Advanced Interconnect Protocols Using High-performance Deadlock-free Id Assignment
ثبت نشده
چکیده
In a latest network-on-chip design, several cores and intellectual properties can be integrated into a one chip. For high-performance interconnects, designers rapidly adopt advance interconnect protocols that support mechanisms of parallel accessing, including outstanding transactions and out-of-order complete of transactions. To produce these novel mechanisms, a master tag an ID to each transaction to decide in-order or out-of-order properties. These advanced protocols may lead to transaction deadlocks that do not occur in existing protocols. To prevent the deadlock problem, current solutions stall suspicious transactions and in some cases, many such stalls can incur serious performance penalty. Here, we propose a novel ID assignment mechanism that guarantee the issue transactions to be deadlock-free and results in rapid reduction in the number of transaction stalls issued by masters. Our experiment results show great performance improvements compared with last works with little hardware and power overheads.
منابع مشابه
Deadlock Free Routing in Mesh Networks on Chip with Regions
There is a seemingly endless miniaturization of electronic components, which has enabled designers to build sophisticated computing structures on silicon chips. Consequently, electronic systems are continuously improving with new and more advanced functionalities. Design complexity of these Systems on Chip (SoC) is reduced by the use of pre-designed cores. However, several problems related to t...
متن کاملEfficient Multithreaded Context ID Allocation in MPI
An important aspect of support for multithreaded MPI executions is the management of communication context identifiers (IDs), which are used to associate MPI communication operations with a communicator. New communicator creation functionality in MPI 3.0 adds complexity to this core resource management problem. We present an efficient algorithm for multithreaded context ID allocation that build...
متن کاملGraceful deadlock-free fault-tolerant routing algorithm for 3D Network-on-Chip architectures
Three-Dimensional Networks-on-Chip (3D-NoC) has been presented as an auspicious solution merging the high parallelism of Network-on-Chip (NoC) interconnect paradigm with the high-performance and lower interconnect-power of 3-dimensional integration circuits. However, 3D-NoC systems are exposed to a variety of manufacturing and design factors making them vulnerable to different faults that cause...
متن کاملThe Offset Cube: A Three-Dimensional Multicomputer Network Topology Using Through-Wafer Optics
Three-dimensional packaging technologies are critical for enabling ultra-compact, massively parallel processors (MPPs) for embedded applications. Through-wafer optical interconnect has been proposed as a useful technology for building ultra-compact MPPs since it provides a simplified mechanism for interconnecting stacked multichip substrates. This paper presents the offset cube, a new network t...
متن کاملAutomatic Veri cation of Asynchronous Retransmission Go-Back-N ARQ Protocols Using the Concurrency Workbench
The Go-Back-N and Selective-Repeat automatic repeat request (ARQ) protocols are widely used for error control in computer-communication networks. They are nearly optimal for channels characterized by low error rates and small propagation delays, and thus suitable for many classical data transmission applications. However, these protocols are not eecient for modern non-conventional channels with...
متن کامل