The Experts below are selected from a list of 19443 Experts worldwide ranked by ideXlab platform

Rickard Holsmark - One of the best experts on this subject based on the ideXlab platform.

  • deadlock free routing algorithms for irregular Mesh Topology noc systems with rectangular regions
    Journal of Systems Architecture, 2008
    Co-Authors: Rickard Holsmark, Maurizio Palesi, Shashi Kumar
    Abstract:

    The simplicity of regular Mesh Topology Network on Chip (NoC) architecture leads to reductions in design time and manufacturing cost. A weakness of the regular shaped architecture is its inability to efficiently support cores of different sizes. A proposed way in literature to deal with this is to utilize the region concept, which helps to accommodate cores larger than the tile size in Mesh Topology NoC architectures. Region concept offers many new opportunities for NoC design, as well as provides new design issues and challenges. One of the most important among these is the design of an efficient deadlock free routing algorithm. Available adaptive routing algorithms developed for regular Mesh Topology cannot ensure freedom from deadlocks. In this paper, we list and discuss many new design issues which need to be handled for designing NoC systems incorporating cores larger than the tile size. We also present and compare two deadlock free routing algorithms for Mesh Topology NoC with regions. The idea of the first algorithm is borrowed from the area of fault tolerant networks, where a network Topology is rendered irregular due to faults in routers or links, and is adapted for the new context. We compare this with an algorithm designed using a methodology for design of application specific routing algorithms for communication networks. The application specific routing algorithm tries to maximize adaptivity by using static and dynamic communication requirements of the application. Our study shows that the application specific routing algorithm not only provides much higher adaptivity, but also superior performance as compared to the other algorithm in all traffic cases. But this higher performance for the second algorithm comes at a higher area cost for implementing network routers.

  • a methodology for design of application specific deadlock free routing algorithms for noc systems
    International Conference on Hardware Software Codesign and System Synthesis, 2006
    Co-Authors: Maurizio Palesi, Rickard Holsmark, Shashi Kumar, Vincenzo Catania
    Abstract:

    In this paper, we present a methodology to specialize the routing algorithm in routing table based NoC routers. It tries to maximize the communication performance while ensuring deadlock free routing for an application. We demonstrate through analysis that routing algorithms generated by our methodology have higher adaptiveness as compared to turn-model based deadlock free routing algorithms for a Mesh Topology NoC architecture. Performance evaluation is carried out by using a flit-accurate simulator on traffic scenarios generated by both synthetic and real applications. The routing algorithms generated by the proposed methodology achieve an improvement in delay close to 50% and 30% over deterministic XY routing algorithm and adaptive Odd-Even routing algorithm respectively.

  • deadlock free routing algorithms for Mesh Topology noc systems with regions
    Digital Systems Design, 2006
    Co-Authors: Rickard Holsmark, Maurizio Palesi, Shashi Kumar
    Abstract:

    Region concept helps to accommodate cores larger than the tile size in Mesh Topology NoC architectures. In addition, it offers many new opportunities for NoC design, as well as provides new design issues and challenges. The most important among these is the design of a deadlock free routing algorithm. In this paper, we present and compare two routing algorithms for Mesh Topology NoC with regions. The first algorithm is borrowed from the area of fault tolerant networks and is adapted for the NoC context. We compare this with an algorithm designed using a methodology for design of application specific routing algorithms for communication networks. Our study shows that the application specific routing algorithm not only provides much higher adaptivity, but also superior performance as compared to the other algorithm in all traffic cases.

