Imperial College London

Displaying 81-120 of 176 results

  • White Papers // Aug 2010

    Random Neural Network for Emergency Management

    The authors consider decision problems in emergency management, such as simultaneously dispatching emergency teams to locations where incidents have occurred, and propose an algorithmic solution using the Random Neural Network. This is an NP-hard optimization problem, but the approach they suggest is solved in polynomial time, and is also distributed...

    Provided By Imperial College London

  • White Papers // Jul 2010

    A Test-Based Scheduling Protocol (TBSP) for Periodic Data Gathering in Wireless Sensor Networks

    The authors propose TBSP, a TDMA protocol for gathering information periodically from multiple data sources to a central location across multiple hops. TBSP incurs a lower overhead than the existing protocols because it only requires the sensor nodes to keep track of their own schedule, whereas in the existing protocols...

    Provided By Imperial College London

  • White Papers // Jul 2010

    Routing and G-Networks to Optimise Energy and Quality of Service in Packet Networks

    The authors formulate the problem of joint energy and Quality of Service (QoS) optimisation in packet networks and show how it can be formalised using the control capabilities inherent in G-network theory, which also includes the overhead due to control traffic. Using known energy consumption characteristics of network routers and...

    Provided By Imperial College London

  • White Papers // Jul 2010

    Activity Related Biometrics Based on Motion Trajectories

    This paper contributes to the concept of activity-related biometric authentication in ambient Intelligence environments. The motivation behind the proposed approach derives from activity-related biometrics and is mainly focusing on everyday activities. The activity sequence is captured by a stereoscopic camera and the resulting 2.5D data are processed to extract valuable...

    Provided By Imperial College London

  • White Papers // Jul 2010

    Shared and Searchable Encrypted Data for Untrusted Servers

    Current security mechanisms are not suitable for organisations that outsource their data management to untrusted servers. Encrypting and decrypting sensitive data at the client side is the normal approach in this situation but has high communication and computation overheads if only a subset of the data is required, for example,...

    Provided By Imperial College London

  • White Papers // Jul 2010

    An Empirical Study of the Scalability of Performance Analysis Tools in the Cloud

    Calculation of performance metrics such as steady state probabilities and response time distributions in large Markov and semi-Markov models can be accomplished using parallel implementations of well-known numerical techniques. In this paper the authors investigate the scalability of two existing parallel performance analysis tools (one based on Laplace transform inversion...

    Provided By Imperial College London

  • White Papers // Jun 2010

    Abstraction and Refinement for Local Reasoning

    Local reasoning has become a well-established technique in program verification, which has been shown to be useful at many different levels of abstraction. In separation logic, the authors use a low-level abstraction that is close to how the machine sees the program state. In context logic, they work with high-level...

    Provided By Imperial College London

  • White Papers // Jun 2010

    Mapping Multiple Multivariate Gaussian Random Number Generators on an FPGA

    A Multi-Variate Gaussian Random Number Generator (MVGRNG) is an essential block for many hardware designs, including monte carlo simulations. These simulations are usually used in applications such as statistical physics and financial mathematics. Field Programmable Gate Arrays (FPGAs) are often used to implement these generators as the design can be...

    Provided By Imperial College London

  • White Papers // May 2010

    Performance Trade-Offs in a Network Coding Router

    The authors consider the problem of optimizing the performance of a network coding router with two stochastic flows. They develop a queuing model which accounts for the fact that coding is not performed when packets are transmitted, but is done by a separate program or hardware which operates independently of...

    Provided By Imperial College London

  • White Papers // May 2010

    A Decentralised, Measurement-Based Admission Control Mechanism for Self-Aware Networks

    This paper presents a decentralised Admission Control (AC) algorithm. The authors' algorithm is a multiple criteria AC algorithm, where each user can specify the QoS metrics that interest him/her, and decides whether a new call should be allowed to enter the network based on measurements of the QoS metrics on...

