ABSTRACT High performance computing clusters have been a critical resource for computational scie... more ABSTRACT High performance computing clusters have been a critical resource for computational science for over a decade and have more recently become integral to large-scale industrial analysis. Despite their well-specified components, the aggregate behavior of clusters is poorly understood. The difficulties arise from complicated interactions between cluster components during operation. These interactions have been studied by many researchers, some of whom have identified the need for holistic multi-scale modeling that simultaneously includes network level, operating system level, process level, and user level behaviors. Each of these levels presents its own modeling challenges, but the user level is the most complex due to the adaptability of human beings. In this vein, there are several major user modeling goals, namely descriptive modeling, predictive modeling and automated weakness discovery. This study shows how multi-agent techniques were used to simulate a large-scale computing cluster at each of these levels.
Proceedings of Spie the International Society For Optical Engineering, 2006
ABSTRACT There are an increasing number of ways optical network devices and IP routers can intera... more ABSTRACT There are an increasing number of ways optical network devices and IP routers can interact with each other during a network fault. To provide continuity of service, the interactions between each component in a network must be cooperative. Consequently, the effect of recovery processes cooperating are the network configurations that have certain structural relationships, which can be elaborated. A conflict detector can prove that service will be restored during a fault scenario by checking whether these structural properties hold. We are using simulation as a method to study the coordination of recovery strategies and whether different coordination strategies will achieve recovery goals attached to a network service. The network service carries a traffic stream, which is injected into and extracted from a network. For multilayer recovery to complete, the cumulative effect of device actions during a failure must be (1) a connected path between the endpoints of a service and (2) a flow traffic delivered to a destination at a quality that matches a service level agreement. We represent Optical and Multiprotocol Label Switching (MPLS) recovery actions as graph-maintenance operations that change the state of a digraph. For example, the actions of forwarding traffic between an access port and a trunk port and selecting traffic from a new trunk port and forwarding it to an access port can be modeled as a sequence of edge additions and deletions. The state of the digraph represents the current configuration of a multilayer network as actions of recovery are performed. In this paper, we define some structural properties that can be observed during a simulation as the network evolves to a final state from an initial state before a failure occurs.
GLOBECOM 97. IEEE Global Telecommunications Conference. Conference Record, 2000
Abstract There is an increasing demand for higher levels of network availability and reliability.... more Abstract There is an increasing demand for higher levels of network availability and reliability. Effective network monitoring is necessary to meet this demand. Whereas most of the network monitoring research to date has been focused on combining the information collected in a meaningful way, in this research we focus on processing the information collected before it is combined. We propose a change detection methodology for each measurement variable, where we can detect changes from the variable's usual behavior. ...
Proceedings of 1995 IEEE Workshop on Neural Networks for Signal Processing, 2000
To improve network management in today's increasingly complex communication networks, the au... more To improve network management in today's increasingly complex communication networks, the authors propose an intelligent monitoring hierarchy. The hierarchy is comprised of hidden Markov models (HMMs) and neural networks. As demonstrated on real network data, this hierarchy can detect abnormal behavior at high levels using only readily available low-level fault models. This allows the node to provide the network manager a complete picture of the nodes health
ABSTRACT High performance computing clusters have been a critical resource for computational scie... more ABSTRACT High performance computing clusters have been a critical resource for computational science for over a decade and have more recently become integral to large-scale industrial analysis. Despite their well-specified components, the aggregate behavior of clusters is poorly understood. The difficulties arise from complicated interactions between cluster components during operation. These interactions have been studied by many researchers, some of whom have identified the need for holistic multi-scale modeling that simultaneously includes network level, operating system level, process level, and user level behaviors. Each of these levels presents its own modeling challenges, but the user level is the most complex due to the adaptability of human beings. In this vein, there are several major user modeling goals, namely descriptive modeling, predictive modeling and automated weakness discovery. This study shows how multi-agent techniques were used to simulate a large-scale computing cluster at each of these levels.