  • a method for router table compression for application specific routing in Mesh Topology noc architectures
    Lecture Notes in Computer Science, 2006
    Co-Authors: Maurizio Palesi, Shashi Kumar, Rickard Holsmark
    Abstract:

    One way to specialize a general purpose multi-core chip built using NoC principles is to provide a mechanism to configure an application specific deadlock free routing algorithm in the underlying communication network. A table in every router, implemented using a writable memory, can provide a possibility of specializing the routing algorithm according to the application requirements. In such an implementation the cost (area) of the router will be proportional to the size of the routing table. In this paper, we propose a method to compress the routing table to reduce its size such that the resulting routing algorithm remains deadlock free as well as has high adaptivity. We demonstrate through simulation based evaluation that our application specific routing algorithm gives much higher performance, in terms of latency and throughput, as compared to general purpose algorithms for deadlock free routing. We also show that a table size of two entries for each output port gives performance within 3% of the uncompressed table.

Shashi Kumar - One of the best experts on this subject based on the ideXlab platform.

  • designing efficient source routing for Mesh Topology network on chip platforms
    Digital Systems Design, 2010
    Co-Authors: Saad Mubeen, Shashi Kumar
    Abstract:

    Efficient on-chip communication is very important for exploiting enormous computing power available on a multi-core chip. Network on Chip (NoC) has emerged as a competitive candidate for implementing on-chip communication. Routing algorithms significantly affect the performance of a NoC. Most of the existing NoC architectural proposals advocate distributed routing algorithms for building NoC platforms. Although source routing offers many advantages, researchers avoided it due to its apparent disadvantage of larger header size requirement that results in lower bandwidth utilization. In this paper we make a strong case for the use of source routing for NoCs, especially for platforms with small sizes and regular topologies. We present a methodology to compute application specific efficient paths for communication among cores with a high degree of load balancing. The methodology first selects the most appropriate deadlock free routing algorithm, from a set of routing algorithms, based on the application’s traffic patterns. Then the selected (possibly adaptive) routing algorithm is used to compute efficient static paths with the goal of link load balancing. We demonstrate through simulation based evaluation that source routing has a potential of achieving higher performance, for example up to 28% lower latency even at medium load, as compared to distributed routing. A simple scheme is proposed for encoding of router ports to reduce the header overhead. A generic simulator was developed for evaluation and performance comparison between source routing and distributed routing. We also designed a router to support source routing for Mesh Topology NoC platforms.

  • deadlock free routing algorithms for irregular Mesh Topology noc systems with rectangular regions
    Journal of Systems Architecture, 2008
    Co-Authors: Rickard Holsmark, Maurizio Palesi, Shashi Kumar
    Abstract:

    The simplicity of regular Mesh Topology Network on Chip (NoC) architecture leads to reductions in design time and manufacturing cost. A weakness of the regular shaped architecture is its inability to efficiently support cores of different sizes. A proposed way in literature to deal with this is to utilize the region concept, which helps to accommodate cores larger than the tile size in Mesh Topology NoC architectures. Region concept offers many new opportunities for NoC design, as well as provides new design issues and challenges. One of the most important among these is the design of an efficient deadlock free routing algorithm. Available adaptive routing algorithms developed for regular Mesh Topology cannot ensure freedom from deadlocks. In this paper, we list and discuss many new design issues which need to be handled for designing NoC systems incorporating cores larger than the tile size. We also present and compare two deadlock free routing algorithms for Mesh Topology NoC with regions. The idea of the first algorithm is borrowed from the area of fault tolerant networks, where a network Topology is rendered irregular due to faults in routers or links, and is adapted for the new context. We compare this with an algorithm designed using a methodology for design of application specific routing algorithms for communication networks. The application specific routing algorithm tries to maximize adaptivity by using static and dynamic communication requirements of the application. Our study shows that the application specific routing algorithm not only provides much higher adaptivity, but also superior performance as compared to the other algorithm in all traffic cases. But this higher performance for the second algorithm comes at a higher area cost for implementing network routers.