    Provided By Imperial College London

  • White Papers // May 2010

    DEFCON: High-Performance Event Processing With Information Security

    In finance and healthcare, event processing systems handle sensitive data on behalf of many clients. Guaranteeing information security in such systems is challenging because of their strict performance requirements in terms of high event throughput and low processing latency. The authors describe DEFCON, an event processing system that enforces constraints...

    Provided By Imperial College London

  • White Papers // Apr 2010

    Enforcing User Privacy in Web Applications Using Erlang

    Social networking applications on the web handle the personal data of a large number of concurrently active users. These applications must comply with complex privacy requirements, while achieving scalability and high performance. Applying constraints to the flow of data through such applications to enforce privacy policy is challenging because individual...

    Provided By Imperial College London

  • White Papers // Apr 2010

    QoI-Aware Wireless Sensor Network Management for Dynamic Multi-Task Operations

    QoI-aware WSN O&M represents a broad area of research challenges that this paper only begins to address. Contrary to other research focusing on the network utility maximization problem with predefined utility functions, this paper employs a novel runtime design perspective where the WSN learns and optimizes the network utility by...

    Provided By Imperial College London

  • White Papers // Apr 2010

    Utility-Based Gateway Deployment for Supporting Multi-Domain DTNs

    Due to technology or policy constraints, communications across network domains usually require the intervention of gateways, and their proper deployment is crucial to the overall performance. In this paper, the authors study the problem of placing static gateways in mobile DTNs consisting of multiple domains. Given a limited gateway budget,...

    Provided By Imperial College London

  • White Papers // Mar 2010

    A Scripting Engine for Combining Design Transformations

    To implement complex designs quickly, designers increasingly turn to high-level design descriptions, which ease design capture and design space exploration, and allow rapid prototyping and fast time to market. This paper describes a scripting engine based on the python language and the ROSE compiler framework. The authors' scripting engine supports...

    Provided By Imperial College London

  • White Papers // Mar 2010

    Automated Precision Analysis: A Polynomial Algebraic Approach

    When migrating an algorithm onto hardware, the potential saving that can be obtained by tuning the precision used in the algorithm to meet a range or error specification is often overlooked; the major reason is that it is hard to choose a number system which can guarantee any such specification...

    Provided By Imperial College London

  • White Papers // Mar 2010

    Object Capabilities and Isolation of Untrusted Web Applications

    A growing number of current web sites combine active content (applications) from untrusted sources, as in so-called mashups. The object-capability model provides an appealing approach for isolating untrusted content: if separate applications are provided disjoint capabilities, a sound object-capability framework should prevent untrusted applications from interfering with each other, without...

    Provided By Imperial College London

  • White Papers // Mar 2010

    Simple Solutions for the Second Decade of Wireless Sensor Networking

    At 2010, the authors are at the effective end of the first decade of Wireless Sensor Network (WSN) research and that the aspects such as energy efficient routing have pretty much had their day as the sensing, computing and energy technologies have now moved on. There now seems to be...

    Provided By Imperial College London

  • White Papers // Mar 2010

    Presentation Abstract: Remote Elderly Care Monitoring in WASP

    The European project "Wireless Accessible Sensor Populations" (WASP) aims to develop cost-efficient solutions to facilitate WSN deployment and application-driven optimization. This paper addresses the deployment set-up and trials undertaken to enable continuous monitoring of elderly persons using unobtrusive (body-worn and ambient) wireless sensors. Such monitoring is intended to support elderly...

    Provided By Imperial College London

  • White Papers // Feb 2010

    Self-Healing for Pervasive Computing Systems

    The development of small wireless sensors and smart-phones, which include various sound, video, motion and location sensors have facilitated new pervasive applications. These pervasive systems are expected to perform in a vast number of environments, ranging from urban to rural, with different capabilities and resources. Application requirements may change dynamically...

    Provided By Imperial College London

  • White Papers // Feb 2010

    Randomized Scheduling Algorithm for Data Aggregation in Wireless Sensor Networks

    The authors consider a wireless sensor network in which a routing tree has been established to transmit the information from a set of source nodes to a data sink. The existing algorithms to schedule the transmission slots in a way that allows the data to be compressed as it moves...