Seamless Interconnection for Universal Services. Global Telecommunications Conference. GLOBECOM'99. (Cat. No.99CH37042), 1999
Automatic protection switching (APS) protocols assigned to different layers in wide-area networks... more Automatic protection switching (APS) protocols assigned to different layers in wide-area networks require interworking functionality in order to restore a wide variety of services and accommodate an evolving network infrastructure. Without some coordination between restoration mechanisms, an outage duration would be lengthened as methods assigned to different layers interfere with each other, and the network would be locked up in
NOMS 2002. IEEE/IFIP Network Operations and Management Symposium. ' Management Solutions for the New Communications World'(Cat. No.02CH37327), 2002
The plethora of new technologies and services such as MPLS, ATM, IP, SONET and WDM allows service... more The plethora of new technologies and services such as MPLS, ATM, IP, SONET and WDM allows services to be restored at different layers and at different costs. Restoration schemes at multiple layers might collide, causing a race condition, where restoration agents at different layers keep trying to establish a connection. We define a model for restoration mechanisms at different layers
ABSTRACT High performance computing clusters have been a critical resource for computational scie... more ABSTRACT High performance computing clusters have been a critical resource for computational science for over a decade and have more recently become integral to large-scale industrial analysis. Despite their well-specified components, the aggregate behavior of clusters is poorly understood. The difficulties arise from complicated interactions between cluster components during operation. These interactions have been studied by many researchers, some of whom have identified the need for holistic multi-scale modeling that simultaneously includes network level, operating system level, process level, and user level behaviors. Each of these levels presents its own modeling challenges, but the user level is the most complex due to the adaptability of human beings. In this vein, there are several major user modeling goals, namely descriptive modeling, predictive modeling and automated weakness discovery. This study shows how multi-agent techniques were used to simulate a large-scale computing cluster at each of these levels.
2011 IEEE Third Int'l Conference on Privacy, Security, Risk and Trust and 2011 IEEE Third Int'l Conference on Social Computing, 2011
Social networking websites have become a vital means of communication that can provide informatio... more Social networking websites have become a vital means of communication that can provide information on various topics. The real time nature of the information published on social networking websites coupled with their accessibility as a publishing platform make them a powerful tool for information gathering. Furthermore, many individuals utilize these sorts of platforms to share their knowledge and opinions with
ABSTRACT High performance computing clusters have been a critical resource for computational scie... more ABSTRACT High performance computing clusters have been a critical resource for computational science for over a decade and have more recently become integral to large-scale industrial analysis. Despite their well-specified components, the aggregate behavior of clusters is poorly understood. The difficulties arise from complicated interactions between cluster components during operation. These interactions have been studied by many researchers, some of whom have identified the need for holistic multi-scale modeling that simultaneously includes network level, operating system level, process level, and user level behaviors. Each of these levels presents its own modeling challenges, but the user level is the most complex due to the adaptability of human beings. In this vein, there are several major user modeling goals, namely descriptive modeling, predictive modeling and automated weakness discovery. This study shows how multi-agent techniques were used to simulate a large-scale computing cluster at each of these levels.
Proceedings of Spie the International Society For Optical Engineering, 2006
ABSTRACT There are an increasing number of ways optical network devices and IP routers can intera... more ABSTRACT There are an increasing number of ways optical network devices and IP routers can interact with each other during a network fault. To provide continuity of service, the interactions between each component in a network must be cooperative. Consequently, the effect of recovery processes cooperating are the network configurations that have certain structural relationships, which can be elaborated. A conflict detector can prove that service will be restored during a fault scenario by checking whether these structural properties hold. We are using simulation as a method to study the coordination of recovery strategies and whether different coordination strategies will achieve recovery goals attached to a network service. The network service carries a traffic stream, which is injected into and extracted from a network. For multilayer recovery to complete, the cumulative effect of device actions during a failure must be (1) a connected path between the endpoints of a service and (2) a flow traffic delivered to a destination at a quality that matches a service level agreement. We represent Optical and Multiprotocol Label Switching (MPLS) recovery actions as graph-maintenance operations that change the state of a digraph. For example, the actions of forwarding traffic between an access port and a trunk port and selecting traffic from a new trunk port and forwarding it to an access port can be modeled as a sequence of edge additions and deletions. The state of the digraph represents the current configuration of a multilayer network as actions of recovery are performed. In this paper, we define some structural properties that can be observed during a simulation as the network evolves to a final state from an initial state before a failure occurs.
