Search this site

Navigation

          
    Chunqiang (CQ) Tang ,   Ph.D.   (唐春强)         
            (Pronounced "Choon-chyahn Tong")    

I am a computer scientist at IBM T.J. Watson Research Center. My research interests lie broadly in the fields of distributed systems, information retrieval, operating systems, computer networks, and services computing. I am a hands-on researcher and spend much of my time on actually building sophisticated real systems of product quality. Much of my research work has been incorporated into advanced commercial products. In addition to  the research work, I also serve IBM's research community as the Watson Chair of the Services Computing PIC (PIC is IBM's internal counterpart of ACM SIG).

Selected Publications (Full List, DBLP, Google Scholar)

USENIX'09
  • "Black-Box Performance Control for High-Volume Non-Interactive Systems"
    		•
    USENIX'09
  • "vPath: Precise Discovery of Request Processing Paths from Black-Box Observations of Thread and Network Activities"
    		•
    SIGIR'08
  • "On Iterative Intelligent Medical Search"
    		•
    SIGMOD'07
  • "Resource-Adaptive Real-Time New Event Detection"
    • WWW'07
  • "Answering Relationship Queries on the Web"
    • WWW'07
  • "A Scalable Application Placement Controller for Enterprise Data Centers"
    • SCC'06 Best Paper            
  • "A Distributed Service Management Infrastructure for Enterprise Data Centers Based on Peer-to-Peer Technology"
    • SIGMETRICS'05
  • "Low Traffic Overlay Networks with Large Routing Tables" 
    • SIGIR'04
  • "On Scaling Latent Semantic Indexing for Large Peer-to-Peer Systems"
    • NSDI'04
  • "Hybrid Global-Local Indexing for Efficient Peer-to-Peer Information Retrieval"
    • SIGCOMM'03
  • "Peer-to-Peer Information Retrieval Using Self-Organizing Semantic Overlay Networks"
    • HotOS'03
  • "Towards a Semantic-Aware File Store"
    		•
    ICDCS'03
  • "Efficient Distributed Shared State for Heterogeneous Machine Architectures"
    		•
    HotNets'02
  • "pSearch: Information Retrieval in Structured Overlays"
    		•
    ICPP'02 Best Paper
  • "Multi-level Shared State for Distributed Systems"

    • Impact of My Research on Commercial Products and Exploratory Systems

    • Peer-to-Peer Enterprise Middleware   (updated 04/2007)
      I built a scalable enterprise middleware based on the principles of P2P. This middleware has been shipped as a subcomponent of a leading IBM commercial product, running on AIX, HP-UX, Linux, Solaris, Windows, and z/OS. This middleware embraces the concept of P2P, but unlike previous structured or unstructured P2P networks, its algorithms are specifically designed for enterprise environments, and its functions go way beyond file sharing. Its implementation is closely related to (but goes beyond) our series of P2P papers. It has significant complexity (the core consists of 41,000 lines of product-quality Java code), and the same code base can run in both emulation mode and real deployment mode. This is probably the first ever P2P enterprise middleware shipped in a commercial product. In addition, it has been used in several exploratory systems at IBM Research.

    • Dynamic Resource Provisioning in Enterprise Data Centers   (updated 04/2007)
      An extended version of the application placement algorithm in our WWW'07 placement paper has been implemented and shipped as a subcomponent of a leading IBM commercial product. In addition, the algorithm has also been implemented in the Apertium machine translation platform by people working on a Google Summer of Code project for Apertium.

    • Streaming News Analysis   (updated 06/2007)
      The framework in our SIGMOD'07 paper for detecting first-story news has been implemented in an exploratory stream processing system, which was deemed "ready to make its way into the marketplace" in a New York Times report.

    • Streaming Event Processing for IT Monitoring and Management   (updated 03/2009)
      I developed the autonomic performance management solution in a leading commercial product for IT service monitoring and management. A distinguishing feature of this solution is that it is throughput centric, as opposed to typical response time centric solutions for Web servers. Part of this work was reported in our USENIX'09 "black-box performance control" paper.

    • Performance Healthcheck (updated 03/2009)
      I am a main contributor to the PARA-medic tool, which checks performance issues for enterprise IT systems. This tool has been successfully used in commercial engagements and also widely featured in news media, including CNN, Fox, MSNBC, Yahoo, AOL, etc. Both a video and a  high-level text description are available online. Related papers include our USENIX'09 vPath paper and ICWS'08 paper.