    Provided By Imperial College London

  • White Papers // Feb 2010

    Public Support For Innovation, Intangible Investment And Productivity Growth In The UK Market Sector

    Pressure on public finances has increased scrutiny of public support for innovation. The authors examine two particular issues. First, there have been many recent calls for the (relatively new) UK R&D subsidy to be extended to other "Research" activities, such as software. Second, argument still rages about the efficacy of...

    Provided By Imperial College London

  • White Papers // Jan 2010

    Towards Automated Verification of Autonomous Networks: A Case Study in Self-Configuration

    In autonomic networks, the self-configuration of network entities is one of the most desirable properties. In this paper, the authors show how formal verification techniques can verify the correctness of self-configuration. As a case study, they describe the configuration of Physical Cell Identifiers (PCIs), a radio configuration parameter in cellular...

    Provided By Imperial College London

  • White Papers // Jan 2010

    Neighbor Discovery With Activity Monitoring in Multichannel Wireless Mesh Networks

    One way of improving performance of wireless mesh networks is to use multiple non-overlapping channels. At the same time, the mesh network must continuously self-adapt to varying radio conditions and topology changes. Thus, the authors propose a neighbor discovery protocol that fits the requirements of multichannel networks while dealing at...

    Provided By Imperial College London

  • White Papers // Jan 2010

    Dynamic Control of Data Ferries Under Partial Observations

    Controlled mobile helper nodes called data ferries have recently been proposed to bridge communications between disconnected nodes in a delay-tolerant manner. While existing work has explored various trajectory designs for the data ferry by assuming either static nodes or full observations at the data ferry, the problem remains open when...

    Provided By Imperial College London

  • White Papers // Dec 2009

    Energy-Efficient Broadcasts in Wireless Sensor Networks With Multiple Virtual Channels

    Multichannel solutions are increasingly used to cope with the problem of low capacity in sensor networks resulting from high contention during wake-up periods of nodes. The authors can consider wake-up schedules as virtual channels, because nodes using different schedules cannot communicate with each other. In this paper, they show that...

    Provided By Imperial College London

  • White Papers // Dec 2009

    A Fused Hybrid Floating-Point and Fixed-Point Dot-product for FPGAs

    Dot-products are one of the essential and recurrent building blocks in scientific computing, and often take-up a large proportion of the scientific acceleration circuitry. The acceleration of dot-products is very well suited for Field Programmable Gate Arrays (FPGAs) since these devices can be configured to employ wide parallelism, deep pipelining...

    Provided By Imperial College London

  • White Papers // Dec 2009

    Optimising Memory Bandwidth Use for Matrix-Vector Multiplication in Iterative Methods

    Computing the solution to a system of linear equations is a fundamental problem in scientific computing, and its acceleration has drawn wide interest in the FPGA community. One class of algorithms to solve these systems, iterative methods, has drawn particular interest, with recent literature showing large performance improvements over General...

    Provided By Imperial College London

  • White Papers // Dec 2009

    Design of a Financial Application Driven Multivariate Gaussian Random Number Generator for an FPGA

    A Multi-Variate Gaussian Random Number Generator (MVGRNG) is a pre-requisite for most monte carlo simulations for financial applications, especially those that involve many correlated assets. In recent years, Field Programmable Gate Arrays (FPGAs) have received a lot of attention as a target platform for the implementation of such a generator...

    Provided By Imperial College London

  • White Papers // Nov 2009

    Distributed Fault Tolerant Controllers

    Distributed applications are often built from sets of distributed components that must be co-ordinated in order to achieve some global behaviour. The common approach is to use a centralised controller for co-ordination, or occasionally a set of distributed entities. Centralised co-ordination is simpler but introduces a single point of failure...

