Showing papers by "Francesco Quaglia published in 2008"

DyMeLoR: Dynamic Memory Logger and Restorer Library for Optimistic Simulation Objects with Generic Memory Layout

Abstract: In this article we focus on checkpoint/restore facilities for optimistic simulation objects with generic memory layout. Specifically, we present the design and implementation of a {\tt C} library, named DyMeLoR (Dynamic Memory Logger and Restorer), that, beyond offering traditional services for dynamic memory allocation/release, additionally supports transparent checkpoint/restore of scattered simulation objects' states. DyMeLoR is well suited for being integrated within optimistic simulation platforms relying on kernel processes, each managing one or more simulation objects, as typical in most implementations of general purpose optimistic simulation platforms. From the point of view of efficiency, DyMeLoR has been designed in order to minimize memory consumption for meta-data describing the current layout of the simulation object's state, and to provide good trade-offs between the cost of meta-data manipulation and the cost of memory-to-memory data copies associated with checkpoint/restore tasks. Also, the library exhibits Piece-Wise-Deterministic (PWD) behavior, thus allowing the employment of (optimized) sparse checkpointing strategies each timethe overlying application software complies with the PWD assumption. We also report the results of an experimental study where DyMeLoR is integrated within the ROme OpTimistic Simulator (ROOT-Sim), and is used to support optimistic simulation of a cellular system.

A Performance Model of Multi-Version Concurrency Control

Abstract: In this article we present a performance model for multi-version concurrency control (MVCC). This type of concurrency control is currently very popular among mainstream commercial and open source database systems thanks to its ability to well cope with read intensive workloads, as in the case of transaction profiles proper of Web applications. To build the model we had to tackle the intrinsic higher complexity of MVCC when compared to traditional concurrency control mechanisms (i.e. 2-phase-locking and optimistic ones), such as the joint use of locks and aborts to resolve direct conflicts among write accesses to the same data item, and the management of multiple data versions. We validate our analytical model via an extensive simulation study, considering both uniform and skewed data accesses, as well as differentiated transaction profiles. To the best of our knowledge, the present study provides the first analytical model of MVCC.

APART: Low Cost Active Replication for Multi-tier Data Acquisition Systems

Abstract: This paper proposes APART (a posteriori active replication), a novel active replication protocol specifically tailored for multi-tier data acquisition systems. Unlike existing active replication solutions, APART does not rely on a-priori coordination schemes determining a same schedule of events across all the replicas, but it ensures replicas consistency by means of an a-posteriori reconciliation phase. The latter is triggered only in case the replicated servers externalize their state by producing an output event towards a different tier. On one hand, this allows coping with non-deterministic replicas, unlike existing active replication approaches. On the other hand, it allows attaining striking performance gains in the case of silent replicated servers, which only sporadically, yet unpredictably, produce output events in response to the receipt of a (possibly large) volume of input messages. This is a common scenario in data acquisition systems, where sink processes, which filter and/or correlate incoming sensor data, produce output messages only if some application relevant event is detected. Further, the APART replica reconciliation scheme is extremely lightweight as it exploits the cross-tier communication pattern spontaneously induced by the application logic to avoid explicit replicas coordination messages.

Fast Computation of Hyper-exponential Approximations of the Response Time Distribution of MMPP/M/1 Queues

Abstract: Input characterization to describe the flow of incoming traffic in network systems, such as the grid and the WWW, is often performed by using Markov modulated poisson processes (MMPP). Therefore, to enact capacity planning and quality-of-service (QoS) oriented design, the model of the servers that receive the incoming traffic is often described as a MMPP/M/1 queue. In a work we have provided an approximate solution for the response time distribution of the MMPP/M/1 queue, which is based on a hyper-exponential process obtained via a weighted superposition of the response time distributions of M/M/l queues. Compared to exact solution methods, or simulative techniques, the aim of this approximation is to provide the potential for more efficient model solution, so to enable, e.g., real-time what-if analysis in system reconfiguration scenarios. In this paper, we show how fast the computation can be supported in practical settings by ad-hoc techniques allowing the hyper-exponential model to be solved with no iterative or numerical costly steps, which would otherwise be required in order to compute the length of transient phases due to state switches in the MMPP arrival process. An application to the context of performance analysis of a grid system is also shown, supporting the efficiency of our proposal.

Integration and evaluation of Multi-Instance-Precommit schemes within postgreSQL

Abstract: Multi-instance-precommit (MIP) has been recently presented as an innovative transaction management scheme in support of reliability for Atomic Transactions in multitier (e.g. Web-based) systems. With this scheme, fail-over of a previously activated transaction can be supported via simple retry logics, which do not require knowledge about whether, and on which sites, the original transaction was precommitted. Mutual deadlock between the original and the retried transaction are prevented via MIP facilities, which also support reconciliation mechanisms for at-most-once transaction execution semantic. In this article we present an extension of the open source PostgreSQL database system in order to support MIP. The extension is based on the exploitation of PostgreSQL native multiversion concurrency control scheme. We also present an experimental evaluation based on the TPC-W benchmark, aimed at quantifying the relative overhead of MIP facilities on transaction execution latency, system throughput and storage usage.

Accuracy vs efficiency of hyper-exponential approximations of the response time distribution of MMPP/M/1 queues

Abstract: The Markov modulated Poisson process (MMPP) has been shown to well describe the flow of incoming traffic in networked systems, such as the Grid and the WWW. This makes the MMPP/M/1 queue a valuable instrument to evaluate and predict the service level of networked servers. In a recent work we have provided an approximate solution for the response time distribution of the MMPP/M/1 queue, which is based on a weighted superposition of M/M/l queues (i.e. a hyper-exponential process). In this article we address the tradeoff between the accuracy of this approximation and its computational cost. By jointly considering both accuracy and cost, we identify the scenarios where such approximate solution could be effectively used in support of network servers (dynamic) configuration and evaluation strategies, aimed at ensuring the agreed dependability levels in case of, e.g., request redirection due to faults.