The Experts below are selected from a list of 12822 Experts worldwide ranked by ideXlab platform
Kamil Sarac - One of the best experts on this subject based on the ideXlab platform.
-
Estimating Network Layer Subnet Characteristics via Statistical Sampling
2012Co-Authors: M. Tozal, Kamil SaracAbstract:Network layer Internet topology consists of a set of routers connected to each other through Subnets. Recently, there has been a significant interest in studying topological characteristics of Subnets in addition to routers in the Internet. However, given the size of the Internet, constructing complete Subnet level topology maps is neither practical nor economical. A viable solution, then, is to sample Subnets in the target domain and estimate their global characteristics. In this study, we propose a sampling framework for Subnets; derive proper estimators for various Subnet characteristics including total number of Subnets, Subnet prefix length distribution, mean Subnet degree, and IP address utilization; and analyze the theoretical and empirical aspects of these estimators.
-
Networking (1) - Estimating network layer Subnet characteristics via statistical sampling
NETWORKING 2012, 2012Co-Authors: M. Tozal, Kamil SaracAbstract:Network layer Internet topology consists of a set of routers connected to each other through Subnets. Recently, there has been a significant interest in studying topological characteristics of Subnets in addition to routers in the Internet. However, given the size of the Internet, constructing complete Subnet level topology maps is neither practical nor economical. A viable solution, then, is to sample Subnets in the target domain and estimate their global characteristics. In this study, we propose a sampling framework for Subnets; derive proper estimators for various Subnet characteristics including total number of Subnets, Subnet prefix length distribution, mean Subnet degree, and IP address utilization; and analyze the theoretical and empirical aspects of these estimators.
-
Subnet level network topology mapping
International Performance Computing and Communications Conference, 2011Co-Authors: Engin M Tozal, Kamil SaracAbstract:Internet topology at the network layer consists of routers and Subnets, i.e., point-to-point or multi-access connections. Network measurement studies have focused on router level maps and derived characteristics of routers such as mean degree, degree distribution, clustering coefficient and betweenness. Considering the fact that Subnets are also important building blocks of the Internet topology, this paper introduces a complementary view of network topologies named Subnet level maps. Subnet level network topology maps represent Subnets as vertices and depict routers as links connecting the vertices/Subnets. Additionally, we introduce a tool, called exploreNET, for Subnet discovery. Although ExploreNET is based on the same principals as our recent work traceNET [21], it differs from traceNET in its utilization in various domains. Particularly, it allows us discover the underlying Subnet level topology map of a network rather than the map dictated by routing dynamics. Finally, we present an evaluation of exploreNET by using it to discover and analyze various Subnet characteristics including degree distribution, capacity distribution and utilization for six geographically disperse public Internet Service Providers (ISPs).
-
IPCCC - Subnet level network topology mapping
30th IEEE International Performance Computing and Communications Conference, 2011Co-Authors: M. Tozal, Kamil SaracAbstract:Internet topology at the network layer consists of routers and Subnets, i.e., point-to-point or multi-access connections. Network measurement studies have focused on router level maps and derived characteristics of routers such as mean degree, degree distribution, clustering coefficient and betweenness. Considering the fact that Subnets are also important building blocks of the Internet topology, this paper introduces a complementary view of network topologies named Subnet level maps. Subnet level network topology maps represent Subnets as vertices and depict routers as links connecting the vertices/Subnets. Additionally, we introduce a tool, called exploreNET, for Subnet discovery. Although ExploreNET is based on the same principals as our recent work traceNET [21], it differs from traceNET in its utilization in various domains. Particularly, it allows us discover the underlying Subnet level topology map of a network rather than the map dictated by routing dynamics. Finally, we present an evaluation of exploreNET by using it to discover and analyze various Subnet characteristics including degree distribution, capacity distribution and utilization for six geographically disperse public Internet Service Providers (ISPs).
Dhabaleswar K. Panda - One of the best experts on this subject based on the ideXlab platform.
