Canada Spends Logo

Explorez notre analyse du budget 2026

Grant: Canada Research Chairs

University of Toronto

$50,000.00

Systems Software

EF 2024-25

Résumé
Software systems are vital to almost every aspect of our lives today, from powering every Internet service like Google and Amazon to running our cars and phones. Unfortunately, failures inevitably occur. The software industry spends $244 billion on failure recovery annually, and programmers today spend 50% of their time diagnosing failures (known as debugging). Moreover, prolonged system outages are not only costly for service providers; they can have deadly consequences when they occur in safety critical systems in cars and aircrafts. Professor Ding Yuan's goal is to automate failure diagnosis, hence minimizing system downtime. If we take a broad perspective on the history of computer technology evolution, it is clear that while almost every other area has experienced fundamental advancements, failure diagnosis is an exception: it still largely remains a manual process. Therefore, this research is addressing one of the most fundamental and impactful problems in computer science and engineering. Specifically, this research proposes on-site, automated failure diagnosis. Today, diagnosis is performed off-site: when a failure occurs on a production system, the log data, which contains diagnostic information, is transmitted to backend servers before being analyzed by engineers. If the logs do not contain sufficient information for engineers to narrow down the root cause, they add logging code to the system to collect additional diagnostic information, and send this newly instrumented system to production so that it outputs more logs, which are then analyzed. This process repeats until the root cause is unambiguously inferred. This research aims to automate this diagnosis process, and design it to be highly efficient so that it can be performed at the same production environment. Doing so would eliminate communication overhead, enabling real-time failure diagnosis. In addition, it will eliminate the cost of storing the logs on the backend server. It also eliminates the need for engineers to replicate the production environment: in off-site diagnosis, engineers need to replicate the environment setup that is the same as the production site in order to reproduce the failure. Such an environment setup is often the most time-consuming process in diagnosis. On-site diagnosis eliminates such complications altogether.
Programme
Canada Research Chairs
Bénéficiaire
Emplacement
Ontario, Canada
Type de bénéficiaire
Academia
Type
Grant
Source
NSERC's Awards Data

Tel que publié

Application id
CRC-2023-00162
Name nom
Yuan, Ding
Department département
Electrical and Computer Engineering - University of Toronto
Organization id
31
Institution établissement
University of Toronto
Province en
Ontario
Province fr
Ontario
Country en
CANADA
Country fr
CANADA
Fiscal year exercice financier
2024
Competition year année de concours
2023
Award amount
50000
Program id
CRC
Program name en
Canada Research Chairs
Program name fr
Chaires de recherche du Canada
Group en
DISCOVERY RESEARCH
Group fr
RECHERCHE AXÉE SUR LA DÉCOUVERTE
Committee code
9998
Committee name en
Canada Research Chair Selection Committee
Committee name fr
Comité de sélection des Chaires de recherche du Canada
Area of application code
9999
Area of application group en
Not available
Area of application group fr
Non disponible
Area of application en
Not available
Area of application fr
Non disponible
Research subject code
9999
Research subject group en
Not available
Research subject group fr
Non disponible
Research subject en
Not available
Research subject fr
Non disponible
Field of research list codes
RDF1020303
Field of research list names en
Software domains (including operating systems and software infrastructure)
Field of research list names fr
Domaines logiciels (y compris les systèmes d'exploitation et l'infrastructure logicielle)
Socioeconomic objective list codes
RDS10610
Socioeconomic objective list names en
Information systems, technologies and services
Socioeconomic objective list names fr
Systèmes, technologies et services d'information
Application title
Systems Software
Keywords
Systems software; Operating systems; Software reliability; Failure diagnosis; Log compression; Log search; Software logging; Observability
Application summary
Software systems are vital to almost every aspect of our lives today, from powering every Internet service like Google and Amazon to running our cars and phones. Unfortunately, failures inevitably occur. The software industry spends $244 billion on failure recovery annually, and programmers today spend 50% of their time diagnosing failures (known as debugging). Moreover, prolonged system outages are not only costly for service providers; they can have deadly consequences when they occur in safety critical systems in cars and aircrafts. Professor Ding Yuan's goal is to automate failure diagnosis, hence minimizing system downtime. If we take a broad perspective on the history of computer technology evolution, it is clear that while almost every other area has experienced fundamental advancements, failure diagnosis is an exception: it still largely remains a manual process. Therefore, this research is addressing one of the most fundamental and impactful problems in computer science and engineering. Specifically, this research proposes on-site, automated failure diagnosis. Today, diagnosis is performed off-site: when a failure occurs on a production system, the log data, which contains diagnostic information, is transmitted to backend servers before being analyzed by engineers. If the logs do not contain sufficient information for engineers to narrow down the root cause, they add logging code to the system to collect additional diagnostic information, and send this newly instrumented system to production so that it outputs more logs, which are then analyzed. This process repeats until the root cause is unambiguously inferred. This research aims to automate this diagnosis process, and design it to be highly efficient so that it can be performed at the same production environment. Doing so would eliminate communication overhead, enabling real-time failure diagnosis. In addition, it will eliminate the cost of storing the logs on the backend server. It also eliminates the need for engineers to replicate the production environment: in off-site diagnosis, engineers need to replicate the environment setup that is the same as the production site in order to reproduce the failure. Such an environment setup is often the most time-consuming process in diagnosis. On-site diagnosis eliminates such complications altogether.