Ketan's Homepage

Ketan Bhardwaj

Research Scientist II
School of Computer Science
College of Computing
Georgia Institute of Technology

SCS / CoC / GT



I enjoy building systems that enable new use cases. My core research interests lie in the areas of computer systems which spans from distributed-decentralized to low-level operating system support, from functional correctness to performance and finally, to systems support for security. These come together in Edge Computing.

I co-lead edge computing research in Kernel Research Group at Georgia Tech. I decided to stay put at GT to continue my research after receiving my PhD in computer systems in December 2016.

I am always looking for talented students. If you also enjoy building systems, please contact me with your CV.

Recent Research Highlights

  • An NSF grant awarded as Co-PI for DEX: Decentralized Edge Exchange.
  • Couper presents a practical system to slice production-ready DNN models across edge and cloud (SEC'19).
  • AppSlicer featured on SCS Website Georgia Tech Researchers Create Tool to Slice Apps.
  • 2 projects funded by GTRC Center for development of applications for Internet of things (CDAIT)
    DIoTe: Decentralized IoT Ensembles using edge.
    Seamless interaction of IoT devices through edge computing.
  • Received the CISCO gift fund to support our research on edge computing based IoT DDoS Solution.
  • My interview on edge computing's impact on industry Peggy Smedley Show.
  • Granted Premier Membership for streaming video alliance(announcement).
  • Received NSF I-Corps grant to carry out customer discovery for dissertation work.

The goal of my research is to build practical, reliable, secure and high-performance systems, and to bridge all layers of the system stack from hardware platforms to applications. In my research group, we conduct systems research for a variety of computing platforms spanning from IoT devices to Virtual Reality to mobiles to large-scale data centers, and build systems that support and enable new and emerging data-intensive applications. Our research contributions have been published at top-tier systems, architecture, and security conferences, some of them are being transferred into products.

Our research is supported in part by NSF, CISCO, VMWare, HP, Intel.

Publications


Organizing Committee

  • Publication Co-Chair USENIX Annual Technical Conference (USENIX ATC), 2020

Program Committee

  • The 2019 European conference on security and privacy (EuroSPEC’19)
  • The USENIX Workshop on Hot Topics in Edge Computing (Usenix HotEdge '18)
  • IEEE Third International Conference on Fog and Mobile Edge Computing (FMEC 2018)
  • The 1st European Workshop on Security and Privacy in Fog and Edge Computing (SPIFEC 2017)
  • The 2nd IEEE International Conference on Fog and Mobile Edge Computing (FMEC 2017)
  • Transactions on Cyber-Physical Systems (TCPS-2016-0124)
  • International Workshop on the IoT for Healthcare colocated with IEEE smartcomp
  • The 1st Workshop on Middleware for Edge Clouds & Cloudlets co-located with ACM/IFOP/Usenix Middleware

  • Reviewer

  • Expert external reviewer USENIX Annual Technical Conference (USENIX ATC), 2018

  • Others

  • Streaming Video Alliance, Membership Grant




Ph.D students

  • Ke-Jou Hsu (Carol)
  • Harshit Daga
  • Misun Park
  • Jim Choncholas
  • Jin Heo

Master students

  • Vaibhav Bhosale

Alumni

  • Matt Saunders (MS, Now at )
  • Nikita Juneja (MS, Now at Apple)
  • Pragya Agarwal (MS, Now at Amazon)
  • Mangala Khandekar








Scheduled Talks

  • November (11/06):Harshit Daga

    Title:

    Abstract:TBD

    About:TBD


  • October (11/13):Chao

    Title:

    Abstract:TBD

    About:TBD


  • October (10/23): Seulbae Kim

    Title: Finding Semantic Bugs in File Systems with an Extensible Fuzzing Framework

    Abstract: File systems are too large to be bug free. Although hand-written test suites have been widely used to stress file systems, they can hardly keep up with the rapid increase in file system size and complexity, leading to new bugs being introduced and reported regularly. These bugs come in various flavors: simple buffer overflows to sophisticated semantic bugs. Although bug-specific checkers exist, they generally lack a way to explore file system states thoroughly. More importantly, no turnkey solution exists that unifies the checking effort of various aspects of a file system under one umbrella. In this talk, he will present the work that highlights the potential of applying fuzzing to find not just memory errors but, in theory, any type of file system bugs with an extensible fuzzing framework: Hydra. Hydra provides building blocks for file system fuzzing, including input mutators, feedback engines, a libOS-based executor, and a bug reproducer with test case minimization. As a result, developers only need to focus on building the core logic for finding bugs of their own interests. We showcase the effectiveness of Hydra with four checkers that hunt crash inconsistency, POSIX violations, logic assertion failures, and memory errors. So far, Hydra has discovered 95 new bugs in Linux file systems, including one in a verified file system (FSCQ), as well as four POSIX violations.

    About: Seulbae Kim received the M.S. and B.S. degrees in Computer Science and Engineering from Korea University, Seoul Korea, in 2018 and 2016. Currently, he is a 2nd year Ph.D. in the School of Computer Science at Georgia Institute of Technology. His research interests include systems security and automated vulnerability detection.


  • October (10/16):Sanidhya Kashyap

    Title: Scalable and Practical Locking With Shuffling

    Abstract:TBD

    About:TBD


  • September (09/11): Andrew McCrabb

    Title: Spatial Locality Accelerator for Dynamic Graph Applications

    Abstract: In this talk, I will discuss the benefits and challenges of applications that use dynamic graphs: vertex-edge graphs with structures that change over time. I will also explain the critical role of partitioning these graphs and how we can use hardware to efficiently partition dynamic graphs and accelerate the application's execution.

    Bio: Andrew McCrabb is a PhD student in Computer Science and Engineering at the University of Michigan. He received his undergraduate degree from Auburn University and is currently working alongside Dr. Valeria Bertacco on hardware algorithms and architectures for graph applications through the Applications Driving Architectures (ADA) research center.