  • a methodology for design of application specific deadlock free routing algorithms for noc systems
    International Conference on Hardware Software Codesign and System Synthesis, 2006
    Co-Authors: Maurizio Palesi, Rickard Holsmark, Shashi Kumar, Vincenzo Catania
    Abstract:

    In this paper, we present a methodology to specialize the routing algorithm in routing table based NoC routers. It tries to maximize the communication performance while ensuring deadlock free routing for an application. We demonstrate through analysis that routing algorithms generated by our methodology have higher adaptiveness as compared to turn-model based deadlock free routing algorithms for a Mesh Topology NoC architecture. Performance evaluation is carried out by using a flit-accurate simulator on traffic scenarios generated by both synthetic and real applications. The routing algorithms generated by the proposed methodology achieve an improvement in delay close to 50% and 30% over deterministic XY routing algorithm and adaptive Odd-Even routing algorithm respectively.

  • deadlock free routing algorithms for Mesh Topology noc systems with regions
    Digital Systems Design, 2006
    Co-Authors: Rickard Holsmark, Maurizio Palesi, Shashi Kumar
    Abstract:

    Region concept helps to accommodate cores larger than the tile size in Mesh Topology NoC architectures. In addition, it offers many new opportunities for NoC design, as well as provides new design issues and challenges. The most important among these is the design of a deadlock free routing algorithm. In this paper, we present and compare two routing algorithms for Mesh Topology NoC with regions. The first algorithm is borrowed from the area of fault tolerant networks and is adapted for the NoC context. We compare this with an algorithm designed using a methodology for design of application specific routing algorithms for communication networks. Our study shows that the application specific routing algorithm not only provides much higher adaptivity, but also superior performance as compared to the other algorithm in all traffic cases.

  • a method for router table compression for application specific routing in Mesh Topology noc architectures
    Lecture Notes in Computer Science, 2006
    Co-Authors: Maurizio Palesi, Shashi Kumar, Rickard Holsmark
    Abstract:

    One way to specialize a general purpose multi-core chip built using NoC principles is to provide a mechanism to configure an application specific deadlock free routing algorithm in the underlying communication network. A table in every router, implemented using a writable memory, can provide a possibility of specializing the routing algorithm according to the application requirements. In such an implementation the cost (area) of the router will be proportional to the size of the routing table. In this paper, we propose a method to compress the routing table to reduce its size such that the resulting routing algorithm remains deadlock free as well as has high adaptivity. We demonstrate through simulation based evaluation that our application specific routing algorithm gives much higher performance, in terms of latency and throughput, as compared to general purpose algorithms for deadlock free routing. We also show that a table size of two entries for each output port gives performance within 3% of the uncompressed table.

Maurizio Palesi - One of the best experts on this subject based on the ideXlab platform.

  • deadlock free routing algorithms for irregular Mesh Topology noc systems with rectangular regions
    Journal of Systems Architecture, 2008
    Co-Authors: Rickard Holsmark, Maurizio Palesi, Shashi Kumar
    Abstract:

    The simplicity of regular Mesh Topology Network on Chip (NoC) architecture leads to reductions in design time and manufacturing cost. A weakness of the regular shaped architecture is its inability to efficiently support cores of different sizes. A proposed way in literature to deal with this is to utilize the region concept, which helps to accommodate cores larger than the tile size in Mesh Topology NoC architectures. Region concept offers many new opportunities for NoC design, as well as provides new design issues and challenges. One of the most important among these is the design of an efficient deadlock free routing algorithm. Available adaptive routing algorithms developed for regular Mesh Topology cannot ensure freedom from deadlocks. In this paper, we list and discuss many new design issues which need to be handled for designing NoC systems incorporating cores larger than the tile size. We also present and compare two deadlock free routing algorithms for Mesh Topology NoC with regions. The idea of the first algorithm is borrowed from the area of fault tolerant networks, where a network Topology is rendered irregular due to faults in routers or links, and is adapted for the new context. We compare this with an algorithm designed using a methodology for design of application specific routing algorithms for communication networks. The application specific routing algorithm tries to maximize adaptivity by using static and dynamic communication requirements of the application. Our study shows that the application specific routing algorithm not only provides much higher adaptivity, but also superior performance as compared to the other algorithm in all traffic cases. But this higher performance for the second algorithm comes at a higher area cost for implementing network routers.

