arXiv subjects

Luis Veiga

Publications and source records attributed to Luis Veiga.

4 recordsLinked to original sources

arXiv2020

Palpatine: Mining Frequent Sequences for Data Prefetching in NoSQL Distributed Key-Value Stores

This paper presents PALPATINE, the first in-memory application-level cache for Distributed Key-Value (DKV) data stores, capable of prefetching data that is likely to be accessed in an immediate future. To predict data accesses, PALPATINE continuously captures frequent access patterns to the back store by means of data mining techniques. With these patterns, PALPATINE builds a stochastic graph of accessed items, and makes prefetching decisions based on it. Experimental evaluation indicates that PALPATINE can improve the latency of a specific DKV store by more that an order of magnitude.

cs.DC↗

arXiv2017

Study on Resource Efficiency of Distributed Graph Processing

Graphs may be used to represent many different problem domains -- a concrete example is that of detecting communities in social networks, which are represented as graphs. With big data and more sophisticated applications becoming widespread in recent years, graph processing has seen an emergence of requirements pertaining data volume and volatility. This multidisciplinary study presents a review of relevant distributed graph processing systems. Herein they are presented in groups defined by common traits (distributed processing paradigm, type of graph operations, among others), with an overview of each system's strengths and weaknesses. The set of systems is then narrowed down to a set of two, upon which quantitative analysis was performed. For this quantitative comparison of systems, focus was cast on evaluating the performance of algorithms for the problem of detecting communities. To help further understand the evaluations performed, a background is provided on graph clustering.

cs.DC↗

arXiv2016

Bandwidth-aware Service Placement in Community Network Clouds

Seamless computing and service sharing in community networks have been gaining momentum due to the emerging technology of community network micro-clouds (CNMCs). However, running services in CNMCs can face enormous challenges such as the dynamic nature of micro-clouds, limited capacity of nodes and links, asymmetric quality of wireless links for services, deployment mod- els based on geographic singularities rather than network QoS, and etc. CNMCs have been increasingly used by network-intensive services that exchange significant amounts of data between the nodes on which they run, therefore the performance heavily relies on the available bandwidth resource in a network. This paper proposes a novel bandwidth-aware service placement algorithm which out- performs the current random placement adopted by Guifi.net. Our preliminary results show that the proposed algorithm consistently outperforms the current random placement adopted in Guifi.net by 35% regarding its bandwidth gain. More importantly, as the number of services increases, the gain tends to increase accordingly.

cs.NI↗

arXiv2008

Optimizing Binary Code Produced by Valgrind (Project Report on Virtual Execution Environments Course - AVExe)

Valgrind is a widely used framework for dynamic binary instrumentation and its mostly known by its memcheck tool. Valgrind's code generation module is far from producing optimal code. In addition it has many backends for different CPU architectures, which difficults code optimization in an architecture independent way. Our work focused on identifying sub-optimal code produced by Valgrind and optimizing it.

cs.PL↗