Grant: Discovery Grants Program - Individual
University of Waterloo
$31,000.00
High-Coverage Performance Analysis of Modern Large-Scale Networks
EF 2023-24
Résumé
Networks have become a critical infrastructure for the day-to-day operation of modern society. Not only do we need networks to work in face of traffic surges and failures, we also need them to work "well", providing each application with a satisfactory level of latency, throughput, jitter, and loss. However, it is challenging to know, at design time, whether a network will perform well in production. This is because some of the important parameters that affect network performance, such as the characteristics of input traffic and failures, are only known at run time. The set of possible run-time scenarios is only growing with time as networks continue to grow in size and complexity to support more users and applications. The long-term goal of this proposal is to enable high-coverage proactive performance analysis of modern large-scale networks. This is motivated by two encouraging trends in recent years. First, formal methods tools and techniques have become increasingly efficient in exploring large state spaces and finding scenarios that violate desired properties. So, they are a promising candidate for analyzing network behavior under all possible run-time scenarios. Second, using high-level domain-specific languages to program networks has become more widespread. This provides an opportunity to significantly scale the analysis by doing it at a higher level of abstraction. Specifically, my team and I will pursue three main objectives. First, we will develop abstract but representative performance models. These models, expressed as logical formulas, should include enough details about the network so that their analysis leads to accurate insights about network performance while remaining tractable for analysis. Second, we will develop scalable tools and techniques which can analyze these models for networks with thousands of components under all possible run-time scenarios. Third, we will design performance-aware network programming platforms that can compile high-level network programs to low-level packet processing instructions with certain performance guarantees. This way, we can reason about their performance by analyzing the high-level program as opposed to low-level instructions or the underlying hardware. The proposed research will create tools and techniques that, compared to the status-quo, can analyze the network under a significantly larger set of run-time scenarios and detect those where the network cannot provide satisfactory performance. As such, it can help uncover performance problems before they go into production and affect live traffic. Such problems can significantly disrupt the operations of networked applications that Canadians and others in the world rely on in their jobs and day-to-day lives. Moreover, as networks grow in size and complexity, there is an increasing need to ensure their robustness through rigorous design and analysis, a role for which my students will be well-equipped by working on the proposed research.
Programme
Discovery Grants Program - Individual
Bénéficiaire
Emplacement
Ontario, Canada
Type de bénéficiaire
Academia
Type
Grant
Source
NSERC's Awards Data
Tel que publié
Application id
RGPIN-2023-03775
Name nom
Tahmasbi Arashloo, Mina
Department département
Cheriton School of Computer Science
Organization id
33
Institution établissement
University of Waterloo
Province en
Ontario
Province fr
Ontario
Country en
CANADA
Country fr
CANADA
Fiscal year exercice financier
2023
Competition year année de concours
2023
Award amount
31000
Program id
RGPIN
Program name en
Discovery Grants Program - Individual
Program name fr
Programme de subventions à la découverte - individuelles
Group en
DISCOVERY RESEARCH
Group fr
RECHERCHE AXÉE SUR LA DÉCOUVERTE
Committee code
1507
Committee name en
Computer Science
Committee name fr
Sciences informatiques
Area of application code
1207
Area of application group en
Advancement of knowledge
Area of application group fr
Avancement des connaissances
Area of application en
Information, computer and communication technologies
Area of application fr
Technologie de l'information, des ordinateurs et des réseaux de communications
Research subject code
9999
Research subject group en
Not available
Research subject group fr
Not available
Research subject en
Non disponible
Research subject fr
Non disponible
Application title
High-Coverage Performance Analysis of Modern Large-Scale Networks
Keyword
Network Algorithms; Network Performance; Network Quality of Service (QoS); Network Testing; Network Verification; Protocol Verification
Application summary
Networks have become a critical infrastructure for the day-to-day operation of modern society. Not only do we need networks to work in face of traffic surges and failures, we also need them to work "well", providing each application with a satisfactory level of latency, throughput, jitter, and loss. However, it is challenging to know, at design time, whether a network will perform well in production. This is because some of the important parameters that affect network performance, such as the characteristics of input traffic and failures, are only known at run time. The set of possible run-time scenarios is only growing with time as networks continue to grow in size and complexity to support more users and applications. The long-term goal of this proposal is to enable high-coverage proactive performance analysis of modern large-scale networks. This is motivated by two encouraging trends in recent years. First, formal methods tools and techniques have become increasingly efficient in exploring large state spaces and finding scenarios that violate desired properties. So, they are a promising candidate for analyzing network behavior under all possible run-time scenarios. Second, using high-level domain-specific languages to program networks has become more widespread. This provides an opportunity to significantly scale the analysis by doing it at a higher level of abstraction. Specifically, my team and I will pursue three main objectives. First, we will develop abstract but representative performance models. These models, expressed as logical formulas, should include enough details about the network so that their analysis leads to accurate insights about network performance while remaining tractable for analysis. Second, we will develop scalable tools and techniques which can analyze these models for networks with thousands of components under all possible run-time scenarios. Third, we will design performance-aware network programming platforms that can compile high-level network programs to low-level packet processing instructions with certain performance guarantees. This way, we can reason about their performance by analyzing the high-level program as opposed to low-level instructions or the underlying hardware. The proposed research will create tools and techniques that, compared to the status-quo, can analyze the network under a significantly larger set of run-time scenarios and detect those where the network cannot provide satisfactory performance. As such, it can help uncover performance problems before they go into production and affect live traffic. Such problems can significantly disrupt the operations of networked applications that Canadians and others in the world rely on in their jobs and day-to-day lives. Moreover, as networks grow in size and complexity, there is an increasing need to ensure their robustness through rigorous design and analysis, a role for which my students will be well-equipped by working on the proposed research.