Cheng Zhong Xu

Load Balancing in Parallel Computers: Theory and Practice

Foreword. Preface. 1. Introduction. 2. A Survey of Nearest-Neighbor Load Balancing Algorithms. 3. The GDE Method. 4. GDE on Tori and Meshes. 5. The Diffusion Method. 6. GDE Versus Diffusion. 7. Termination Detection of Load Balancing. 8. Remapping with the GDE Method. 9. Load Distribution in Combinatorial Optimizations. 10. Conclusions. References. Index.

Exploring event correlation for failure prediction in coalitions of clusters

In large-scale networked computing systems, component failures become norms instead of exceptions. Failure prediction is a crucial technique for self-managing resource burdens. Failure events in coalition systems exhibit strong correlations in time and space domain. In this paper, we develop a spherical covariance model with an adjustable timescale parameter to quantify the temporal correlation and a stochastic model to describe spatial correlation. We further utilize the information of application allocation to discover more correlations among failure instances. We cluster failure events based on their correlations and predict their future occurrences. We implemented a failure prediction framework, called PREdictor of Failure Events Correlated Temporal-Spatially (hPREFECTs), which explores correlations among failures and forecasts the time-between-failure of future instances. We evaluate the performance of hPREFECTs in both offline prediction of failure by using the Los Alamos HPC traces and online prediction in an institute-wide clusters coalition environment. Experimental results show the system achieves more than 76% accuracy in offline prediction and more than 70% accuracy in online prediction during the time from May 2006 to April 2007.

Energy-aware parallel task scheduling in a cluster

Reducing energy consumption for high end computing can bring various benefits such as reducing operating costs, increasing system reliability, and environmental respect. This paper aims to develop scheduling heuristics and to present application experience for reducing power consumption of parallel tasks in a cluster with the Dynamic Voltage Frequency Scaling (DVFS) technique. In this paper, formal models are presented for precedence-constrained parallel tasks, DVFS-enabled clusters, and energy consumption. This paper studies the slack time for non-critical jobs, extends their execution time and reduces the energy consumption without increasing the tasks execution time as a whole. Additionally, Green Service Level Agreement is also considered in this paper. By increasing task execution time within an affordable limit, this paper develops scheduling heuristics to reduce energy consumption of a tasks execution and discusses the relationship between energy consumption and task execution time. Models and scheduling heuristics are examined with a simulation study. Test results justify the design and implementation of proposed energy aware scheduling heuristics in the paper.

Performance and energy modeling for live migration of virtual machines

Live migration of virtual machine (VM) provides a significant benefit for virtual server mobility without disrupting service. It is widely used for system management in virtualized data centers. However, migration costs may vary significantly for different workloads due to the variety of VM configurations and workload characteristics. To take into account the migration overhead in migration decision-making, we investigate design methodologies to quantitatively predict the migration performance and energy consumption. We thoroughly analyze the key parameters that affect the migration cost from theory to practice. We construct application-oblivious models for the cost prediction by using learned knowledge about the workloads at the hypervisor (also called VMM) level. This should be the first kind of work to estimate VM live migration cost in terms of both performance and energy in a quantitative approach. We evaluate the models using five representative workloads on a Xen virtualized environment. Experimental results show that the refined model yields higher than 90% prediction accuracy in comparison with measured cost. Model-guided decisions can significantly reduce the migration cost by more than 72.9% at an energy saving of 73.6%.

Energy-Aware Modeling and Scheduling for Dynamic Voltage Scaling with Statistical Real-Time Guarantee

Dynamic voltage scaling (DVS) is a promising technique for battery-powered systems to conserve energy consumption. Most existing DVS algorithms assume information about task periodicity or a priori knowledge about the task set to be scheduled. This paper presents an analytical model of general tasks for DVS assuming job timing information is known only after a task release. It models the voltage scaling process as a transfer function-based filtering system, which facilitates the design of two efficient scaling algorithms. The first is a time-invariant scaling policy and it is proved to be a generalization of several popular DVS algorithms for periodic, sporadic, and aperiodic tasks. A more energy efficient policy is a time-variant scaling algorithm for aperiodic tasks. It is optimal in the sense that it is online without assumed information about future task releases. The algorithm turns out to be a water-filling process with a linear time complexity. It can be applied to scheduling based on worst-case execution times as well as online slack distribution when jobs complete earlier. We further establish two relationships between computation capacity and deadline misses to provide a statistical real-time guarantee with reduced capacity

Live Virtual Machine Migration via Asynchronous Replication and State Synchronization

Live migration of virtual machines (VM) across physical hosts provides a significant new benefit for administrators of data centers and clusters. Previous memory-to-memory approaches demonstrate the effectiveness of live VM migration in local area networks (LAN), but they would cause a long period of downtime in a wide area network (WAN) environment. This paper describes the design and implementation of a novel approach, namely, CR/TR-Motion, which adopts checkpointing/recovery and trace/replay technologies to provide fast, transparent VM migration for both LAN and WAN environments. With execution trace logged on the source host, a synchronization algorithm is performed to orchestrate the running source and target VMs until they reach a consistent state. CR/TR-Motion can greatly reduce the migration downtime and network bandwidth consumption. Experimental results show that the approach can drastically reduce migration overheads compared with memory-to-memory approach in a LAN: up to 72.4 percent on application observed downtime, up to 31.5 percent on total migration time, and up to 95.9 percent on the data to synchronize the VM state. The application performance overhead due to migration is kept within 8.54 percent on average. The results also show that for a variety of workloads migrated across WANs, the migration downtime is less than 300 milliseconds.