    Provided By Imperial College London

  • White Papers // Nov 2009

    A Decision Support System for Disaster Management in Buildings

    Information systems that provide decision support in an efficient and timely manner can prove beneficial for emergency response operations. In this paper, the authors propose the use of a system that provides movement decision support to evacuees by directing them through the shortest or less hazardous routes to the exit...

    Provided By Imperial College London

  • White Papers // Nov 2009

    Portfolio Decisions With Higher Order Moments

    In this paper, the authors address the global optimization of two interesting nonconvex problems in finance. They relax the normality assumption underlying the classical Markowitz mean-variance portfolio optimization model and consider the incorporation of skewness (third moment) and kurtosis (fourth moment). The investor seeks to maximize the expected return and...

    Provided By Imperial College London

  • White Papers // Oct 2009

    Creating a Digital Ecosystem: Service-Oriented Architectures With Distributed Evolutionary Computing

    The authors start with a discussion of the relevant literature, including Nature Inspired Comuting as a framework in which to understand this work, and the process of biomimicry to be used in mimicking the necessary biological processes to create Digital Ecosystems. They then consider the relevant theoretical ecology in creating...

    Provided By Imperial College London

  • White Papers // Sep 2009

    Route Capacity Estimation Based Admission Control and QoS Routing for Mesh Networks

    Wireless Mesh Networks (WMNs) is a promising key technology for next generation wireless backhauling that is expected to support various types of applications with different Quality-of-Service (QoS) demands. Advanced antenna techniques, novel scheduling algorithms and routing schemes have attracted increased research interests aiming to optimize the performance of WMNs and...

    Provided By Imperial College London

  • White Papers // Sep 2009

    Modelling Zoned RAID Systems using Fork-Join Queueing Simulation

    RAID systems are ubiquitously deployed in storage environments, both as standalone storage solutions and as fundamental components of virtualized storage platforms. Accurate models of their performance are crucial to delivering storage infrastructures that meet given quality of service requirements. To this end, this paper presents a flexible fork-join queuing simulation...

    Provided By Imperial College London

  • White Papers // Sep 2009

    Data Placement and Migration Strategies for Virtualised Data Storage Systems

    In this paper, the authors details a simulation intended to explore the viability of storage systems where data is placed intelligently depending on data-specific quality of service requirements and the characteristics of the underlying storage devices. Their implementation provides a framework for the automatic pro ling of a virtualized storage...

    Provided By Imperial College London

  • White Papers // Sep 2009

    A Performance Model of Zoned Disk Drives with I/O Request Reordering

    Disk drives are a common performance bottleneck in modern storage systems. To alleviate this, disk manufacturers employ a variety of I/O request scheduling strategies which aim to reduce disk head positioning time by dynamically reordering queuing requests. An analytical model of this phenomenon is best represented by an M/G/1 queue...

    Provided By Imperial College London

  • White Papers // Sep 2009

    Truncation of Passage Time Calculations in Semi-Markov Models

    Calculation of passage time distributions in large semi-Markov models can be accomplished by means of a previously-presented iterative algorithm, the core of which is repeated sparse matrix - vector multiplication. The algorithm's performance is therefore highly dependent on the number of multiplications of matrix and vector elements that must be...

    Provided By Imperial College London

  • White Papers // Sep 2009

    Methodology for Designing Statically Scheduled Application-Specific SDRAM Controllers using Constrained Local Search

    In this paper the authors present an automatic method for generating valid SDRAM command schedules which obey the timing restrictions of DDR2 memory from a set of memory references. These generated schedules can be implemented using a static memory controller. A complete knowledge of the sequence of memory references in...

    Provided By Imperial College London

  • White Papers // Sep 2009

    Automatic Optimisation of MapReduce Designs by Geometric Programming

    Many important applications can be expressed using the MapReduce pattern, where a computation is decomposed into a map phase on which each element of source data is independently operated, followed by a reduce phase in which the mapped elements are combined with an associative operator. The authors develop an approach...

