A Low Latency Arbitration Circuit
نویسندگان
چکیده
Indexing terms: Asynchronous circuits, Arbiters We present an asynchronous circuit for an arbiter cell that can be used to construct cascaded multi-way arbitration circuits. The circuit is completely speed-independent. It has a short response delay at the input request-grant handshake link due to both (a) the propagation of requests in parallel with starting arbitration, and (b) the concurrent resetting of request-grant handshakes in diierent cascades of a request-grant propagation chain. Arbitration circuits are commonplace in digital systems with a single resource allocated to diier-ent user processes. The typical examples are systems with shared busses, multi-port memories, packet routers to name but a few. An asynchronous arbiter is deened as a circuit that dynamically allocates a single shared resource to the user components in a system which is free from common clock. Each user, when requires the resource, issues an asynchronous request and waits until the arbiter produces a grant. The user then uses the resource and after nishing its action releases its request. This results in a subsequent release of the grant, after which the user can issue another request and so on. The arbiter, when it receives a number of active requests from diierent users, generates after some nite delay a grant to exactly one of them and leaves other requests pending until the granted user has released the request. The arbiter then releases the grant and, if there are pending requests, produces another active grant, again on a mutually exclusive basis. We consider a standard basic cell of a multi-way arbiter that arbitrates between two users. Multi-way arbitration is organised by building a cascade of such cells to form a tree or a linear structure. Each cell thus propagates the request in the direction from the lower level to the higher level of the structure, while the grants are generated in the opposite direction. Figure 1.(a) shows one such cell, a 2-way arbiter, with its three request-grant handshake links (R1,G1), (R2,G2) and (R,G), where (R1,G1) and (R2,G2) stand for the links with lower level cascades, generating competing requests at R1 and R2, and the (R,G) pair is the link with the higher level cascade. An example of a 4-way arbiter, shown in Figure 1.(b), illustrates the regular way in which a cascaded multi-way arbiter can be composed from the basic cells. Figure 1.(c) illustrates the handshaking protocol between the links. After a rst request by R1 is granted …
منابع مشابه
An Accommodative Adaptive Arbitration Algorithm for Maximum CPU Utilization, Fair Bandwidth Allocation and Low Latency
An adaptive algorithm for fair bandwidth allocation, low latency and maximum CPU utilization is proved to be a promising approach for designing system-on-chip for future applications. Adaptive arbitration focuses on user oriented as well as system oriented performance therefore it is more advantageous then the other conventional arbitration algorithms for several reasons; these include fair ban...
متن کاملTowards zero latency photonic switching in shared memory networks
Photonic networks-on-chip based on silicon photonics have been proposed to reduce latency and power consumption in future chip multi-core processors (CMP). However, high performance CMPs use a shared memory model which generates large numbers of short messages, creating high arbitration latency overhead for photonic switching networks. In this paper we explore techniques which intelligently use...
متن کاملNetwork-Level FPGA Acceleration of Low Latency Market Data Feed Arbitration
Financial exchanges provide market data feeds to update their members about changes in the market. Feed messages are often used in time-critical automated trading applications, and two identical feeds (A and B feeds) are provided in order to reduce message loss. A key challenge is to support A/B line arbitration efficiently to compensate for missing packets, while offering flexibility for vario...
متن کاملSynthesis and Analysis of a Delay-Insensitive Folded FIFO
Darren C. Cronquist and Steven M. Burns Department of Computer Science and Engineering University of Washington Box 352350 Seattle, WA 98195-2350 fdarrenc,[email protected] Abstract Fall-through FIFOs provide high data throughput with a small overhead for control but also yield a latency dependent on the size of the FIFO. Nonfallthrough FIFOs remove this latency dependency by adding more...
متن کاملThe effect of the router arbitration policy on the scalability of ServerNet™ topologies
In this paper we extend a previously introduced method for optimizing the arbitration policy employed by ServerNet routers and we evaluate the method's effect on scalability. The ServerNet System Area Network (SAN), developed by Tandem Computers Inc., is a wormhole-routed, packet-switched, point-to-point network, with special attention paid to reducing latency and to assuring reliability. The S...
متن کاملPerformance of a Speculative Transmission Scheme for Arbitration Latency Reduction
Low latency is a critical requirement in some switching applications, specifically in parallel computer interconnection networks. The minimum latency in switches with centralized arbitration comprises two components, namely, the control-path latency and the data-path latency, which in a practical high-capacity, distributed switch implementation can be far greater than the cell duration. We intr...
متن کاملذخیره در منابع من
با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید
عنوان ژورنال:
دوره شماره
صفحات -
تاریخ انتشار 2007