Locality-Aware and Churn-Resilient Load-Balancing Algorithms in Structured Peer-to-Peer Networks

Structured peer-to-peer overlay networks, like distributed hash tables (DHTs), map data items to the network based on a consistent hashing function. Such mapping for data distribution has an inherent load balance problem. Data redistribution algorithms based on randomized matching of heavily loaded nodes with light ones can deal with the dynamics of DHTs. However, they are unable to consider the proximity of the nodes simultaneously. There are other methods that rely on auxiliary networks to facilitate locality-aware load redistribution. Due to the cost of network construction and maintenance, the locality-aware algorithms can hardly work for DHTs with churn. This paper presents a locality-aware randomized load-balancing algorithm to deal with both the proximity and network churn at the same time. We introduce a factor of randomness in the probing of lightly loaded nodes in a range of proximity. We further improve the efficiency by allowing the probing of multiple candidates (d-way) at a time. Simulation results show the superiority of the locality-aware two-way randomized algorithm in comparison with other random or locality-aware algorithms. In DHTs with churn, it performs no worse than the best chum-resilient algorithm. It takes advantage of node capacity heterogeneity and achieves good load balance effectively even in a skewed distribution of items

URL: A unified reinforcement learning approach for autonomic cloud management

Cloud computing is emerging as an increasingly important service-oriented computing paradigm. Management is a key to providing accurate service availability and performance data, as well as enabling real-time provisioning that automatically provides the capacity needed to meet service demands. In this paper, we present a unified reinforcement learning approach, namely URL, to automate the configuration processes of virtualized machines and appliances running in the virtual machines. The approach lends itself to the application of real-time autoconfiguration of clouds. It also makes it possible to adapt the VM resource budget and appliance parameter settings to the cloud dynamics and the changing workload to provide service quality assurance. In particular, the approach has the flexibility to make a good trade-off between system-wide utilization objectives and appliance-specific SLA optimization goals. Experimental results on Xen VMs with various workloads demonstrate the effectiveness of the approach. It can drive the system into an optimal or near-optimal configuration setting in a few trial-and-error iterations.

Predictive Directional Greedy Routing in Vehicular Ad hoc Networks

VANETs (vehicle ad hoc networks) are highly mobile wireless ad hoc networks targeted to support vehicular safety and other commercial applications. Conventional routing protocols in MANETs (mobile ad hoc networks) are unable to fully address the unique characteristics in vehicular networks. On the other hand, some characteristics in VANETs, like mobility constraints and predicable mobility, can benefit routing in vehicular networks. It is known that vehicles tend to move more regularly. We propose a new position-based routing strategy with the consideration of nodes moving direction for VANETs, called DGR (directional greedy routing). Considering the fact that vehicles often have predicable mobility, we propose PDGR (predictive directional greedy routing) to forward packet to the most suitable next hop based on both current and predicable future situations. We evaluate the performance of the solutions via simulations with realistic mobility models in open environments. Simulation results show that our solutions outperform existing ones in terms of packet delivery ratio, end-to-end delay, and routing overhead.