-
performance modeling of Subnet management on fat tree infiniband networks using opensm
International Parallel and Distributed Processing Symposium, 2005Co-Authors: Abhinav Vishnu, Amith R. Mamidala, Hyun-wook Jin, Dhabaleswar K. PandaAbstract:InfiniBand is becoming increasingly popular in the area of cluster computing due to its open standard and high performance. Fat tree is a primary interconnection topology for building large scale InfiniBand clusters. Instead of using a shared bus approach, InfiniBand employs an arbitrary switched point-to-point topology. In order to manage the Subnet, InfiniBand specifies a basic management infrastructure responsible for discovery, configuration and maintaining the active state of the network. In the literature, simulation studies have been done on irregular topologies to characterize the Subnet management mechanism. However, there is no study to model Subnet management mechanism on regular topologies using actual implementations. In this paper, we take up the challenge of modeling Subnet management mechanism for fat tree InfiniBand networks using a popular Subnet manager OpenSM. We present the timings for various Subnet management phases namely topology discovery, path computation and path distribution for large scale fat tree InfiniBand Subnets and present basic performance evaluation on small scale InfiniBand cluster. We verify our model with the basic set of results obtained, and present the results for the model by varying different parameters on fat trees.
-
IPDPS - Performance modeling of Subnet management on fat tree InfiniBand networks using OpenSM
19th IEEE International Parallel and Distributed Processing Symposium, 1Co-Authors: Abhinav Vishnu, Amith R. Mamidala, Hyun-wook Jin, Dhabaleswar K. PandaAbstract:InfiniBand is becoming increasingly popular in the area of cluster computing due to its open standard and high performance. Fat tree is a primary interconnection topology for building large scale InfiniBand clusters. Instead of using a shared bus approach, InfiniBand employs an arbitrary switched point-to-point topology. In order to manage the Subnet, InfiniBand specifies a basic management infrastructure responsible for discovery, configuration and maintaining the active state of the network. In the literature, simulation studies have been done on irregular topologies to characterize the Subnet management mechanism. However, there is no study to model Subnet management mechanism on regular topologies using actual implementations. In this paper, we take up the challenge of modeling Subnet management mechanism for fat tree InfiniBand networks using a popular Subnet manager OpenSM. We present the timings for various Subnet management phases namely topology discovery, path computation and path distribution for large scale fat tree InfiniBand Subnets and present basic performance evaluation on small scale InfiniBand cluster. We verify our model with the basic set of results obtained, and present the results for the model by varying different parameters on fat trees.
Abhinav Vishnu - One of the best experts on this subject based on the ideXlab platform.
-
performance modeling of Subnet management on fat tree infiniband networks using opensm
International Parallel and Distributed Processing Symposium, 2005Co-Authors: Abhinav Vishnu, Amith R. Mamidala, Hyun-wook Jin, Dhabaleswar K. PandaAbstract:InfiniBand is becoming increasingly popular in the area of cluster computing due to its open standard and high performance. Fat tree is a primary interconnection topology for building large scale InfiniBand clusters. Instead of using a shared bus approach, InfiniBand employs an arbitrary switched point-to-point topology. In order to manage the Subnet, InfiniBand specifies a basic management infrastructure responsible for discovery, configuration and maintaining the active state of the network. In the literature, simulation studies have been done on irregular topologies to characterize the Subnet management mechanism. However, there is no study to model Subnet management mechanism on regular topologies using actual implementations. In this paper, we take up the challenge of modeling Subnet management mechanism for fat tree InfiniBand networks using a popular Subnet manager OpenSM. We present the timings for various Subnet management phases namely topology discovery, path computation and path distribution for large scale fat tree InfiniBand Subnets and present basic performance evaluation on small scale InfiniBand cluster. We verify our model with the basic set of results obtained, and present the results for the model by varying different parameters on fat trees.
-
IPDPS - Performance modeling of Subnet management on fat tree InfiniBand networks using OpenSM
19th IEEE International Parallel and Distributed Processing Symposium, 1Co-Authors: Abhinav Vishnu, Amith R. Mamidala, Hyun-wook Jin, Dhabaleswar K. PandaAbstract:InfiniBand is becoming increasingly popular in the area of cluster computing due to its open standard and high performance. Fat tree is a primary interconnection topology for building large scale InfiniBand clusters. Instead of using a shared bus approach, InfiniBand employs an arbitrary switched point-to-point topology. In order to manage the Subnet, InfiniBand specifies a basic management infrastructure responsible for discovery, configuration and maintaining the active state of the network. In the literature, simulation studies have been done on irregular topologies to characterize the Subnet management mechanism. However, there is no study to model Subnet management mechanism on regular topologies using actual implementations. In this paper, we take up the challenge of modeling Subnet management mechanism for fat tree InfiniBand networks using a popular Subnet manager OpenSM. We present the timings for various Subnet management phases namely topology discovery, path computation and path distribution for large scale fat tree InfiniBand Subnets and present basic performance evaluation on small scale InfiniBand cluster. We verify our model with the basic set of results obtained, and present the results for the model by varying different parameters on fat trees.
