Difference between revisions of "CSE730x Research Seminar"
| Line 42: | Line 42: | ||
=== September 25, 2009 - Weijun (Vincent) Guo === | === September 25, 2009 - Weijun (Vincent) Guo === | ||
| − | <blockquote> | + | Paper: |
| − | + | <blockquote>Chen, Y., Gnawali, O., Kazandjieva, M., Levis, P., and Regehr, J. 2009. Surviving sensor network software faults. In Proceedings of the ACM SIGOPS 22nd Symposium on Operating Systems Principles (Big Sky, Montana, USA, October 11 - 14, 2009). SOSP '09. ACM, New York, NY, 235-246. DOI= http://doi.acm.org/10.1145/1629575.1629598 | |
| + | <blockquote> | ||
| − | |||
Abstract | Abstract | ||
| − | We describe Neutron, a version of the TinyOS operating system that efficiently recovers from memory safety bugs. | + | <blockquote> |
| − | Where existing schemes reboot an entire node on an error, | + | We describe Neutron, a version of the TinyOS operating system that efficiently recovers from memory safety bugs. Where existing schemes reboot an entire node on an error, Neutron's compiler and runtime extensions divide programs into recovery units and reboot only the faulting unit. The TinyOS kernel itself is a recovery unit: a kernel safety violation appears to applications as the processor being unavailable for 10-20 milliseconds. |
| − | into recovery units and reboot only the faulting unit. The TinyOS kernel itself is a recovery unit: a kernel safety violation | + | |
| − | appears to applications as the processor being unavailable for | + | Neutron further minimizes safety violation cost by supporting "precious" state that persists across reboots. Application data, time synchronization state, and routing tables can all be declared as precious. Neutron's reboot sequence conservatively checks that precious state is not the source of a fault before preserving it. Together, recovery units and precious state allow Neutron to reduce a safety violation's cost to time synchronization by 94% and to a routing protocol by 99.5%. Neutron also protects applications from losing data. Neutron provides this recovery on the very limited resources of a tiny, low-power microcontroller. |
| − | can all be declared as precious. | ||
| − | a fault before preserving it. Together, recovery units and precious state allow Neutron to reduce a safety | ||
| − | to time synchronization by 94% and to a routing protocol by 99 | ||
| − | Neutron provides this recovery on the very limited resources of a tiny, low-power microcontroller. | ||
</blockquote> | </blockquote> | ||
Revision as of 02:34, 24 December 2009
- Instructors: Chenyang Lu and Gruia-Catalin Roman
- Time: Friday at 2-3:30pm, Location: Bryan 509C
This seminar examines fundamental and emerging concepts in concurrency and distribution by studying seminal papers and recent research results. Broad topics of interest include models of concurrency, mobile computing, parallel architectures, sensor networks, distributed algorithms, and specialized protocols. Each semester, the seminar emphasizes different themes reflecting the current research interests of the participants.
The theme of this semester's seminar is Wireless Sensor Networks. We will read and discuss papers from recent major conferences on mobile, wireless, and sensor networks and systems. These conferences include:
- SenSys
- IPSN (ACM Archive)
- MobiCom (ACM archive)
- MobiSys (2006, ACM archive)
- NSDI (USENIX Website)
- SOSP
- OSDI (Archive website)
- RTSS (IEEE Archive)
- RTAS (IEEE Archive: part1 part2)
- SIGCOMM (Website)
When choosing a paper to present, you may look through the conferences mentioned above, or view the list of potential papers.
August 28, 2009 - Abu Sayeed Saifullah
WirelessHART
September 4, 2009 - Octav Chipara
Reliable Clinical Monitoring.
September 11, 2009 - Yong Fu
Feedback Thermal Control for Real-time Systems
September 18, 2009 - Chengjie Wu
Miroslav Pajic and Rahul Mangharam, Anti-Jamming for Embedded Wireless Networks, IPSN 09, Apr. 2009.
Paper: [1]
September 25, 2009 - Weijun (Vincent) Guo
Paper:
Chen, Y., Gnawali, O., Kazandjieva, M., Levis, P., and Regehr, J. 2009. Surviving sensor network software faults. In Proceedings of the ACM SIGOPS 22nd Symposium on Operating Systems Principles (Big Sky, Montana, USA, October 11 - 14, 2009). SOSP '09. ACM, New York, NY, 235-246. DOI= http://doi.acm.org/10.1145/1629575.1629598
Abstract
We describe Neutron, a version of the TinyOS operating system that efficiently recovers from memory safety bugs. Where existing schemes reboot an entire node on an error, Neutron's compiler and runtime extensions divide programs into recovery units and reboot only the faulting unit. The TinyOS kernel itself is a recovery unit: a kernel safety violation appears to applications as the processor being unavailable for 10-20 milliseconds.
Neutron further minimizes safety violation cost by supporting "precious" state that persists across reboots. Application data, time synchronization state, and routing tables can all be declared as precious. Neutron's reboot sequence conservatively checks that precious state is not the source of a fault before preserving it. Together, recovery units and precious state allow Neutron to reduce a safety violation's cost to time synchronization by 94% and to a routing protocol by 99.5%. Neutron also protects applications from losing data. Neutron provides this recovery on the very limited resources of a tiny, low-power microcontroller.
October 2, 2009 - Sisu Xi
October 9, 2009 - N/A
Canceled due to RTAS deadline.
October 16, 2009 - N/A
Fall Break
October 23, 2009 - Louis Thomas
- The Hitchhiker's Guide to Successful Wireless Sensor Network Deployments" by Barrenetxea et al.
October 30, 2009 - N/A
Canceled due to IPSN deadline
November 6, 2009 - Mo Sha
November 13, 2009 - Yong Fu
November 20, 2009 - Greg Hackmann
RACNet: A High-Fidelity Data Center Sensing Network. Chieh-Jan Mike Liang, Jie Liu, Liqian Luo, Andreas Terzis, Feng Zhao. SenSys 2009.
November 27, 2009 - N/A
Thanksgiving Break
December 4, 2009 - Chien-Liang Fok
Ramanathan, N., Schoellhammer, T., Kohler, E., Whitehouse, K., Harmon,
T., and Estrin, D. 2009. Suelo: human-assisted sensing for exploratory soil monitoring studies. In Proceedings of the 7th ACM Conference on Embedded Networked Sensor Systems (Berkeley, California, November 04 - 06, 2009). SenSys '09. ACM, New York, NY, 197-210. DOI=
December 11, 2009 - Abu Sayeed Saifullah
December 18, 2009 - N/A
Winter Break
Previous Semesters
