Analysis of Minimal Path Routing Schemes in the Presence of Faults

نویسنده

  • Jesse M. Gordon
چکیده

Gordon, J.M., Analysis of minimal path routing schemes in the presence of faults, Discrete Applied Mathematics 37138 (1992) 245-263. The design and analysis of fault tolerant message routing schemes for large paraIIe1 systems has been the focus of much recent research. In this paper, we present a framework for the analysis of routing schemes in distributed memory multiprocessor systems containing faulty or unusable components. We introduce techniques for the derivation of the probabilities of successfully routing a single inessage using minimal path routing schemes. Using this framework, we derive closed form solutions for a wide range of routing schemes on the hypercube and on the twodimensional mesh. The results obtained show the surprising resilience of the hypercube to a potentially large number of faults while demonstrating the inability of the mesh to tolerate a comparatively smaller number of faults. component failure e Since the corresponding mean time to failure is likely to be short, off-line fault diagnosis and the subsequent replacement of failed units is not an attractive alternative for dealing with the problem. Rather, we would like to be able to continue system operation in the presence of such failures. A fundamental component of the operation of distributed memory multiprocessor systems is the routing of messages. The large size of these parallel systems mitigates against the selection G; routing algorithms which require the propagation of global fault information to individual processors. Rather, we would prefer to use approaches which require Dnly local fault infxmation, such as the fault status of immediate neighbors. For similar reasons, we would prefer to use distributed (as opposed to centralized) routing schemes, i.e., ones in which routing decisions are made at the individual processors. * Supported in part by NS. r crant No. DCR-8507785 1 and a fellowship from the Unisys Corporation. Current address: IBM Corporation, 11400 Burnet Road, Zip 9641, Austin, TX 78758, USA. 0166-218X/92/$05.00

برای دانلود رایگان متن کامل این مقاله و بیش از 32 میلیون مقاله دیگر ابتدا ثبت نام کنید

ثبت نام

اگر عضو سایت هستید لطفا وارد حساب کاربری خود شوید

منابع مشابه

CAFT: Cost-aware and Fault-tolerant routing algorithm in 2D mesh Network-on-Chip

By increasing, the complexity of chips and the need to integrating more components into a chip has made network –on- chip known as an important infrastructure for network communications on the system, and is a good alternative to traditional ways and using the bus. By increasing the density of chips, the possibility of failure in the chip network increases and providing correction and fault tol...

متن کامل

A Novel Protection Guaranteed, Quality of Transmission Aware Routing and Wavelength Assignment Algorithm for All-optical Networks

Transparent All Optical Networks carry huge traffic and any link failure can cause the loss of gigabits of data; hence protection and its guarantee becomes necessary at the time of failure. Many protection schemes were presented in the literature, but none of them speaks about protection guarantee. Also, in all optical networks, due to absence of  regeneration capabilities, the physical layer i...

متن کامل

A Routing and Broadcasting Scheme on Faulty Star Graphs

In this paper we present a routing algorithm that uses the depth rst search approach combined with a backtracking technique to route messages on the star graph in the presence of faulty links. The algorithm is distributed and requires no global knowledge of faults. The only knowledge required at a node is the state of its incident links. The routed message carries information about the followed...

متن کامل

Design of a Router for Fault-Tolerant Networks

As interconnection networks grow larger and larger, the need for reliable message delivery in the presence of faults grows as well. Unfortunately, most network routing schemes currently in use do not provide graceful tolerance of even the most common faults. Because routing messages around failed components requires non-minimal routing, it makes sense to examine routers which, by design, allow ...

متن کامل

Congestion estimation of router input ports in Network-on-Chip for efficient virtual allocation

Effective and congestion-aware routing is vital to the performance of network-on-chip. The efficient routing algorithm undoubtedly relies on the considered selection strategy. If the routing function returns a number of more than one permissible output ports, a selection function is exploited to choose the best output port to reduce packets latency. In this paper, we introduce a new selection s...

متن کامل

ذخیره در منابع من


  با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید

عنوان ژورنال:
  • Discrete Applied Mathematics

دوره 37/38  شماره 

صفحات  -

تاریخ انتشار 1992