PERFORMANCE ANALYSIS OF CREDIT-BASED FLOW CONTROL IN INFINIBAND INTERCONNECTION NETWORKS

2006 ◽  
Vol 07 (04) ◽  
pp. 535-548 ◽  
Author(s):  
Shihang Yan ◽  
Geyong Min ◽  
Irfan Awan
Keyword(s):  
Performance Analysis ◽  
Flow Control ◽  
Network Model ◽  
Interconnection Networks ◽  
High Speed ◽  
High Performance ◽  
Performance Metrics ◽  
Queueing Network ◽  
Design And Optimization ◽  
Control Scheme

Credit-based flow control scheme that can be used to support both end-to-end and link-level flow control is becoming increasingly popular in high speed system area networks (SAN), e.g. InfiniBand networks where multiple processor nodes and I/O devices are interconnected using switched point-to-point links. By virtue of such a scheme, the downstream node sends credits to the upstream node indicating the availability of buffer spaces. Upon receiving credits, the upstream node injects packets into the networks. Performance analysis of credit-based flow control scheme plays an important role for the design and optimization of InfiniBand interconnection networks which have been widely used in many high-performance cluster, Grid and P2P computing systems. This study develops a new queueing network model for performance evaluation of credit-based flow control in InfiniBand networks. The performance metrics to be derived include the mean queue length, throughput and response time of the system. Simulation experiments have been used to validate the accuracy of the queueing network model. Results obtained from the analytical model have showed that this model can effectively evaluate the performance of credit-based flow control in InfiniBand networks.

Reliability Analysis of Window Flow Control Based on Packet Transmission Interval with ECN Considering Packet Loss

2020 ◽  
Vol 28 (01) ◽  
pp. 2150001
Author(s):  
Mitsutaka Kimura ◽  
Mitsuhiro Imaizumi ◽  
Toshio Nakagawa
Keyword(s):  
Flow Control ◽  
Congestion Control ◽  
Packet Loss ◽  
High Performance ◽  
Window Size ◽  
Packet Transmission ◽  
Network Congestion ◽  
Constant Number ◽  
Control Scheme ◽  
Window Flow Control

This paper discusses the reliability model of a window flow control scheme using High-performance and Flexible Protocol (HpFP) with Explicit Congestion Notification (ECN) considering packet loss. HpFP is an important techniques as congestion control scheme in a radio environment and video stream communication. HpFP has the character that throughput is adjusted by changing a packet transmission interval. We have already discussed some reliability models of a window flow control scheme based on a packet transmission interval. In these models, if some packets has failed at a first-time transmission, the packet transmission interval is prolonged. On the other hand, the server checks the state of network congestion by ECN bit. That is, if ECN bit has been set during connection, a packet transmission interval is also prolonged. We consider an extended stochastic model of a window flow control scheme based on a packet transmission interval with ECN considering packet loss. That is, the server checks ECN bit during connection and if the server detects the network congestion, the server executes congestion control that a packet transmission interval is prolonged. Thereafter, if a constant number of the retransmission has failed, or a constant number of packets has failed, the server checks it again. We derive the mean time until packet transmissions succeed, and discuss analytically a window size which maximizes the amount of packets per unit of mean transmission time.

scholarly journals A Comprehensive Performance Analysis of Neurodegenerative diseases Incidence based on Epidemiological Study in the Female subjects over varied data

2021 ◽  
Vol 10 (2) ◽  
Author(s):  
Afreen Khan ◽  
Swaleha Zubair ◽  
Samreen Khan
Keyword(s):  
Machine Learning ◽  
Alzheimer’S Disease ◽  
Alzheimer's Disease ◽  
Performance Analysis ◽  
Neurodegenerative Diseases ◽  
High Performance ◽  
Performance Metrics ◽  
Mean Squared Error ◽  
Kappa Statistic ◽  
Comprehensive Performance

Neurodegenerative diseases such as Alzheimer’s disease and dementia are gradually becoming more prevalent chronic diseases, characterized by the decline in cognitive and behavioral symptoms. Machine learning is revolu-tionising almost all domains of our life, including the clinical system. The application of machine learning has the potential to enormously augment the reach of neurodegenerative care thus building it more proficient. Throughout the globe, there is a massive burden of Alzheimer’s and demen-tia cases; which denotes an exclusive set of difficulties. This provides us with an exceptional opportunity in terms of the impending convenience of data. Harnessing this data using machine learning tools and techniques, can put scientists and physicians in the lead research position in this area. The ob-jective of this study was to develop an efficient prognostic ML model with high-performance metrics to better identify female candidate subjects at risk of having Alzheimer’s disease and dementia. The study was based on two diverse datasets. The results have been discussed employing seven perfor-mance evaluation measures i.e. accuracy, precision, recall, F-measure, Re-ceiver Operating Characteristic (ROC) area, Kappa statistic, and Root Mean Squared Error (RMSE). Also, a comprehensive performance analysis has been carried out later in the study.

scholarly journals A Two-Level Flow Control Scheme for High Speed Networks

1994 ◽  
Vol 3 (3) ◽  
pp. 261-284 ◽  
Author(s):  
Fengmin Gong ◽  
Gurudatta Parulkar
Keyword(s):  
Flow Control ◽  
High Speed ◽  
High Speed Networks ◽  
Control Scheme
Sign in / Sign up

Export Citation Format

Share Document