The Experts below are selected from a list of 81 Experts worldwide ranked by ideXlab platform
Mohamed Abid - One of the best experts on this subject based on the ideXlab platform.
-
SCPN modeling and comparative performance evaluation of bussed and crossbar-based Switched Fabric CAN
Computer Standards & Interfaces, 2014Co-Authors: Mohamed Mazouzi, Oussama Kallel, Salem Hasnaoui, Ihsen Ben Mbarek, Mohamed AbidAbstract:Abstract In recent years, there has been a growth in a category of performance-critical distributed embedded systems and applications. These systems gain complexity when they are equipped with many microcontrollers which oversee many electronic control units (ECU). High performance and predictability are the main criteria of choice for any large-scale networked system dependent on real-time data processing and analysis. Switched-Fabric networks can provide fast and highly scalable hardware solutions and are now being increasingly used in distributed systems. In this paper, SCPN models of Switched Fabric and bussed CAN network are presented using timed colored Petri nets. These models are then evaluated and verified for the desired properties using CPN Tools. Comparison is then applied to both of the models to extract information on network performance metrics.
-
Design, implementation and simulation of CAN-based crossbar Switched Fabric
2012 2nd IEEE International Conference on Parallel Distributed and Grid Computing, 2012Co-Authors: Mohamed Mazouzi, Salem Hasnaoui, Ihsen Ben Mbarek, Dhia Belhajali, Mohamed AbidAbstract:Nowadays, CAN bus, as a well-known, fieldbus, reaches its limits when dealing with sophisticated embedded systems that require reliability and high-speed data transfer. Switched Fabric can be a fast and reliable hardware solution for existing CAN-bus limits. In fact, it allows multiple packets to be transferred simultaneously ensuring sufficient bandwidth. To resolve this problem when keeping the use of CAN protocol, we remove the CAN access technique module from an existing VHDL implementation of a CAN controller component and we replace it by a crossbar Switched Fabric, implemented in VHDL also. In this paper, we designed a Crossbar Switched Fabric CAN based on VHDL modules and we simulated it under ModelSim tools and we give of proof-of-concept of a CAN controller based on crossbar Switched Fabric.
-
SCPN evaluation of a Switched Fabric CAN Network
2010Co-Authors: Mohamed Mazouzi, Oussama Kallel, Salem Hasnaoui, Mohamed AbidAbstract:Nowadays distributed computing in complex embedded systems gain complexity when they are equipped with many microcontrollers which oversee many Electronic Control Units (ECU). High performance and predictability are the prerequisites for any large-scale networked system dependent on real-time data processing and analysis. Switched-Fabric networks can provide fast and highly scalable hardware solutions and are now being increasingly used in distributed systems. In this paper, we proposed a Switched Fabric CAN Network based on CAN Controllers and switch Fabric. We have implemented in VHDL a CAN Controller and used an existing 8x8 Switch Fabric. We also evaluated the overall architecture using CPNTools. We separately have modeled the CAN Controller correspondently to its VHDL implementation and a SCPN model of the Switched Fabric CAN Network. Experiments are then applied to the model
Gregory Kerr - One of the best experts on this subject based on the ideXlab platform.
-
Dissecting a small InfiniBand application using the Verbs API. arXiv:1105.1827v2
2016Co-Authors: Gregory KerrAbstract:InfiniBand is a Switched Fabric interconnect. The InfiniBand specification does not define an API. However the OFED package, libibverbs, has become the default API on Linux and Solaris systems. Sparse documentation exists for the verbs API. The simplest InfiniBand program provided by OFED, ibv rc pingpong, is about 800 lines long. The semantics of using the verbs API for this program is not obvious to the first time reader. This paper will dissect the ibv rc pingpong program in an attempt to make clear to users how to interact with verbs. This work was motivated by an ongoing project to include direct InfiniBand support for the DMTCP checkpointing package [1].
-
Dissecting a Small InfiniBand Application Using the Verbs API
arXiv: Distributed Parallel and Cluster Computing, 2011Co-Authors: Gregory KerrAbstract:InfiniBand is a Switched Fabric interconnect. The InfiniBand specification does not define an API. However the OFED package, libibverbs, has become the default API on Linux and Solaris systems. Sparse documentation exists for the verbs API. The simplest InfiniBand program provided by OFED, ibv_rc_pingpong, is about 800 lines long. The semantics of using the verbs API for this program is not obvious to the first time reader. This paper will dissect the ibv_rc_pingpong program in an attempt to make clear to users how to interact with verbs.
Zhao Zhihong - One of the best experts on this subject based on the ideXlab platform.