    Provided By Imperial College London

  • White Papers // Sep 2009

    An Overview of Low-Power Techniques for Field-Programmable Gate Arrays

    This paper provides an overview of low-power techniques for Field-Programmable Gate Arrays (FPGAs). It covers system-level design techniques and device-level design techniques that have targeted current commercial devices. It also describes current research on circuit-level and architecture-level design techniques. Recent studies on power modelling and on low-power Computer-Aided Design (CAD)...

    Provided By Imperial College London

  • White Papers // Apr 2009

    Context-Aware Publish-Subscribe: Model, Implementation, and Evaluation

    Complex communication patterns often need to take into account the situation in which the information to be communicated is produced or consumed. Publish-subscribe, and particularly its content-based incarnation, is often used to convey this information by encoding the "Context" of the publisher into the published messages. In this paper, the...

    Provided By Imperial College London

  • White Papers // Apr 2009

    Improving Federation Executions With Migrating HLA/RTI Central Runtime Components

    Simulation tools are popular in the design and study of communication systems and networks. The implementation of analytic distributed simulations commonly requires a central entity to support certain tasks (e.g., time management). The HLA/RTI uses a Central Runtime Component (CRC) to this end. Because of network variability, the location of...

    Provided By Imperial College London

  • White Papers // Dec 2011

    Migrating Auctioneers on Internet Auctions for Improved Utility and Performance

    The paper studies a technique to improve the utility and performance of an automated auction application where the auctioneer and bidders communicate through the Internet. The lack of quality-of-service guarantees from site to site can severely influence the results of an auction by affecting the seller's income rate, auction fairness,...

    Provided By Imperial College London

  • White Papers // Apr 2007

    Detecting Denial of Service Attacks with Bayesian Classifiers and the Random Neural Network

    Denial of Service (DoS) is a prevalent threat in today's networks. While such an attack is not difficult to launch, defending a network resource against it is disproportionately difficult, and despite the extensive research in recent years, DoS attacks continue to harm. The first goal of any protection scheme against...

    Provided By Imperial College London

  • White Papers // Oct 2013

    Managing Emergencies Optimally Using a Random Neural Network-Based Algorithm

    Emergency rescues require that first responders provide support to evacuate injured and other civilians who are obstructed by the hazards. In this case, the emergency personnel can take actions strategically in order to rescue people maximally, efficiently and quickly. The paper studies the effectiveness of a Random Neural Network (RNN)-based...

    Provided By Imperial College London

  • White Papers // Jan 2009

    Adaptability and Failure Resilience of the Cognitive Packet Network

    Self Aware Networks (SAN) is a proposal of QoS-enabled networks with enhanced monitoring and self improvement capabilities that use adaptive packet routing proto-cols. Such an example is Cognitive Packet Network (CPN), which addresses QoS by using adaptive techniques based on online measurements. It is a distributed protocol that provides QoS-driven...

    Provided By Imperial College London

  • White Papers // May 2007

    A Biologically Inspired Denial of Service Detector Using the Random Neural Network

    Several of today's computing challenges have been met by resorting to and adapting optimal solutions that have evolved in nature. For example, autonomic communication networks have started applying biologically-inspired methods to achieve some of their self-properties. The authors build upon such methods to solve the recent problem of detection of...

    Provided By Imperial College London

  • White Papers // Jul 2010

    Shared and Searchable Encrypted Data for Untrusted Servers

    Current security mechanisms are not suitable for organisations that outsource their data management to untrusted servers. Encrypting and decrypting sensitive data at the client side is the normal approach in this situation but has high communication and computation overheads if only a subset of the data is required, for example,...

    Provided By Imperial College London

  • White Papers // Mar 2006

    UNITE: Uniform Hardware-Based Network Intrusion DeTection Engine

    Current software implementations of network intrusion detection reach a maximum network connection speed of about 1Gbps (Gigabits per second). This paper analyses the Snort software network intrusion detection system to highlight the bottlenecks of such systems. It proposes a novel packet processing engine called UNITE that deploys uniform hardware architecture...