  • a methodology for design of application specific deadlock free routing algorithms for noc systems
    International Conference on Hardware Software Codesign and System Synthesis, 2006
    Co-Authors: Maurizio Palesi, Rickard Holsmark, Shashi Kumar, Vincenzo Catania
    Abstract:

    In this paper, we present a methodology to specialize the routing algorithm in routing table based NoC routers. It tries to maximize the communication performance while ensuring deadlock free routing for an application. We demonstrate through analysis that routing algorithms generated by our methodology have higher adaptiveness as compared to turn-model based deadlock free routing algorithms for a Mesh Topology NoC architecture. Performance evaluation is carried out by using a flit-accurate simulator on traffic scenarios generated by both synthetic and real applications. The routing algorithms generated by the proposed methodology achieve an improvement in delay close to 50% and 30% over deterministic XY routing algorithm and adaptive Odd-Even routing algorithm respectively.

  • deadlock free routing algorithms for Mesh Topology noc systems with regions
    Digital Systems Design, 2006
    Co-Authors: Rickard Holsmark, Maurizio Palesi, Shashi Kumar
    Abstract:

    Region concept helps to accommodate cores larger than the tile size in Mesh Topology NoC architectures. In addition, it offers many new opportunities for NoC design, as well as provides new design issues and challenges. The most important among these is the design of a deadlock free routing algorithm. In this paper, we present and compare two routing algorithms for Mesh Topology NoC with regions. The first algorithm is borrowed from the area of fault tolerant networks and is adapted for the NoC context. We compare this with an algorithm designed using a methodology for design of application specific routing algorithms for communication networks. Our study shows that the application specific routing algorithm not only provides much higher adaptivity, but also superior performance as compared to the other algorithm in all traffic cases.

  • a method for router table compression for application specific routing in Mesh Topology noc architectures
    Lecture Notes in Computer Science, 2006
    Co-Authors: Maurizio Palesi, Shashi Kumar, Rickard Holsmark
    Abstract:

    One way to specialize a general purpose multi-core chip built using NoC principles is to provide a mechanism to configure an application specific deadlock free routing algorithm in the underlying communication network. A table in every router, implemented using a writable memory, can provide a possibility of specializing the routing algorithm according to the application requirements. In such an implementation the cost (area) of the router will be proportional to the size of the routing table. In this paper, we propose a method to compress the routing table to reduce its size such that the resulting routing algorithm remains deadlock free as well as has high adaptivity. We demonstrate through simulation based evaluation that our application specific routing algorithm gives much higher performance, in terms of latency and throughput, as compared to general purpose algorithms for deadlock free routing. We also show that a table size of two entries for each output port gives performance within 3% of the uncompressed table.

Dimas Priyambodho - One of the best experts on this subject based on the ideXlab platform.

  • Analisis Kinerja EIGRP dan OSPF pada Topologi Ring dan Mesh
    Teknik Elektro Institut Teknologi Nasional Bandung, 2016
    Co-Authors: Dwi Aryanta, Arsyad Ramadhan Darlis, Dimas Priyambodho
    Abstract:

