XML and JSON have become the default formats to exchange the information for web application or within enterprises. Keyword Search over XML data has been motivated by the need to relieve users from writing difficult XQueries since otherwise users are required to know the complex XML schema. In existing XML keyword search techniques the XML nodes returned for a keyword query are the Lowest Common Ancestor (LCA) nodes for the query keywords. In this paper, we argue that the LCA based techniques still require users to be well versed with the XML schema and also the data to be able to obtain meaningful query results. To address these shortcomings, we present a novel system, Generic Keyword Search (GKS), for a given keyword query Q, instead of identifying (and returning information) only from LCA nodes, GKS returns ‘meaningful’ information from any XML node, which contains a subset of keywords in the search query Q. GKS response includes LCA nodes, if any, that would have been returned by LCA based techniques. GKS is also able to find highly relevant keywords and XML schema elements, deeper analytical insights called DI in the XML data in the context of the user query. DI enables users to navigate the XML data and to refine their queries even if they are not familiar with the data and the schema. Our experiments on real data sets show that GKS is able to return highly relevant responses to keyword queries efficiently.

Classical XML keyword search based on the Lowest Common Ancestor (LCA) framework requires users to be well versed with data and semantic relationships between the query keywords to extract meaningful response, restricting its applicability. GKS (Generic Keyword Search), on the other hand, allows users to browse and navigate XML data without such constraints. GKS enables discovery of deeper insights (DI) in the XML data, found in the context of the search results. Such insights not only expose patterns hidden in the search results but also help users tune their queries, thus enabling the navigation of complex XML repositories with ease. We further show how, for a search query, different insights can be discovered from the data by varying a single parameter.

Due to their real time nature, microblog streams are a rich source of dynamic information, for example, about emerging events . Existing techniques for discovering such events from a microblog stream in real time (such as Twitter trending topics), have several lacunae when used for discovering emerging events; extant graph based event detection techniques are not practical in microblog settings due to their complexity; and conventional techniques, which have been developed for blogs, web-pages, etc., involving the use of keyword search, are only useful for finding information about known events. Hence, in this paper, we present techniques to discover events that are unraveling in microblog message streams in real time so that such events can be reported as soon as they occur. We model the problem as discovering dense clusters in highly dynamic graphs. Despite many recent advances in graph analysis, ours is the first technique to identify dense clusters in massive and highly dynamic graphs in real time. Given the characteristics of microblog streams, in order to find clusters without missing any events, we propose and exploit a novel graph property which we call short-cycle property . Our algorithms find these clusters efficiently in spite of rapid changes to the microblog streams. Further we present a novel ranking function to identify the important events. Besides proving the correctness of our algorithms we show their practical utility by evaluating them using real world microblog data. These demonstrate our technique's ability to discover, with high precision and recall, emerging events in high intensity data streams in real time. Many recent web applications create data which can be represented as massive dynamic graphs. Our technique can be easily extended to discover, in real time, interesting patterns in such graphs.

We present a data-driven approach for the problem of approximate data collection in sensor networks within the disjoint clique framework. The objective is to obtain estimations of attribute values from sensor nodes within certain error bounds while spending minimal amount of energy. Chu et al. [3] showed that a disjoint clique model is a natural and effective way of exploiting temporal and spatial correlations in the attribute values. However their approach assumes the prior knowledge of probabilistic models representing the sensor data. This often seems to be a tedious task in real world deployments. This paper presents a data driven solution using disjoint clique model which does not assume any such prior knowledge. Using a preliminary experimental evaluation on a real life and a synthetic dataset, we show that the proposed data driven approach is feasible and is easy to deploy.

In enterprise environments, it is critical to find the root cause of performance failures as quickly as possible. More often, root cause for these failures are the change events that are done deliberately by system administrator to fix existing problems in the system. However, these changes may themselves manifest as faults with the change in operating conditions over time. Hence, time lag between the changes that manifest themselves as faults and resulting failures may be unbounded. Most existing approaches fail to find such root cause faults as they consider only time correlated symptom events and real root cause event(s) are not even considered. In this paper we present a novel approach to identify such changes, from the set of changes done over the time, which may be the root cause of current failures. Our system automatically associates failures with these changes without any time window constraint. The approach presented in this paper does not require existence of any baseline model. As per our understanding, ours is a first system that associates changes with current failures without these assumptions. We have implemented this system in a real life test bed and it shows promising results.

