Yijun Lu

Adaptive Consistency Guarantees for Large-Scale Replicated Services

To maintain consistency, designers of replicated services have traditionally been forced to choose from either strong consistency guarantees or none at all. Realizing that a continuum between strong and optimistic consistencies is semantically meaningful for a broad range of network services, previous research has proposed a continuous consistency model for replicated services to support the tradeoff between the guaranteed consistency level, performance and availability. However, to meet changing application needs and to make the model useful for interactive users of large-scale replicated services, the adaptability and the swiftness of inconsistency resolution are important and challenging. This paper presents IDEA (an infrastructure for detection-based adaptive consistency guarantees) for adaptive consistency guarantees of large-scale, Internet-based replicated services. The main functions enabled by IDEA include quick inconsistency detection and resolution, consistency adaptation and quantified consistency level guarantees. Through experimentation on the Planet-Lab, IDEA is evaluated from two aspects: its adaptive consistency guarantees and its performance for inconsistency resolution. Results show that IDEA is able to provide consistency guarantees adaptive to userpsilas changing needs, and it achieves low delay for inconsistency resolution and incurs small communication overhead.

An efficient, low-cost inconsistency detection framework for data and service sharing in an Internet-scale system

In this paper, we argue that a broad range of Internet-scale distributed applications can benefit from an underlying low-cost consistency detection framework that is an alternative to inconsistency avoidance and can detect inconsistency among nodes sharing data or services in a timely manner. This paper first presents an overview of the inconsistency detection framework. Then, it discusses the detailed design of the two-layer inconsistency detection module, the core component of this framework, which can detect inconsistency among nodes in a timely manner. The proposed two-layer inconsistency detection module is evaluated both analytically and via simulations, which shows that this module can significantly reduce the time to detect inconsistency among nodes without adding much maintenance cost. Finally, this paper outlines the possible mechanisms to discern the application semantics and to resolve the detected inconsistencies

IDEA:: an infrastructure for detection-based adaptive consistency control in replicated services

In Internet-scale distributed and replicated services, poor consistency results in poor QoS or even monetary loss. Recent research focuses on enforcing a certain consistency level, instead of perfect consistency, to strike a balance between consistency guarantee and systems scalability. In this paper, we argue that it is equally, if not more, important to achieve adaptability. I.e., the system adjusts its consistency level on the fly to suit applications. ongoing need. This paper presents IDEA (an Infrastructure for DEtection-based Adaptive consistency control), a protocol that adaptively controls consistency in replicated services by detecting inconsistency among nodes in a timely manner via an inconsistency detection framework and resolving the detected inconsistencies efficiently when necessary. Through experimentation on Planet-Lab, IDEA is evaluated from two aspects: its adaptive interface and its performance of inconsistency resolution. Results show that IDEA achieves adaptability by adjusting the consistency level according to users. preference on-demand, and it achieves low inconsistency resolution delay and incurs minimal communication cost.

An analytical study of FairOM: a fair overlay multicast protocol for Internet-scale distributed systems

Fairness emerges as an important research issue in overlay multicast because spreading the multicasting load evenly among participants can eliminate potential traffic hot spots, thus improving the systems quality of service (QoS). In this paper, we analyze FairOM and compare it with non-FairOM approaches from two aspects: tree height and number of sessions that can be supported, which measure FairOM from a single-sessions and multiple-sessions point of view, respectively

FairOM: enforcing proportional contributions among peers in internet-scale distributed systems

The viability of overlay multicasting has been established by previous research. However, in order to apply overlay multicast to Internet-scale distributed systems, such as the Grid and Peer-to-Peer systems, the issue of effectively enforcing fairness among peers so as to optimize overall performance remains as a challenge. This paper argues that simply applying a multiple-tree scheme does not provide sufficient fairness, in terms of performance. Instead, we believe that a better way to define fairness, for performance’s sake, is to factor in peers’ proportional contributions as it provides the opportunity to support many simultaneous multicasting sessions. This paper then presents a protocol, called FairOM (Fair Overlay Multicast), to enforce proportional contributions among peers in Internet-scale distributed systems. By exploiting the notion of staged spare capacity group and deploying a two-phase multicast forest construction process, FairOM enforces proportional contributions among peers, which enables more simultaneous multicasting sessions and alleviates potential hot-spots. The simulation results of a large multicast group with 1000 members show that FairOM achieves the goal of enforcing proportional contributions among peers and does not overwhelm the peers, including the multicast source. FairOM also achieves low delay penalty for peers and high path diversity.

Accurate Performance Modeling and Guidance to the Adoption of an Inconsistency Detection Framework

With the increased popularity of replica-based services in distributed systems such as the Grid, consistency control among replicas becomes more and more important. To this end, IDF (inconsistency detection framework), a two-layered overlay-based architecture, has been proposed as a new way to solve this problem-instead of enforcing a predefined consistency control protocol, IDF detects inconsistency in a timely manner when it occurs and resolves it based on applicationspsila semantics.This paper presents a comprehensive analytical study of the two-layer detection mechanism. We develop an analytical model to characterize IDF and, based on this model, evaluating the successful rate of inconsistency detection within the top layer, which directly impacts the performance of IDF. We also derive a unified formula to model the performance of IDF with regard to a wide range of applications. Based on the modeling results, we illustrate how practitioners can use these quantitative insights to tune the IDF parameters for specific applications.