-
Design and Implementation of HDFS over Infiniband with RDMA
Wired Wireless Internet Communication, 2013Co-Authors: Dong Buyun, Fang Pei, Luo Bin, Fu Xiao, Zhao ZhihongAbstract:Nowadays more and more data have been generated every day in some enterprises such as facebook and google. These data need to be collected and analyzed in time. So the speed of transmitting data must be very high and the latency must be very low. Hadoop is applied in these enterprises and they use several data centers to store and process these data. But if the amount of data is growing fast or we will use only one data center then the bandwidth of the Ethernet Hadoop Distributed File System (HDFS) using cannot meet the need. The bandwidth of the Ethernet is going to become the performance bottleneck of HDFS. In order to solve this problem we will introduce a relatively new Switched Fabric communication link–—Infiniband in this paper. Based on Infiniband we have designed a new communication mechanism of HDFS and implemented it by modifying the code of HDFS. We use remote direct memory access (RDMA) to send and receive data rather than socket. The new HDFS will not use original stream mode to transmit data. Instead it will dynamically expand buffer and use changeable threshold. In this way the new HDFS will make CPU idle and improve performance. Unlike IPoIB which only uses Infiniband hardware device, our optimized HDFS is not only based on Infiniband hardware but also changes the code of HDFS to use RDMA. Our HDFS uses socket to transmit control message and RDMA to transmit data to make full use of the bandwidth of Infiniband. So applying the Infiniband with RDMA network bandwidth has not been the performance bottleneck of HDFS any more. According to the experiment results we have found that the network bandwidth of HDFS over Infiniband is 60 percent higher than the Ethernet and our optimized HDFS has much better performance than the HDFS over the Ethernet. On the other hand, the performance of our HDFS is also higher than the one which only use IPoIB.
-
Design and Implementation of HDFS over Infiniband with RDMA
Wired Wireless Internet Communication, 2013Co-Authors: Dong Buyun, Fang Pei, Luo Bin, Fu Xiao, Zhao ZhihongAbstract:Nowadays more and more data have been generated every day in some enterprises such as facebook and google. These data need to be collected and analyzed in time. So the speed of transmitting data must be very high and the latency must be very low. Hadoop is applied in these enterprises and they use several data centers to store and process these data. But if the amount of data is growing fast or we will use only one data center then the bandwidth of the Ethernet Hadoop Distributed File System (HDFS) using cannot meet the need. The bandwidth of the Ethernet is going to become the performance bottleneck of HDFS. In order to solve this problem we will introduce a relatively new Switched Fabric communication link–—Infiniband in this paper. Based on Infiniband we have designed a new communication mechanism of HDFS and implemented it by modifying the code of HDFS. We use remote direct memory access (RDMA) to send and receive data rather than socket. The new HDFS will not use original stream mode to transmit data. Instead it will dynamically expand buffer and use changeable threshold. In this way the new HDFS will make CPU idle and improve performance. Unlike IPoIB which only uses Infiniband hardware device, our optimized HDFS is not only based on Infiniband hardware but also changes the code of HDFS to use RDMA. Our HDFS uses socket to transmit control message and RDMA to transmit data to make full use of the bandwidth of Infiniband. So applying the Infiniband with RDMA network bandwidth has not been the performance bottleneck of HDFS any more. According to the experiment results we have found that the network bandwidth of HDFS over Infiniband is 60 percent higher than the Ethernet and our optimized HDFS has much better performance than the HDFS over the Ethernet. On the other hand, the performance of our HDFS is also higher than the one which only use IPoIB.
José G. Delgado-frias - One of the best experts on this subject based on the ideXlab platform.
-
Fault Tolerant Interleaved Switching Fabrics For Scalable High-Performance Routers
IEEE Transactions on Parallel and Distributed Systems, 2007Co-Authors: José G. Delgado-friasAbstract:Scalable high-performance routers and switches are required to provide a larger number of ports, higher throughput, and good reliability. Most of today's routers and switches are implemented using single crossbar as the Switched Fabric. The single crossbar complexity increases at O(N2) in terms of crosspoint number, which might become unacceptable for scalability as the port number (N) increases. A delta class self-routing multistage interconnection network (MIN) with the complexity of O(N times log2N) has been widely used in the asynchronous transfer mode switches. However, the reduction of the crosspoint number results in considerable internal blocking. A number of scalable methods have been proposed to solve this problem. One of them uses more stages with recirculation architecture to reroute the deflected packets, which greatly increase the latency. In this paper, we propose an interleaved multistage switching Fabrics architecture and assess its throughput with an analytical model and simulations. We compare this novel scheme with some previous parallel architectures and show its benefits. From extensive simulations under different traffic patterns and fault models, our interleaved architecture achieves better performance than its counterpart of single panel Fabric. Our interleaved scheme achieves speedups (over the single panel Fabric) of 3.4 and 2.25 under uniform and hot-spot traffic patterns, respectively, at maximum load (p = 1). Moreover, the interleaved Fabrics show great tolerance against internal hardware failures.
G. Holt - One of the best experts on this subject based on the ideXlab platform.
-
Extending the SCA core framework inside the modem architecture of a software defined radio
IEEE Communications Magazine, 2004Co-Authors: Lee Pucker, G. HoltAbstract:Extension of the Software Communications Architecture (SCA) core framework inside of the modem architecture of a software defined radio requires special consideration for managing the non-CORBA-enabled devices. This is especially true of devices that maintain direct hardware connections between each other in support of the high-speed low-latency communications required by many waveforms. This article illustrates the application of the SCA core framework for these types of modem architectures, including aggregating devices in support of direct hardware interconnects between components and the incorporation of a Switched Fabric communications infrastructure within the overall modem architecture.
