Da Silva wins two awards for dependability and event-based system papers

2/21/2013 Elise King, CSL Communications

University of Illinois graduate Gabriela Jacques da Silva won two prestigious awards this past summer for papers she wrote on computer dependability and event-based systems.

Written by Elise King, CSL Communications

University of Illinois graduate Gabriela Jacques da Silva won two prestigious awards this past summer for papers she wrote on computer dependability and event-based systems.

Gabriela Jacques da Silva
Gabriela Jacques da Silva
Gabriela Jacques da Silva

In June, da Silva won the William C. Carter Award at the IEEE/IFIP International Conference on Dependable Systems and Networks, or DSN conference, which was held in Hong Kong. Then in July she won the Best Paper Award at the ACM International Conference on Distributed Event-Based Systems, or DEBS conference, which was held in New York.

Da Silva won the Carter Award for her dissertation titled, “Modeling Stream Processing Applications for Dependability Evaluation,” which is about “a modeling framework that has an equivalent representation of the basic elements of a streaming application, occurrence of faults and error propagation,” she said. Da Silva used these basic elements to create a full streaming application, and with the resulting model a developer can make informed decisions regarding fault tolerance techniques for the given application, she said.

Streaming applications are paradigms that process live incoming data such as business feeds and Twitter feeds. Fault tolerance techniques refer to efforts made to ensure that an application operates properly in the event of a hardware or software failure.

The paper that da Silva co-authored with Buğra Gedik, Henrique Andrade, Kun-Lung Wu and Ravi Iyer for the DEBS conference is titled, “Fault Injection-based Assessment of Partial Fault Tolerance in Stream Processing Applications.” This paper is about a fault injection approach that imitates the effects that a given fault tolerance technique can have on a streaming application.

“During the experimentation phase, faults are injected into different parts of the application,” da Silva said. “We devised four different metrics to characterize how each part of the application behaves under faults.” With the metric results, da Silva said that the developer can understand which parts of the application are more sensitive to faults and need to have stronger protection.

Da Silva graduated from the University of Illinois at Urbana-Champaign in 2010 with a Ph.D. in Electrical and Computer Engineering. While attending the University she was part of CSL researcher Ravi Iyer’s DEPEND group, which researchers various aspects of fault-tolerant and secure systems such as hardware, operations and applications. She also said she focused her PhD on developing fault tolerance techniques for stream processing applications.

Da Silva has been attending the DSN conference almost every year since she started her Ph.D.; however, this was her first year attending the DEBS conference.

Prior to attending the University, da Silva received her bachelor’s degree in computer science from the Federal University of Santa Maria and her master’s degree also in computer science from the Federal University of Rio Grande do Sul; both schools are in Brazil.


Share this story

This story was published February 21, 2013.