    ABSTRAK EIGRP (Enhanced Interior Gateway Routing Protocol) dan OSPF (Open Shortest Path Fisrt) adalah routing protokol yang banyak digunakan pada suatu jaringan komputer. EIGRP hanya dapat digunakan pada perangkat Merk CISCO, sedangkan OSPF dapat digunakan pada semua merk jaringan. Pada penelitian ini dibandingkan delay dan rute dari kedua routing protokol yang diimplementasikan pada topologi Ring dan Mesh. Cisco Packet Tracer 5.3 digunakan untuk mensimulasikan kedua routing protokol ini. Skenario pertama adalah perancangan jaringan kemudian dilakukan pengujian waktu delay 100 kali dalam 5 kasus. Skenario kedua dilakukan pengujian trace route untuk mengetahui jalur yang dilewati paket data lalu memutus link utama. Pada skenario kedua juga dilakukan perbandingan nilai metric dan cost hasil simulasi dengan perhitungan rumus. Skenario ketiga dilakukan pengujian waktu konvergensi untuk setiap routing protokol pada setiap topologi. Hasilnya EIGRP lebih cepat 386 µs daripada OSPF untuk topologi Ring sedangkan OSPF lebih cepat 453 µs daripada EIGRP untuk topologi Mesh. Hasil trace route menunjukan rute yang dipilih oleh routing protokol yaitu nilai metric dan cost yang terkecil. Waktu konvergensi rata-rata topologi Ring pada EIGRP sebesar 12,75 detik dan 34,5 detik pada OSPF sedangkan topologi Mesh di EIGRP sebesar 13 detik dan 35,25 detik di OSPF. Kata Kunci : EIGRP, OSPF, Packet Tracer 5.3, Ring, Mesh, Konvergensi ABSTRACT EIGRP (Enhanced Interior Gateway Routing Protocol) and OSPF (Open Shortest Path Fisrt) is the routing protocol that is widely used in a computer network. EIGRP can only be used on devices Brand CISCO, while OSPF can be used on all brands of network. In this study comparison of both the delay and the routing protocol implemented on Ring and Mesh Topology. Cisco Packet Tracer 5.3 is used to simulate both the routing protocol. The first scenario is the design of the network and then do the test of time delay 100 times in 5 cases. The second scenario tested trace route to determine the path of the data packet and then disconnect the main link. In the second scenario also conducted a cost comparison of metrics and the simulation results with the calculation formula. The third scenario testing time for each routing protocol convergence on any Topology. The result EIGRP faster than 386 microseconds for a ring Topology while OSPF OSPF 453 microseconds faster than EIGRP for Mesh Topology. The results showed trace route chosen by the routing protocol metric value and cost is the smallest. Average convergence time in the EIGRP Topology Ring of 12.75 seconds and 34.5 seconds, while the Mesh Topology in an OSPF EIGRP for 13 seconds and 35.25 seconds in OSPF. Keywords : EIGRP,OSPF, Packet Tracer 5.3, Ring, Mesh, Convergence

  • Analisis Kinerja EIGRP dan OSPF pada Topologi Ring dan Mesh
    Teknik Elektro Institut Teknologi Nasional Bandung, 2014
    Co-Authors: Dwi Aryanta, Arsyad Ramadhan Darlis, Dimas Priyambodho
    Abstract:

    ABSTRAK EIGRP (Enhanced Interior Gateway Routing Protocol) dan OSPF (Open Shortest Path Fisrt) adalah routing protokol yang banyak digunakan pada suatu jaringan komputer. EIGRP hanya dapat digunakan pada perangkat Merk CISCO, sedangkan OSPF dapat digunakan pada semua merk jaringan. Pada penelitian ini dibandingkan delay dan rute dari kedua routing protokol yang diimplementasikan pada topologi Ring dan Mesh. Cisco Packet Tracer 5.3 digunakan untuk mensimulasikan kedua routing protokol ini. Skenario pertama adalah perancangan jaringan kemudian dilakukan pengujian waktu delay 100 kali dalam 5 kasus. Skenario kedua dilakukan pengujian trace route untuk mengetahui jalur yang dilewati paket data lalu memutus link utama. Pada skenario kedua juga dilakukan perbandingan nilai metric dan cost hasil simulasi dengan perhitungan rumus. Skenario ketiga dilakukan pengujian waktu konvergensi untuk setiap routing protokol pada setiap topologi. Hasilnya EIGRP lebih cepat 386 µs daripada OSPF untuk topologi Ring sedangkan OSPF lebih cepat 453 µs daripada EIGRP untuk topologi Mesh. Hasil trace route menunjukan rute yang dipilih oleh routing protokol yaitu nilai metric dan cost yang terkecil. Waktu konvergensi rata-rata topologi Ring pada EIGRP sebesar 12,75 detik dan 34,5 detik pada OSPF sedangkan topologi Mesh di EIGRP sebesar 13 detik dan 35,25 detik di OSPF. Kata Kunci: EIGRP, OSPF, Packet Tracer 5.3, Ring, Mesh, Konvergensi ABSTRACT EIGRP (Enhanced Interior Gateway Routing Protocol) and OSPF (Open Shortest Path Fisrt) is the routing protocol that is widely used in a computer network. EIGRP can only be used on devices Brand CISCO, while OSPF can be used on all brands of network. In this study comparison of both the delay and the routing protocol implemented on Ring and Mesh Topology. Cisco Packet Tracer 5.3 is used to simulate both the routing protocol. The first scenario is the design of the network and then do the test of time delay 100 times in 5 cases. The second scenario tested trace route to determine the path of the data packet and then disconnect the main link. In the second scenario also conducted a cost comparison of metrics and the simulation results with the calculation formula. The third scenario testing time for each routing protocol convergence on any Topology. The result EIGRP faster than 386 microseconds for a ring Topology while OSPF OSPF 453 microseconds faster than EIGRP for Mesh Topology. The results showed trace route chosen by the routing protocol metric value and cost is the smallest. Average convergence time in the EIGRP Topology Ring of 12.75 seconds and 34.5 seconds, while the Mesh Topology in an OSPF EIGRP for 13 seconds and 35.25 seconds in OSPF. Keywords: EIGRP,OSPF, Packet Tracer 5.3, Ring, Mesh, Convergenc

