ABMS methodology, here applied to a target system, composed of a communication network and of a power grid. Moreover we conclude the paper with a thorough discussion of implementation issues.