Ravi B. Konuru

Thin locks: featherweight synchronization for Java

Language-supported synchronization is a source of serious performance problems in many Java programs. Even single-threaded applications may spend up to half their time performing useless synchronization due to the thread-safe nature of the Java libraries. We solve this performance problem with a new algorithm that allows lock and unlock operations to be performed with only a few machine instructions in the most common cases. Our locks only require a partial word per object, and were implemented without increasing object size. We present measurements from our implementation in the JDK 1.1.2 for AIX, demonstrating speedups of up to a factor of 5 in micro-benchmarks and up to a factor of 1.7 in real programs.

MetaFac: community discovery via relational hypergraph factorization

This paper aims at discovering community structure in rich media social networks, through analysis of time-varying, multi-relational data. Community structure represents the latent social context of user actions. It has important applications in information tasks such as search and recommendation. Social media has several unique challenges. (a) In social media, the context of user actions is constantly changing and co-evolving; hence the social context contains time-evolving multi-dimensional relations. (b) The social context is determined by the available system features and is unique in each social media website. In this paper we propose MetaFac (MetaGraph Factorization), a framework that extracts community structures from various social contexts and interactions. Our work has three key contributions: (1) metagraph, a novel relational hypergraph representation for modeling multi-relational and multi-dimensional social data; (2) an efficient factorization method for community extraction on a given metagraph; (3) an on-line method to handle time-varying relations through incremental metagraph factorization. Extensive experiments on real-world social data collected from the Digg social media website suggest that our technique is scalable and is able to extract meaningful communities based on the social media contexts. We illustrate the usefulness of our framework through prediction tasks. We outperform baseline methods (including aspect model and tensor analysis) by an order of magnitude.

Deterministic replay of distributed Java applications

Execution behavior of a Java application can be nondeterministic due to concurrent threads of execution, thread scheduling, and variable network delays. This nondeterminism in Java makes the understanding and debugging of multi-threaded distributed Java applications a difficult and a laborious process. It is well accepted that providing deterministic replay of application execution is a key step towards programmer productivity and program under-standing. Towards this goal, we developed a replay framework based on logical thread schedules and logical intervals. An application of this framework was previously published in the context of a system called Deja Vu that provides deterministic replay of multi-threaded Java programs on a single Java Virtual Machine (JVM). In contrast, this paper focuses on distributed Deja Vu that provides deterministic replay of distributed Java applications running on multiple JVMs. We describe the issues and present the design, implementation and preliminary performance results of distributed Deja Vu that supports both multi-threaded and distributed Java applications.

Cooperative middleware specialization for service oriented architectures

Service-oriented architectures (SOA) will provide the basis of thenext generation of distributed software systems, and have already gained enormous traction in the industry through an XML--based instantiation, Web services. A central aspect of SOAs is the looser coupling between applications (services) that is achieved when services publish their functional and non-functional behavioral characteristics in a standardized, machine readable format. In this paper we argue that in the basic SOA model access to metadata is too static and results in inflexible interactions between requesters and providers. We propose specific extensions to the SOA model to allow service providers and requestors to dynamically expose and negotiate their public behavior, resulting in the ability to specialize and optimize the middleware supporting an interaction. We introduce a middleware architecture supporting this extended SOA functionality, and describe a conformant implementation based on standard Web services middleware. Finally, we demonstrate the advantages of this approach with a detailed real world scenario.

Diversified ranking on large graphs: an optimization viewpoint

Diversified ranking on graphs is a fundamental mining task and has a variety of high-impact applications. There are two important open questions here. The first challenge is the measure - how to quantify the goodness of a given top-k ranking list that captures both the relevance and the diversity? The second challenge lies in the algorithmic aspect - how to find an optimal, or near-optimal, top-k ranking list that maximizes the measure we defined in a scalable way? In this paper, we address these challenges from an optimization point of view. Firstly, we propose a goodness measure for a given top-k ranking list. The proposed goodness measure intuitively captures both (a) the relevance between each individual node in the ranking list and the query; and (b) the diversity among different nodes in the ranking list. Moreover, we propose a scalable algorithm (linear wrt the size of the graph) that generates a provably near-optimal solution. The experimental evaluations on real graphs demonstrate its effectiveness and efficiency.