M. Tozal - One of the best experts on this subject based on the ideXlab platform.
-
Estimating Network Layer Subnet Characteristics via Statistical Sampling
2012Co-Authors: M. Tozal, Kamil SaracAbstract:Network layer Internet topology consists of a set of routers connected to each other through Subnets. Recently, there has been a significant interest in studying topological characteristics of Subnets in addition to routers in the Internet. However, given the size of the Internet, constructing complete Subnet level topology maps is neither practical nor economical. A viable solution, then, is to sample Subnets in the target domain and estimate their global characteristics. In this study, we propose a sampling framework for Subnets; derive proper estimators for various Subnet characteristics including total number of Subnets, Subnet prefix length distribution, mean Subnet degree, and IP address utilization; and analyze the theoretical and empirical aspects of these estimators.
-
Networking (1) - Estimating network layer Subnet characteristics via statistical sampling
NETWORKING 2012, 2012Co-Authors: M. Tozal, Kamil SaracAbstract:Network layer Internet topology consists of a set of routers connected to each other through Subnets. Recently, there has been a significant interest in studying topological characteristics of Subnets in addition to routers in the Internet. However, given the size of the Internet, constructing complete Subnet level topology maps is neither practical nor economical. A viable solution, then, is to sample Subnets in the target domain and estimate their global characteristics. In this study, we propose a sampling framework for Subnets; derive proper estimators for various Subnet characteristics including total number of Subnets, Subnet prefix length distribution, mean Subnet degree, and IP address utilization; and analyze the theoretical and empirical aspects of these estimators.
-
IPCCC - Subnet level network topology mapping
30th IEEE International Performance Computing and Communications Conference, 2011Co-Authors: M. Tozal, Kamil SaracAbstract:Internet topology at the network layer consists of routers and Subnets, i.e., point-to-point or multi-access connections. Network measurement studies have focused on router level maps and derived characteristics of routers such as mean degree, degree distribution, clustering coefficient and betweenness. Considering the fact that Subnets are also important building blocks of the Internet topology, this paper introduces a complementary view of network topologies named Subnet level maps. Subnet level network topology maps represent Subnets as vertices and depict routers as links connecting the vertices/Subnets. Additionally, we introduce a tool, called exploreNET, for Subnet discovery. Although ExploreNET is based on the same principals as our recent work traceNET [21], it differs from traceNET in its utilization in various domains. Particularly, it allows us discover the underlying Subnet level topology map of a network rather than the map dictated by routing dynamics. Finally, we present an evaluation of exploreNET by using it to discover and analyze various Subnet characteristics including degree distribution, capacity distribution and utilization for six geographically disperse public Internet Service Providers (ISPs).
Jorge Garcia-vidal - One of the best experts on this subject based on the ideXlab platform.
-
EuroNGI Workshop - Subnet formation and address allocation approach for a routing with Subnets scheme in MANETs
Wireless Systems and Network Architectures in Next Generation Internet, 2006Co-Authors: Johann López, José M. Barceló, Jorge Garcia-vidalAbstract:Due to MANET topological routing algorithms are not scalable respect to the number of nodes in the network, we evaluate the potential use of a Subnet structure in MANETs. We show scenarios in which a MANET with Subnet structure is applicable, present the main technical challenges for the application of this structure and show that a reduction of routing overhead from a factor of N2 to N2/k (being N the number of nodes and k the number of Subnets) is reachable. In this paper we propose a Subnet Formation and Address Allocation mechanism and evaluate analytically the influence of this proposal on the overhead generated. This solution is only one of the many challenges to fix for putting in use a MANET with Subnets structure.
-
Subnet formation and address allocation approach for a routing with Subnets scheme in MANETs
Lecture Notes in Computer Science, 2006Co-Authors: Johann López, José M. Barceló, Jorge Garcia-vidalAbstract:Due to MANET topological routing algorithms are not scalable respect to the number of nodes in the network, we evaluate the potential use of a Subnet structure in MANETs. We show scenarios in which a MANET with Subnet structure is applicable, present the main technical challenges for the application of this structure and show that a reduction of routing overhead from a factor of N 2 to N 2 /k (being N the number of nodes and k the number of Subnets) is reachable. In this paper we propose a Subnet Formation and Address Allocation mechanism and evaluate analytically the influence of this proposal on the overhead generated. This solution is only one of the many challenges to fix for putting in use a MANET with Subnets structure.