Data Integration engines increasingly need to provide sophisticated processing options for XML data. In the past, it was adequate for these engines to support basic shredding and XML generation capabilities. However, with the steady growth of XML in applications and databases, integration platforms need to provide more direct operations on XML as well as improve the scalability and efficiency of these operations. In this paper, we describe a robust and comprehensive framework for performing Extract-Transform-Load (ETL) of XML. This includes (i) full computational model and engine capabilities to perform these operations in an ETL flow, (ii) an approach to pushing down XML operations into a database engine capable of supporting XML processing, and (iii) methods to apply partitioning techniques to provide scalable, parallel processing for large XML documents. We describe experimental results showing the effectiveness of these techniques.

In modern enterprise system environment when systemspsila performance degrades, detecting the anomaly is a hard problem. In this replicated environment, there can be hundreds or even thousands of server nodes for a single application. These nodes have implicit as well as explicit interdependencies with each other. Further due to heterogeneous capacities of nodes in the cluster, same fault may produce vastly different effect on the monitored metrics of different nodes. In case of performance problem, finding faulty node(s) in this environment is tedious and time consuming exercise with constantly changing workload, topology and SLA requirements. In this paper we present a novel eigen space based technique to detect anomaly in enterprise environment without any extra monitoring overhead. We monitor certain metrics on each of the node in cluster which are available in enterprise environment. We need a small number of most recent samples of each of these monitored metrics as our only historical information. Our technique adapts well in dynamic conditions, simple to operate and in case of an anomaly, automatically produces a list of faulty node(s). We have implemented this method in a 3-tier cluster environment with total 13 nodes. We have tested our algorithm by introducing faults in front tier, middle tier and backend tier. Our method is always able to separate out faulty nodes with high accuracy and precision.

Key areas in managing e-commerce systems are problem prediction, root cause analysis, and automated problem remediation. Anticipating SLO violations by proactive problem determination (PD) is particularly important since it can significantly lower the business impact of application performance problems. The main contribution of this paper is to investigate proactive PD based on two important concepts: dependency graphs and dynamic runtime performance characteristics of resources that comprise an I/T environment. The authors show how one can calculate and use the contribution of all supporting resources for a transaction to the end-to-end SLO for that transaction. Higher order moments of these components' contributions are further tracked for proactive alerting. An important aspect of this process is the classification of user transactions based on the profile of their resource usage, enabling one to set appropriate thresholds for the different classes only. Combined with the complete or semi-complete dependency information, our approach confines the scope of potential root causes to a small set of components, thus enabling efficient performance problem anticipation and quick remediation.

Clustered enterprise middleware systems employing dynamic workload scheduling are susceptible to a variety of application malfunctions that can manifest themselves in a counterintuitive fashion and cause debilitating damage. Until now, diagnosing problems in that domain involves investigating log files and configuration settings and requires in-depth knowledge of the middleware architecture and application design. This paper presents a method for problem determination using change point detection techniques and problem signatures consisting of a combination of changes (or absence of changes) in different metrics. We implemented this approach on a clustered middleware system and applied it to the detection of the storm drain condition: a debilitating problem encountered in clustered systems with counterintuitive symptoms. Our experimental results show that the system detects 93% of storm drain faults with no false positives.

Managing the service level objectives (SLO) in environments that implement e-commerce systems is a challenging task. It typically involves a clear understanding of how user transactions are supported by the I/T resources that comprise the e-commerce system. This paper investigates a subset of this important management problem. Using transaction to resource dependencies, the authors show how one can experimentally calculate the extent to which supporting resources for a transaction contribute to the end-to-end SLOs for that transaction. An important aspect of this process is the classification of user transactions, based on the profile of their resource usage, enabling one to set appropriate thresholds for different classes. This approach is then used to aid the detection and remediation of application performance bottlenecks.

We propose a simple active queue management algorithm, LUBA (link utilization based approach), which identifies malicious flows based on instantaneous link utilization. The algorithm maintains mean arrival rate information for a small number of flows to classify malicious and non-malicious flows. Malicious flows are penalized in proportion to their excess inflow. We show that the algorithm gives good throughput with fair share between TCP and UDP sources. The algorithm shows jitter suppression properties, which can be used in per flow guaranteed services.

The growing popularity of e-business has stimulated web sites to evolve from static content servers to complex multi-tier systems built from heterogeneous server platforms. E-businesses now spend a large fraction of their IT budgets maintaining, troubleshooting, and optimizing these web sites. It has been shown that such system management activities may be simplified or automated to various extents if a dynamic dependency graph of the system were available. Currently, all known solutions to the dynamic dependency graph extraction problem are intrusive in nature, i.e. require modifications at application or middleware level. In this paper, we describe non-intrusive techniques based on data mining, which process existing monitoring data generated by server platforms to automatically extract the system component dependency graphs in multi-tier e-business platforms, without any additional application or system modification.