GLOBECOM 97. IEEE Global Telecommunications Conference. Conference Record, 2000
Abstract There is an increasing demand for higher levels of network availability and reliability.... more Abstract There is an increasing demand for higher levels of network availability and reliability. Effective network monitoring is necessary to meet this demand. Whereas most of the network monitoring research to date has been focused on combining the information collected in a meaningful way, in this research we focus on processing the information collected before it is combined. We propose a change detection methodology for each measurement variable, where we can detect changes from the variable's usual behavior. ...
Proceedings of 1995 IEEE Workshop on Neural Networks for Signal Processing, 2000
To improve network management in today's increasingly complex communication networks, the au... more To improve network management in today's increasingly complex communication networks, the authors propose an intelligent monitoring hierarchy. The hierarchy is comprised of hidden Markov models (HMMs) and neural networks. As demonstrated on real network data, this hierarchy can detect abnormal behavior at high levels using only readily available low-level fault models. This allows the node to provide the network manager a complete picture of the nodes health
ABSTRACT High performance computing clusters have been a critical resource for computational scie... more ABSTRACT High performance computing clusters have been a critical resource for computational science for over a decade and have more recently become integral to large-scale industrial analysis. Despite their well-specified components, the aggregate behavior of clusters is poorly understood. The difficulties arise from complicated interactions between cluster components during operation. These interactions have been studied by many researchers, some of whom have identified the need for holistic multi-scale modeling that simultaneously includes network level, operating system level, process level, and user level behaviors. Each of these levels presents its own modeling challenges, but the user level is the most complex due to the adaptability of human beings. In this vein, there are several major user modeling goals, namely descriptive modeling, predictive modeling and automated weakness discovery. This study shows how multi-agent techniques were used to simulate a large-scale computing cluster at each of these levels.
Seamless Interconnection for Universal Services. Global Telecommunications Conference. GLOBECOM'99. (Cat. No.99CH37042), 1999
Automatic protection switching (APS) protocols assigned to different layers in wide-area networks... more Automatic protection switching (APS) protocols assigned to different layers in wide-area networks require interworking functionality in order to restore a wide variety of services and accommodate an evolving network infrastructure. Without some coordination between restoration mechanisms, an outage duration would be lengthened as methods assigned to different layers interfere with each other, and the network would be locked up in
NOMS 2002. IEEE/IFIP Network Operations and Management Symposium. ' Management Solutions for the New Communications World'(Cat. No.02CH37327), 2002
The plethora of new technologies and services such as MPLS, ATM, IP, SONET and WDM allows service... more The plethora of new technologies and services such as MPLS, ATM, IP, SONET and WDM allows services to be restored at different layers and at different costs. Restoration schemes at multiple layers might collide, causing a race condition, where restoration agents at different layers keep trying to establish a connection. We define a model for restoration mechanisms at different layers
ABSTRACT High performance computing clusters have been a critical resource for computational scie... more ABSTRACT High performance computing clusters have been a critical resource for computational science for over a decade and have more recently become integral to large-scale industrial analysis. Despite their well-specified components, the aggregate behavior of clusters is poorly understood. The difficulties arise from complicated interactions between cluster components during operation. These interactions have been studied by many researchers, some of whom have identified the need for holistic multi-scale modeling that simultaneously includes network level, operating system level, process level, and user level behaviors. Each of these levels presents its own modeling challenges, but the user level is the most complex due to the adaptability of human beings. In this vein, there are several major user modeling goals, namely descriptive modeling, predictive modeling and automated weakness discovery. This study shows how multi-agent techniques were used to simulate a large-scale computing cluster at each of these levels.
2011 IEEE Third Int'l Conference on Privacy, Security, Risk and Trust and 2011 IEEE Third Int'l Conference on Social Computing, 2011
Social networking websites have become a vital means of communication that can provide informatio... more Social networking websites have become a vital means of communication that can provide information on various topics. The real time nature of the information published on social networking websites coupled with their accessibility as a publishing platform make them a powerful tool for information gathering. Furthermore, many individuals utilize these sorts of platforms to share their knowledge and opinions with
Uploads
Papers by Cindy Hood