Tony Q S Quek - One of the best experts on this subject based on the ideXlab platform.

  • secret group key generation at physical layer for multi antenna Mesh Topology
    IEEE Transactions on Information Forensics and Security, 2019
    Co-Authors: Chan Dai Truyen Thai, Jemin Lee, Jay Prakash, Tony Q S Quek
    Abstract:

    In this paper, we propose a secret group-key generation scheme in physical layer, where an arbitrary number of multi-antenna LNs (LN) exist in Mesh Topology with a multi-antenna passive eavesdropper. In the first phase of the scheme, pilot signals are transmitted from selected antennas of all nodes and each node estimates channels linked to it. In the second phase, each node sequentially broadcasts a weighted combination of the estimated channel information using selected coefficients. The other LNs can obtain the channel information used for group-key generation while the eavesdropper cannot. Each node then can generate a group key by quantizing and encoding the estimated channels into keys. We apply well-known quantization schemes, such as scalar and vector quantizations, and compare their performance. To further enhance the key-generation performance, we also provide how to determine the antennas at each node used for group-key generation and the coefficients used in the broadcast phase. The simulation results verify the performance of the proposed secret group-key generation scheme using various key-related metrics. We also verify the practical robustness of our scheme by implementing a testbed using universal software radio peripheral. After generating secret common key among three nodes, we also test it using the National Institute of Standards and Technology test suit. The generated key passes the test and it is random enough for communication secrecy.

  • secret group key generation in physical layer for Mesh Topology
    Global Communications Conference, 2014
    Co-Authors: Chan Dai Truyen Thai, Jemin Lee, Tony Q S Quek
    Abstract:

    Secret group key generation based on physical layer in wireless communications has a lot of practical applications. A few studies focused on the theoretical bounds of the secret key rate based on the sets of received signals rather than a particular secret key generation scheme. We propose a secret group key generation scheme for an arbitrary number of legitimate nodes, n, in Mesh Topology in the presence of a passive eavesdropper. In the scheme, after general pilot signal transmissions at all nodes, each node broadcasts a weighted combination of its received signals with optimized coefficients, so that legitimate nodes can obtain the information of channels used for group key generation while the eavesdropper cannot. We also apply different quantization schemes for quantizing and encoding the estimated channels into keys. To provide detailed transmissions and processing steps of group key generation, we also describe the proposed scheme for 4-node Mesh Topology case. The simulation results show that the proposed scheme achieves a higher secret group key rate and a lower key disagreement rate than a benchmark scheme.