    Provided By Imperial College London

  • White Papers // Mar 2012

    MODACLOUDS: A Model-Driven Approach for the Design and Execution of Applications on Multiple Clouds

    Cloud computing is emerging as a major trend in the ICT industry. While most of the attention of the research community is focused on considering the perspective of the cloud providers, offering mechanisms to support scaling of resources and interoperability and federation between clouds, the perspective of developers and operators...

    Provided By Imperial College London

  • White Papers // Mar 2012

    MODACLOUDS, a Model-Driven Approach for the Design and Execution of Applications on Multiple Clouds

    Cloud computing is emerging as a major trend in the ICT industry. While most of the attention of the research community is focused on considering the perspective of the Cloud providers, offering mechanisms to support scaling of resources and interoperability and federation between Clouds, the perspective of developers and operators...

    Provided By Imperial College London

  • White Papers // Dec 2006

    Response Time Densities and Quantiles in Large Markov and Semi-Markov Models

    Response time quantiles reflect user-perceived quality of service more accurately than mean or average response time measures. Consequently, online transaction processing benchmarks, telecommunications service level agreements and emergency services legislation all feature stringent 90th percentile response time targets. This paper describes a range of techniques for extracting response time densities...

    Provided By Imperial College London

  • White Papers // May 2009

    Distributed Response Time Analysis of GSPN Models with MapReduce

    Generalized Stochastic Petri Nets (GSPNs) are widely used in the performance analysis of computer and communications systems. Response time densities and quantiles are often key outputs of such analysis. These can be extracted from a GSPN's underlying semi-markov process using a method based on numerical Laplace transform inversion. This method...

    Provided By Imperial College London

  • White Papers // Apr 2013

    Bayesian Service Demand Estimation with Gibbs Sampling

    Performance modeling of web applications involves the task of estimating service demands of requests at physical resources, such as CPUs. In this paper, the authors propose a service demand estimation algorithm based on a Markov Chain Monte Carlo (MCMC) technique, Gibbs sampling. Their methodology is widely applicable as it requires...

    Provided By Imperial College London

  • White Papers // Apr 2013

    An Offline Demand Estimation Method for Multi-Threaded Applications

    Parameterizing performance models for multithreaded enterprise applications requires finding the service rates offered by worker threads to the incoming requests. Statistical inference on monitoring data is here helpful to reduce the overheads of application profiling and to infer missing information. While linear regression of utilization data is often used to...

    Provided By Imperial College London

  • White Papers // Nov 2012

    OFBench: an Enterprise Application Benchmark for Cloud Resource Management Studies

    The authors introduce OFBench, a new research benchmark for enterprise applications. OFBench is a load generator for the demo e-commerce component of the Apache OFBiz Enterprise Resource Planning (ERP) framework. ERP applications are increasingly important in the cloud market due to the growing popularity of Software-as-a-Service (SaaS) solutions; hence OFBench...

    Provided By Imperial College London

  • White Papers // Mar 2013

    Modelling Exogenous Variability in Cloud Deployments

    Describing exogenous variability in the resources used by a cloud application leads to stochastic performance models that are difficult to solve. In this paper, the authors describe the blending algorithm, a novel approximation for queuing network models immersed in a random environment. Random environments are Markov chain-based descriptions of time-varying...

    Provided By Imperial College London

  • White Papers // Sep 2010

    IO Performance Prediction in Consolidated Virtualized Environments

    The authors proposed a trace-driven approach to predict the performance degradation of disk request response times due to storage device contention in consolidated virtualized environments. Their performance model evaluates a queueing network with fair share scheduling using trace-driven simulation. The model parameters can be deduced from measurements obtained inside Virtual...

    Provided By Imperial College London

  • White Papers // Apr 2013

    Fitting Second-Order Acyclic Marked Markovian Arrival Processes

    Markovian Arrival Processes (MAPs) are a tractable class of point-processes useful to model correlated time series, such as those commonly found in network traces and system logs used in performance analysis and reliability evaluation. Marked MAPs (MMAPs) generalize MAPs by further allowing the modeling of multi-class traces, possibly with cross-correlation...