Emerging Directions in Embedded and Ubiquitous Computing

Trustworthiness, Reliability and Services in Ubiquitous and Sensor Networks.- Attack-Resilient Random Key Distribution Scheme for Distributed Sensor Networks.- A Critical Approach to Privacy Research in Ubiquitous Environments - Issues and Underlying Assumptions.- The Case Study of Information Security System for International Airports.- Quantitative Evaluation of Intrusion Tolerant Systems Subject to DoS Attacks Via Semi-Markov Cost Models.- An Efficient Mutual Authentication Scheme for EPCglobal Class-1 Generation-2 RFID System.- UPS - An Ubiquitous Proximity eService for Trust Collaboration.- Obligations for Privacy and Confidentiality in Distributed Transactions.- Multi-channel Enhancements for IEEE 802.11-Based Multi-hop Ad-Hoc Wireless Networks.- An Intelligent Event-Driven Interface Agent for Interactive Digital Contents in Ubiquitous Environments.- A Loop-Based Key Management Scheme for Wireless Sensor Networks.- A MAC Protocol with Little Idle Listening for Wireless Sensor Networks.- Security Technologies Based on Home Gateway for Making Smart Home Secure.- Layered Peer to Peer Streaming Using Hidden Markov Models.- Optimum Power Controller for Random Number Generator in the Crypto Module of Ubiquitous Computing Environment.- Problem Localization for Automated System Management in Ubiquitous Computing.- System and Software for Wireless SoC.- A High Speed Analog to Digital Converter for Ultra Wide Band Applications.- Design and DSP Software Implementation of Mobile WiMAX Baseband Transceiver Functions.- Cross-Layer Design for IEEE 802.16-2005 System Using Platform-Based Methodologies.- A Dynamic Frequency Allocation Scheme for IEEE 802.16 OFDMA-Based WMANs Using Hungary Algorithm.- Wireless Network Management System for WiMAX / Wi-Fi Mesh Networks.- An Implementation of QoS Framework for Heterogeneous Networks.- An Energy-Efficient MAC Design for IEEE 802.15.4-Based Wireless Sensor Networks.- A Cross-Layer Signaling and Middleware Platform for Multi-interface Mobile Devices.- Enhanced Sleep Mode Operations for Energy Saving in IEEE 802.16e.- Enhanced Fingerprint-Based Location Estimation System in Wireless LAN Environment.- Improving Channel Scanning Procedures for WLAN Handoffs.- Network Centric Ubiquitous Systems.- A Multicast Extension for Enhanced Mobile IP by Home Agent Handover.- Autonomic Multi-server Distribution in Flash Crowds Alleviation Network.- Generic Energy-Efficient Geographic Routing for Ad-Hoc Wireless Networks.- Description of a New Feature Meta-model.- Studying of Multi-dimensional Based Replica Management in Object Storage System.- Communication Model Exploration for Distributed Embedded Systems and System Level Interpretations.- An End-to-End QoS Adaptation Architecture for the Integrated IntServ and DiffServ Networks.- Ubiquitous Laboratory: A Research Support Environment for Ubiquitous Learning Based on Sensor Networks.- Intelligent Monitoring Using Wireless Sensor Networks.- On the Design of Micro-mobility for Mobile Network.- ANSWER: Adaptive Network Selection in WLAN/UMTS EnviRonment.- Self-authorized Public Key Management for Home Networks.- A Cross-Layered Diagnostician in OSGi Platform for Home Network.- Ubiquitous Underwater Acoustic-Sensor Network.- LaMSM: Localization Algorithm with Merging Segmented Maps for Underwater Sensor Networks.- TinyOS-Based Gateway for Underwater Acoustics/Radio Frequency Communication.- An Energy Scheduling Algorithm for Ensuring the Pre-determined Lifetime in Sensor Network.- Underwater Acoustic Communication and Modem-Based Navigation Aids.- State-of-the-Art in MAC Protocols for Underwater Acoustics Sensor Networks.- An Ultrasonic Sensor Based Low-Power Acoustic Modem for Underwater Communication in Underwater Wireless Sensor Networks.- UWA-NAV - Energy Efficient Error Control Scheme for Underwater Acoustic Sensor Network.- Underwater Wideband Source Localization Using the Interference Pattern Matching.- A New Virtual Select Database Operation for Wireless Sensor Networks.- GT2 - Reduced Wastes Time Mechanism for Underwater Acoustic Sensor Network.- RFID and Ubiquitous Sensor Networks.- Comparative Evaluation of Probabilistic and Deterministic Tag Anti-collision Protocols for RFID Networks.- An Efficient Mutual Authentication Protocol on RFID Tags.- HGLAP - Hierarchical Group-Index Based Lightweight Authentication Protocol for Distributed RFID System.- Target Classification in Sparse Sampling Acoustic Sensor Networks Using IDDC Algorithm.- Scriptable Sensor Network Based Home-Automation.- Applying Situation Awareness to Mobile Proactive Information Delivery.- Embedded Software Optimization.- Energy-Efficiency on a Variable-Bitrate Device.- The Secure DAES Design for Embedded System Application.- Software Power Peak Reduction on Smart Card Systems Based on Iterative Compiling.- Simultaneous Operation Scheduling and Operation Delay Selection to Minimize Cycle-by-Cycle Power Differential.- A Simple Approach to Robust Optimal Pole Assignment of Decentralized Stochastic Singularly-Perturbed Computer Controlled Systems.- Assured-Timeliness Integrity Protocols for Distributable Real-Time Threads with in Dynamic Distributed Systems.- Evaluating Modeling Solutions on Their Ability to Support the Partitioning of Automotive Embedded Systems.- Security in Ubiquitous Computing.- Security Analysis of the Certificateless Signature Scheme Proposed at SecUbiq 2006.- New Efficient Certificateless Signature Scheme.- A Practical Identity-Based Signature Scheme from Bilinear Map.- Linkable Ring Signatures from Linear Feedback Shift Register.- A Simple and Efficient Key Exchange Scheme Against the Smart Card Loss Problem.- A Key Distribution Scheme Preventing Collusion Attacks in Ubiquitous Heterogeneous Sensor Networks.- Token-Based Authenticated Key Establishment Protocols for Three-Party Communication.- Two Approaches on Pairwise Key Path Establishment for Sensor Networks.- An Efficient Authentication Protocol for RFID Systems Resistant to Active Attacks.- Low-Cost and Strong-Security RFID Authentication Protocol.- A Ticket Based Binding Update Authentication Method for Trusted Nodes in Mobile IPv6 Domain.