20-12-2009, 12:03 PM
Abstract:
The use of Internet becomes increased and the transmission gap become increased .Now achieve data transfer rates up to several terabits per second. As packet interarrival times shrink to a few tens of nanoseconds, network systems must address a transmission-processing gap by providing extremely fast data paths as well as high-performance subsystems to implement such functions as protocol processing, memory management, and scheduling. Today, network processors are an important class of embedded processors, used all across the network systems space-from personal to local and wide area networks. Network processor architectures focus on exploiting parallelism to achieve high performance. They usually employ conventional architectural concepts to accelerate the processing required to switch packets between different protocol stacks. The architectures support the mechanisms that network protocols implement in a specific stack by providing efficient data paths and by executing many intelligent network or more homogeneous links ,so in order to increase the speed special processors are used here mainly explaining the TRIPOD registers structure .By using this register we can increase the speed. Here mainly implement the function such as protocol processing, memory management and scheduling. The architecture of the TRIPOD has three register files that establish a pipeline to improve protocol processing performance. The first register processing the IP header, the other two register files are respectively loading and storing packet headers. The packet processing is used for demanding operations. Independently of such special operations, multiple PEs enable parallel execution of several instructions per packet, multithreading supports the assignment of one thread per packet to achieve fast context switching. Tramance subsystems to implement such functions as protocol processing, memory management, and scheduling
Protocol is an agreement between user and interface. The basic mechanism for transmitting information and for the receiver to detect the presence of any transmission errors when a transmission error is detected, even if it is only a single bit, then the complete data block must be discarded. This type of scheme is thus known as best try transmission or connectionless transmission. Network systems have employed embedded processors to offload protocol processing and computationally expensive operations for more than a decade. In the past few years, however, the computer industry has been developing specialized network processors to close the transmission-processing gap in network systems. Today, network processors are an important class of embedded processors, used all across the network systems space from personal to local and wide area networks. They accommodate both the Internet's explosive growth and the proliferation of network-centric system architectures in environments ranging from embedded networks for cars and surveillance systems to mobile enterprise and consumer networks. The protocol processing is oblivious to register file management because the register file structure is transparent to the protocol code. The transparency originates in a simple mechanism that changes the working of register file. The memory system includes all parts of the computer that store information. It consists of primary and secondary memory. The primary memory can be referenced one byte at a time, has relatively fast access time, and is usually volatile. The secondary memory refers to collection of storage devices. The modern memory managers automatically transfer information back and forth between the primary and secondary memory using virtual memory. CPU scheduling refers to the task of managing CPU sharing among a community of processors. The scheduling policy would be selected by each system administrator so it reflects the way that particular computer will be used. Two types of scheduling are preemptive and non-preemptive scheduling. in preemptive scheduling use the interval timer and scheduler to interrupt a running process in order to reallocate the CPU to a higher priority ready process. In non-preemptive scheduling algorithm allow a process to run to completion once it obtains the processor