    Provided By Imperial College London

  • White Papers // Apr 2011

    Approximate Analysis of Blocking Queueing Networks with Temporal Dependence

    In this paper, the authors extend the class of MAP queueing networks to include blocking models, which are useful to describe the performance of service instances which have a limited concurrency level. They consider two different blocking mechanisms: Repetitive Service-Random Destination (RS-RD) and Blocking After Service (BAS). They propose a...

    Provided By Imperial College London

  • White Papers // Mar 2012

    WIQ:Work-Intensive Query Scheduling for In-Memory Database Systems

    The authors propose a novel admission control policy for database queries. Their methodology uses system measurements of CPU utilization and query backlogs to determine interference between queries in execution on the same database server. Query interference may arise due to the concurrent access of hardware and software resources and can...

    Provided By Imperial College London

  • White Papers // Apr 2011

    Markovian Workload Characterization for QoS Prediction in the Cloud

    Resource allocation in the cloud is usually driven by performance predictions, such as estimates of the future incoming load to the servers or of the Quality-of-Service (QoS) offered by applications to end users. In this context, characterizing web workload fluctuations in an accurate way is fundamental to understand how to...

    Provided By Imperial College London

  • White Papers // Sep 2009

    Modelling Zoned RAID Systems using Fork-Join Queueing Simulation

    RAID systems are ubiquitously deployed in storage environments, both as standalone storage solutions and as fundamental components of virtualized storage platforms. Accurate models of their performance are crucial to delivering storage infrastructures that meet given quality of service requirements. To this end, this paper presents a flexible fork-join queuing simulation...

    Provided By Imperial College London

  • White Papers // Sep 2009

    Data Placement and Migration Strategies for Virtualised Data Storage Systems

    In this paper, the authors details a simulation intended to explore the viability of storage systems where data is placed intelligently depending on data-specific quality of service requirements and the characteristics of the underlying storage devices. Their implementation provides a framework for the automatic pro ling of a virtualized storage...

    Provided By Imperial College London

  • White Papers // Sep 2009

    A Performance Model of Zoned Disk Drives with I/O Request Reordering

    Disk drives are a common performance bottleneck in modern storage systems. To alleviate this, disk manufacturers employ a variety of I/O request scheduling strategies which aim to reduce disk head positioning time by dynamically reordering queuing requests. An analytical model of this phenomenon is best represented by an M/G/1 queue...

    Provided By Imperial College London

  • White Papers // Jun 2014

    Self-Adaptive Containers: Functionality Extensions and Further Case Study

    As the number of execution environments and application contexts rises exponentially, ever-changing non-functional requirements can lead to repeated code refactoring. In addition, scaling up software to support large input sizes may require major modification of code. To address these challenges, the authors have previously proposed a framework of self-adaptive containers...

    Provided By Imperial College London

  • White Papers // May 2013

    PEPERCORN: Inferring Performance Models from Location Tracking Data

    Stochastic performance models are widely used to analyze the performance of systems that process customers and resources. However, the construction of such models is traditionally manual and therefore expensive, intrusive and prone to human error. In this paper, the authors introduce PEPERCORN, a Petri Net Performance Model (PNPM) construction tool,...

    Provided By Imperial College London

  • White Papers // Apr 2011

    Data Allocation Strategies for the Management of Quality of Service in Virtualised Storage Systems

    The amount of data managed by organizations continues to grow relentlessly. Driven by the high costs of maintaining multiple local storage systems, there is a well established trend towards storage consolidation using multi-tier Virtualized Storage Systems (VSSs). At the same time, storage infrastructures are increasingly subject to stringent Quality of...

    Provided By Imperial College London

  • White Papers // Jun 2008

    Modelling and Validation of Response Times in Zoned RAID

    The authors present and validate an enhanced analytical queuing network model of zoned RAID. The model focuses on RAID levels 01 and 5, and yields the distribution of I/O request response time. Whereas their previous paper could only support arrival streams of I/O requests of the same type, the model...