An information exploration tool for performance analysis of Java programs

The diagnosis of performance and memory problems can require the analysis of large and complex data sets describing a programs execution. An analysis tool must help the user both find the right organization of the data to uncover useful information, and work with the data through a lengthy and unpredictable discovery process. We present Jinsight EX, a tool for analyzing Java performance, that adopts techniques that have been successfully used to explore large data sets in other application domains, and adapts them specifically to the needs of program execution analysis. We introduce execution slices, a high-level organizing abstraction that the user may define and then easily reuse in various settings. We illustrate techniques that allow the user to perform a range of common analysis tasks and to structure a longer analysis process, using this abstraction. We present the tool, its implementation and initial experience of its use.

Community Discovery via Metagraph Factorization

This work aims at discovering community structure in rich media social networks through analysis of time-varying, multirelational data. Community structure represents the latent social context of user actions. It has important applications such as search and recommendation. The problem is particularly useful in the enterprise domain, where extracting emergent community structure on enterprise social media can help in forming new collaborative teams, in expertise discovery, and in the long term reorganization of enterprises based on collaboration patterns. There are several unique challenges: (a) In social media, the context of user actions is constantly changing and coevolving; hence the social context contains time-evolving multidimensional relations. (b) The social context is determined by the available system features and is unique in each social media platform; hence the analysis of such data needs to flexibly incorporate various system features. In this article we propose MetaFac (MetaGraph Factorization), a framework that extracts community structures from dynamic, multidimensional social contexts and interactions. Our work has three key contributions: (1) metagraph, a novel relational hypergraph representation for modeling multirelational and multidimensional social data; (2) an efficient multirelational factorization method for community extraction on a given metagraph; (3) an online method to handle time-varying relations through incremental metagraph factorization. Extensive experiments on real-world social data collected from an enterprise and the public Digg social media Web site suggest that our technique is scalable and is able to extract meaningful communities from social media contexts. We illustrate the usefulness of our framework through two prediction tasks: (1) in the enterprise dataset, the task is to predict users’ future interests on tag usage, and (2) in the Digg dataset, the task is to predict users’ future interests in voting and commenting on Digg stories. Our prediction significantly outperforms baseline methods (including aspect model and tensor analysis), indicating the promising direction of using metagraphs for handling time-varying social relational contexts.

On demand web-client technologies

This paper describes a comprehensive set of technologies that enables rich interaction paradigms for Web applications. These technologies improve the richness of user interfaces and the responsiveness of user interactions. Furthermore, they support disconnected or weakly connected modes of operation. The technologies can be used in conjunction with many Web browsers and client platforms, interacting with a J2EE™ server. The approach is based on projecting the server-side model-view-controller paradigm onto the client. This approach is firmly rooted in the Web paradigm and proposes a series of incremental extensions. Most of the described technologies have been adopted by IBM server (WebSphere®) and client products.

A programming framework for mobilizing enterprise applications

Mobile applications often need to synchronize their data with backend servers. Synchronization semantics have typically been set at the level of backend datastores. This is a major hindrance for enterprise applications that cannot assume a uniform storage model. In this paper, we present the SodaSync framework that provides a generic synchronization model for mobile enterprise applications that use heterogeneous backend stores. SodaSync exploits a unifying higher-level data model of service data objects (SDO) and introduces a persistence and synchronization framework for the model. It allows application programmers to express data and consistency requirements in terms of the SDO model and thereby emancipates them from the replication nuances of various backend stores. We present the major features of SodaSync, its architecture, and the status of our implementation.

Outline wizard: presentation composition and search

Presentation material is a commonly-performed task. Yet current tools provide inadequate support - search tools are unable to return individual slides, and the linear model employed by presentation creation tools lacks structure and context. We propose a novel method for presentation creation, implemented in a tool called Outline Wizard, which enables outline-based composition and search. An Outline Wizard user enters a hierarchically-structured outline of a presentation; using that structure, the tool extracts user requests to formulate contextual queries, matches them against presentations within a repository, taking into account both content and structures of the presentations, and presents the user with sets of slides that are appropriate for each outline topic. At the heart of Outline Wizard is an outline-based search technique, which conducts content search within the context derived from the hierarchical structures of both user requests and presentations. We present a heuristic outline-extraction technique, which is used to reverse engineer the structures of presentations, thereby making the structures available for our search engine. Evaluations show that the outline extraction technique and outline-based search both perform well, and that users report a satisfying experience when using Outline Wizard to compose presentations from libraries of existing material.