Michael Kaminsky

Biography

Michael Kaminsky is a senior research scientist in Intel Labs and an adjunct faculty member of the Computer Science Department at Carnegie Mellon University. He is currently part of the Intel Science and Technology Center for Visual Cloud Systems (ISTC-VCS), based in Pittsburgh, PA at Carnegie Mellon. Previously, Michael was part of the ISTC for Cloud Computing (ISTC-CC) and, before that, the Intel University Lablet at CMU. Michael joined Intel in Summer 2004 after completing his Ph.D. in Computer Science at MIT. He received a S.M. (Masters) from MIT in Spring 2000 and a B.S. from the University of California, Berkeley in Spring 1998. Michael is generally interested in computer science systems research, including distributed systems, networking, and operating systems. His recent research efforts have examined systems support for machine learning, memory efficiency, key-value storage, energy-efficiency, low-power servers, and cross-data center consistency.

Research

I am generally interested in computer science systems research. More specifically, I enjoy working on distributed systems, networking, operating systems and network/systems security. I am working on a handful of projects under the broad umbrella of Fast Array of Wimpy Nodes (FAWN), including systems challenges in machine learning, memory-efficient algorithms and data structures, key-value storage, RDMA, and cross-data center consistency. Much of the code from these projects is available in our group's github repositories. Previous projects include software building blocks for non-volatile RAM (e.g., PCM), Neighborhood-Aware Networking (NaN), CloudConnect, Reliable Email (Re:) and Data-Oriented Transfer (DOT). Most of my graduate work involved a secure, decentralized network filesystem called SFS.

My Ph.D. thesis involved two pieces of work. First, I worked on the problem of how to do user authentication in a global file system. See our SOSP paper for more details. Second, I worked on REX, a remote-execution extension to SFS which incorporates the flexible key management and security already available in the filesystem. This extension is integrated with the SFS agent architecture and the filesystem; we hope with REX+SFS to provide a complete distributed computing environment where moving between machines is transparent. See our USENIX paper for more details.

In summer 1998, I did some really interesting and fun work related to the Interactive Barney doll from Actimates—check out this page for more info.

Contact Information

Senior Research Scientist
Intel Science and Technology Center for Visual Cloud Systems (ISTC-VCS)
Intel Labs

Adjunct Research Scientist
Computer Science Department
Carnegie Mellon University

RMCIC Bldg, Carnegie Mellon University
4720 Forbes Avenue, Suite 4100
Pittsburgh, PA 15213

Tel: 412-268-5023
Email: michael dot e dot kaminsky at intel dot com
Email: kaminsky at cs dot cmu dot edu

Selected Publications

2018
  • Putting the "Micro" Back in Microservice. Sol Boucher, Anuj Kalia, David G. Andersen, Michael Kaminsky. To appear in 2018 USENIX Annual Technical Conference (USENIX ATC'18), Boston, MA, July 2018.
  • Mainstream: Dynamic Stem-Sharing for Multi-Tenant Video Processing. Angela Jiang, Daniel L.-K. Wong, Christopher Canel, Ishan Misra, Michael Kaminsky, Michael A. Kozuch, Padmanabhan Pillai, David G. Andersen, Gregory R. Ganger. To appear in 2018 USENIX Annual Technical Conference (USENIX ATC'18), Boston, MA, July 2018.
  • SuRF: Practical Range Query Filtering with Fast Succinct Tries. Awarded Best Paper! Huanchen Zhang, Hyeontaek Lim, Viktor Leis, David G. Andersen, Michael Kaminsky, Kimberly Keeton, Andrew Pavlo. To appear in ACM SIGMOD International Conference on Management of Data 2018 (SIGMOD'18), Houston, TX, June 2018. [PDF]
  • Building A Bw-Tree Takes More Than Just Buzz Words. Ziqi Wang, Andrew Pavlo, Hyeontaek Lim, Viktor Leis, Huanchen Zhang, Michael Kaminsky, David G. Andersen. To appear in ACM SIGMOD International Conference on Management of Data 2018 (SIGMOD'18), Houston, TX, June 2018. [PDF]
  • Picking Interesting Frames in Streaming Video. Christopher Canel, Thomas Kim, Giulio Zhou, Conglong Li, Hyeontaek Lim, David G. Andersen, Michael Kaminsky, Subramanya R. Dulloor. In SysML Conference 2018 (extended/poster abstract), Stanford, CA, February 2018. [PDF]
  • Efficient Multi-Tenant Inference on Video using Microclassifiers. Giulio Zhou, Thomas Kim, Christopher Canel, Conglong Li, Hyeontaek Lim, David G. Andersen, Michael Kaminsky, Subramanya R. Dulloor. In SysML Conference 2018 (extended/poster abstract), Stanford, CA, February 2018. [PDF]
  • Dynamic Stem-Sharing for Multi-Tenant Video Processing. Angela Jiang, Christopher Canel, Daniel Wong, Michael Kaminsky, Michael A. Kozuch, Padmanabhan Pillai, David G. Andersen, Gregory R. Ganger. In SysML Conference 2018 (extended/poster abstract), Stanford, CA, February 2018. [PDF]
2017
2016
2015
2014
2013
2012
  • Using Vector Interfaces to Deliver Millions of IOPS from a Networked Key-value Storage Server. Vijay Vasudevan, Michael Kaminsky, David G. Andersen. In ACM Symposium on Cloud Computing (SOCC'12), San Jose, CA, October 2012. [PDF]
  • FAWNSort: Energy-efficient Sorting of 10GB, 100GB, and 1TB. Padmanabhan Pillai, Michael Kaminsky, Michael A. Kozuch, David G. Andersen. Winner of 2012 10GB, 100GB, and 1TB JouleSort, Daytona and Indy categories. [PDF]
2011
2010
2009
2008
2007
2006
2004
  • User Authentication and Remote Execution Across Administrative Domains. Michael Kaminsky. Ph.D. Thesis, MIT, September 2004. [PDF]
  • REX: Secure, Extensible Remote Execution. Michael Kaminsky, Eric Peterson, Daniel B. Giffin, Kevin Fu, David Mazières, and M. Frans Kaashoek. In Proceedings of the 2004 USENIX Annual Technical Conference (USENIX'04), pages 199-212, Boston, Massachusetts, June 2004. [PDF]
2003
2002
  • Using SFS for a Secure Network File System. Kevin Fu, Michael Kaminsky, and David Mazières. In USENIX ;login:, December 2002. [PDF]
2000
  • Flexible Key Management with SFS Agents. Michael Kaminsky. Master's Thesis, MIT, May 2000. [PDF]
1999