    Provided By Imperial College London

  • White Papers // Jan 2014

    Simulation and Modelling of RAID 0 System Performance

    RAID systems are fundamental components of modern storage infrastructures. It is therefore important to model their performance effectively. This paper describes a simulation model which predicts the cumulative distribution function of I/O request response time in a RAID 0 system consisting of homogeneous zoned disk drives. The model is constructed...

    Provided By Imperial College London

  • White Papers // Jul 2008

    A Parallel and Distributed Analysis Pipeline for Performance Tree Evaluation

    Performance Trees (PTs) provide the means to express a wide variety of queries on stochastic models. Graphically, PTs are represented as tree structures consisting of operation and value nodes. Performance trees are a unifying framework for the specification of performance queries involving measures and requirements. This paper describes an evaluation...

    Provided By Imperial College London

  • White Papers // Apr 2009

    Towards the Automated Inference of Queueing Network Models From High-Precision Location Tracking Data

    Traditional methods for deriving performance models of customer flow in real-life systems are manual, time consuming and prone to human error. This paper proposes an automated four-stage data processing pipeline which takes as input raw high-precision location tracking data and which outputs a queuing network model of customer flow. The...

    Provided By Imperial College London

  • White Papers // Sep 2009

    Truncation of Passage Time Calculations in Semi-Markov Models

    Calculation of passage time distributions in large semi-Markov models can be accomplished by means of a previously-presented iterative algorithm, the core of which is repeated sparse matrix - vector multiplication. The algorithm's performance is therefore highly dependent on the number of multiplications of matrix and vector elements that must be...

    Provided By Imperial College London

  • White Papers // May 2009

    Towards a QoS-aware Virtualised Storage System

    Every organization depends critically on reliable high-performance storage. Driven by the high costs of maintaining and managing multiple local storage systems, there is a trend towards virtualized multi-tier storage infrastructures. The main limitation of such centralized solutions is their inability to guarantee application-level Quality of Service (QoS) without extensive and...

    Provided By Imperial College London

  • White Papers // Jun 2008

    State-Space Size Estimation By Least-Squares Fitting

    The authors present a method for estimating the number of states in the Continuous Time Markov Chains (CTMCs) underlying high-level models using least-squares fitting. Their paper improves on existing techniques by producing a numerical estimate of the number of states rather than classifying the state space into on of three...

    Provided By Imperial College London

  • White Papers // Jun 2008

    Validation of Large Zoned RAID Systems

    Building on the authors' prior paper in, they present an improved model for large partial stripe following full stripe writes in RAID 5. This was necessary because they observed that their previous model tended to underestimate measured results. To date, they have only validated these models against RAID systems with...

    Provided By Imperial College London

  • White Papers // Mar 2007

    Tackling Large State Spaces in Performance Modelling

    Stochastic performance models provide a powerful way of capturing and analyzing the behavior of complex concurrent systems. Traditionally, performance measures for these models are derived by generating and then analyzing a (semi-)Markov chain corresponding to the model's behavior at the state-transition level. However, and especially when analyzing industrial-scale systems, workstation...

    Provided By Imperial College London

  • White Papers // May 2006

    Distributed Computation of Transient State Distributions and Passage Time Quantiles in Large Semi-Markov Models

    Semi-Markov Processes (SMPs) are expressive tools for modeling parallel and distributed systems; they are a generalization of markov processes that allow for arbitrarily distributed sojourn times. This paper presents an iterative technique for transient and passage time analysis of large structurally unrestricted semi-Markov processes. The authors' method is based on...

    Provided By Imperial College London

  • White Papers // Jul 2006

    A General Graph Model For Representing Exact Communication Volume in Parallel Sparse Matrix-Vector Multiplication

    In this paper, the authors present a new graph model of sparse matrix decomposition for parallel sparse matrix - vector multiplication. Their model differs from previous graph-based approaches in two main respects. Firstly, their model is based on edge coloring rather than vertex partitioning. Secondly, their model is able to...

    Provided By Imperial College London