R. Klasing, "Simulating large cube-connected cycles and large butterfly networks on smaller ones", Master Thesis (1990), Universitat-GH Paderborn, Fachbereich 17 -- Mathematik/Informatik, Germany. Home/Search Document Not in Database Summary Related Articles Check |
This paper is cited in the following contexts:
Compressing Cube-Connected Cycles and Butterfly Networks - Klasing, Lüling, Monien (1990) Self-citation (Klasing) (Correct)
....at most 2 Delta j 1 3 2 l Gammak k 2 2 l Gammak cycles of BFN(l) a load of l 5 3 2 l Gammak m is achieved. II. l Gamma k odd: With similar techniques as in Case I, a load of l 5 3 2 l Gammak m can be derived. A detailed discussion of the different subcases can be found in [11]. In every case, it must be made sure that at most 2 l Gammak cycles of BFN(l) are affected by the rearrangements anywhere in the cycle in BFN(k) B2) 2p Gamma4 p Gamma1 l k 2p Gamma2 p for p 2 f7; 8; g By choosing (i) l il k m for 0 i k Gamma 1 in the first stage, ....
....most 2 l Gammak cycles of BFN(l) are affected by the rearrangements anywhere in the cycle in BFN(k) It turns out that the cases n 2 f4r 3; 4r 4; 4r 5; 4r 6g, for r 2 IN 0 , have to be distinguished. Here, we only state the case n = 4r 3. All the other cases work in a similar way (cf. [11]) In each case, the load derived is l 2n Gamma1 n 2 l Gammak m . Let n = 4r 3, and let j 1 n 2 l Gammak k be abbreviated by L. Then the load in each section can be balanced as follows: Gamma Gamma Gamma Gamma Gamma Gamma . Gamma Gamma Gamma Gamma ....
[Article contains additional citation context not shown here]
R. Klasing, "Simulating large cube-connected cycles and large butterfly networks on smaller ones", Master Thesis (1990), Universitat-GH Paderborn, Fachbereich 17 -- Mathematik/Informatik, Germany.
Compressing Cube-Connected Cycles and Butterfly Networks - Klasing, Lüling, Monien (1990) Self-citation (Klasing) (Correct)
.... ba i 1 : a l Gamma1 ) b 2 fa i ; a i g onto (d(i 1; a k : a i Gamma1 ba i 1 : a l Gamma1 ) a 0 : a k Gamma1 ) d(i 1; a k : a i Gamma1 ba i 1 : a l Gamma1 ) a 0 : a k Gamma1 ) By using l k 2 and property ( of d, it can easily be verified (cf. [11]) that d ij l k 2 k ; afi j Gamma d ij l k 2 k Gamma 1; afl j 1 for all a 2 f0; 1g; fi; fl 2 f0; 1g l Gammak Gamma1 : Hence, the two image nodes of 1. i; ff) and (i 1; ff) i = j l k 2 k Gamma 1; 2. i; ff) and (i 1; ff(i) i = j l k 2 k Gamma 1; have ....
....k times (thus affecting at most 2 Delta j 1 3 2 l Gammak k 2 2 l Gammak cycles) a load of l 5 3 2 l Gammak m is achieved. II. l Gamma k odd: With similar techniques as in Case I, a load of l 5 3 2 l Gammak m can be derived. A detailed discussion can be found in [11]. The only thing which has to be checked in every case is that at most 2 l Gammak cycles of BFN(l) are affected by the rearrangements anywhere in the cycle in BFN(k) B2) 2p Gamma4 p Gamma1 l k 2p Gamma2 p ; 7 p 2k By choosing (i) l il k m for 0 i k Gamma 1 in the first ....
[Article contains additional citation context not shown here]
R. Klasing, "Simulating large cube-connected cycles and large butterfly networks on smaller ones", Master Thesis (1990), Universit¨at-GH Paderborn, Fachbereich 17 -- Mathematik/Informatik, Germany.
Online articles have much greater impact More about CiteSeer.IST Add search form to your site Submit documents Feedback
CiteSeer.IST - Copyright Penn State and NEC