Delay-Insensitive Pipelined Communicatioon on Parallel Buses
نویسندگان
چکیده
Abstruct-Cons:ler a communication channel that consists of several subchannels transmitting simultaneously and asynchronously. As an example of this scheme, we can consider a board with several chips. The subchannels represent wires connecting between the chips where differences in the lengths of the wires might result in asynchronous reception. In current technology, the receiver acknowledges reception of the message before the transmitter sends the following message. Namely, pipelined utilization of the channel is not possible. Our main contribution is a scheme that enables transmission without an acknowledgment of the message, therefore enabling pipelined communication and providing a higher bandwidth. Moreover, our scheme allows for a certain number of transitions from a second message to arrive before reception of the current message has been completed, a condition that we call skew. We have derived necessary and sufficient conditions for codes that can tolerate a certain amount of skew among adjacent messages (therefore, allowing for continuous operation) and detect a larger amount of skew when the original skew is exceeded. These results generalize previously known results. We have constructed codes that satisfy the necessary and sufficient conditions, studied their optimality, and devised efficient decoding algorithms. To the best of our knowledge, this is the first known scheme that permits efficient asynchronous communications without acknowledgment. Potential applications are in onchip, on-board, and board to board communications, enabling much higher communication bandwidth.
منابع مشابه
Mutual Exclusion on Optical Buses
This paper presents two algorithms for mutual exclusion on optical bus architectures including the folded one-dimensional bus, the one-dimensional array with pipeleined buses (1D APPB), and the twodimensional array with pipelined buses (2D APPB). The rst algorithm guarantees mutual exclusion, while the second guarantees both mutual exclusion and fairness. Both algorithms exploit the predictabil...
متن کاملEfficient implementation of low time complexity and pipelined bit-parallel polynomial basis multiplier over binary finite fields
This paper presents two efficient implementations of fast and pipelined bit-parallel polynomial basis multipliers over GF (2m) by irreducible pentanomials and trinomials. The architecture of the first multiplier is based on a parallel and independent computation of powers of the polynomial variable. In the second structure only even powers of the polynomial variable are used. The par...
متن کاملPipelined Communications in Optically Interconnected Arrays
Two synchronous multiprocessor architectures based on pipelined optical bus interconnections are presented. The first is a linear pipeline with enhanced control strategies which make optimal use of the available communication bandwidth of the optical bus. The second is a two-dimensional architecture in which processors are placed in a square grid and interconnected to one another through horizo...
متن کاملSIMULATING A PIPELINED RECONFIGURABLE MESH ON A LINEAR ARRAY WITH A RECONFIGURABLE PIPELINED BUS SYSTEM by MATHURA GOPALAN
Due to the unidirectional nature of propagation and predictable delays, optically pipelined buses have been gaining more attention. There have been many models that have been proposed over time that use reconfigurable optically pipelined buses which in turn function based on numerous parallel algorithms. These models are well suited for parallel processing due to the high bandwidth available by...
متن کاملSimulating a Pipelined Reconfigurable Mesh on a Linear Array with a Reconfigurable Pipelined Bus System
Due to the unidirectional nature of propagation and predictable delays, optically pipelined buses have been gaining more attention. There have been many models that have been proposed over time that use reconfigurable optically pipelined buses which in turn function based on numerous parallel algorithms. These models are well suited for parallel processing due to the high bandwidth available by...
متن کاملذخیره در منابع من
با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید
برای دانلود متن کامل این مقاله و بیش از 32 میلیون مقاله دیگر ابتدا ثبت نام کنید
ثبت ناماگر عضو سایت هستید لطفا وارد حساب کاربری خود شوید
ورودعنوان ژورنال:
- IEEE Trans. Computers
دوره 44 شماره
صفحات -
تاریخ انتشار 1995