main

Recent Funding
	PI- National Science Foundation (NSF)-Award: CRI-1305258: 8/1/13 - 7/31/15, “Compute Infrastructure for Large-Scale Data Analytics”, $189,952.

Recent Research Projects
* InsightNotes: Large-Scale Annotation Management. * Redoop: Recurring Queries in MapReduce Infrastructure. * HandsOn DB: Managing Human-Involved Dependencies in RDBMS. * CoHadoop & E3: Hadoop-Based Query Optimizations.

My expertise that I am keen to bring to WPI and integrate into both the research and teaching activities fall under the broad areas of big data management and analytics. I enjoy system-oriented research in which my team builds prototype engines and techniques—possibly within larger systems such as relational database, Hadoop, and NoSQL engines—and deploy these prototypes to serve diverse real-world applications.

My research interests span the areas of big data infrastructres, scalable data analytics, query optimization, scalable data curation and metadata management, scientific data management, and scalable graph (network) analytics.

Specific areas of recent focus:

Next-Generation of Data Curation Engines:
Data curation and metadata management play significant roles in most modern scientific applications. The creation and maintenance of annotated and curated databases require a great deal of effort (and cost) from many scientists and domain experts. Yet, the gain from the maintained annotations is still very limited, and the virtue of the annotations' hidden knowledge is still uncharted. This is especially true because the growing volume, profound complexity, increasing heterogeneity, and hidden semantics of the emerging annotation repositories create unprecedented challenges to annotation management techniques.

In addition, the rapid evolution of the big data infrastructure, e.g., MapReduce, Spark, and NoSQL databases, creates numerous challenges and research opportunities for curation, annotation, and metadata management including distributed metadata processing, metadata migration across different infrastructures, provenance tracking through complex workflows over possibly several platforms, proactive curation techniques, and many others.

In my research, we rethink the whole design methodology and processing cycle of metadata management. We address various challenges including new metadata models, extensible semantic extraction and summarization, advanced query processing and manipulation, proactive and automated data curation, distributed metadata processing, and metadata quality and verification.

Related Papers: EDBT '16, VLDBJ '16, SIGMOD '15, EDBT '15, SIGMOD '14, EDBT '09

Big Data Management & Analytics:
Big data is one of the most critical and highly promising areas of both research and teaching. Almost all modern and emerging applications are collecting and/or generating massive volumes of data. Therefore, innovations in these applications is only hindered by their ability to analyze and discover knowledge from the data in a timely and scalable fashion. Big data warrant the need for new infrastructures, innovative optimizations, and efficient indexing techniques.

My team is exploring various optimizations and advances to big data infrastructures, and more specifically on Hadoop.

Related Papers: ICDE'17, PVLDB '16, SSDBM '16, VLDBJ '16, PVLDB '15, PVLDB '14, EDBT '14, EDBT '13, PVLDB '11