The -dimensional Möbius cube is an important variant of the hypercube , which possesses some properties superior to the hypercube. This paper investigates the fault-tolerant edge-pancyclicity of , and shows that if () contains at most faulty vertices and/or edges then, for any fault-free edge in and any integer with , there is a fault-free cycle of length containing the edge , where is the number of faulty vertices. The result is optimal in some senses.
It is well known that interconnection networks are of interest in
parallel and distributed computing systems because they determine the
performance of the systems on a large scale. As topological structures,
interconnection networks can be represented by a graph , where is the vertex-set of and is the edge-set of . and denote the numbers of vertices and
edges of , respectively. A path
denoted by is a sequence of vertices where two consecutive
vertices are adjacent in . A cycle
is a path where .
A graph is
pancyclic if, for every , has a cycle of length . A graph is edge-pancyclic
if, for any edge of and every , has a cycle of length containing . A graph is
vertex-pancyclic if, for any vertex of and every , has a cycle of length containing . Obviously, if a graph is
edge-pancyclic, it is also vertex-pancyclic. Edge-pancyclic and
vertex-pancyclic on various interconnection networks were studied,
including hypercubes, crossed cubes, twisted cubes, locally twisted
cubes, augmented cubes, star graphs, and others.
Edge and/or vertex failures are inevitable when a large parallel and
distributed computer system is running. Thus, the fault-tolerant
capacity of network is an important issue for parallel and distributed
computing. A graph is
-fault-tolerant
pancyclic(resp., vertex-pancyclic, edge-pancyclic) if is still pancyclic(resp.,
vertex-pancyclic, edge-pancyclic) for any with . Pancyclicity and
fault-tolerant pancyclicity have been widely studied for many well-known
networks, see Xu and Ma [14]
for a detail survey on these topics.
The Möbius cube has many properties superior to the hypercube. Though
both the Möbius cubes and the ordinary hypercube have the same number of
vertices and the same vertex degree, the diameter of the Möbius cube is
approximately half that of the ordinary hypercube. Due to nearly half
the diameter and better graph embedding capability as compared with its
hypercube counterpart of the same size, the Mö¡§bius cubes have been
proposed as promising candidates for interconnection topology, and have
received considerable attention [1,2,3,6,7,8,11,13,5,4].
With regard to the fault-tolerant Hamiltonicity of Möbius cubes,
Huang et al. [7] showed that
an -dimensional Möbius cube is
Hamiltonian in the presence of up to node and edge faults. As concerns the
pancyclicity and fault-tolerant pancyclicity of Möbius cubes, Fan [3] proved that Möbius cubes are
four-pancyclic. Hsieh and Chen [6] found that an -dimensional Möbius cube with up to
edge faults is
four-pancyclic.
This paper investigates the fault-tolerant edge-pancyclicity of , and shows that if () contains at most faulty vertices and/or edges then,
for any fault-free edge in and any integer with , there
is a fault-free cycle of length containing the edge , where is the number of faulty vertices.
The remainder of this paper is organized as follows. In Section 2, we
recall the definition of , and
introduce some properties of
to be used in our proofs. In Section 3, we give the proof of our result.
Finally, we give some concluding remarks in Section 4.
2. Definitions and properties
The -dimensional Möbius cube,
denoted by , is such an
undirected graph, its vertex set is the same as the vertex set of , the vertex connects to other vertices , (), where each
satisfies one of the following equations:
From the above definition,
connects to by complementing
the bit if or by complementing all bits
of if . The connection between and is undefined, so we can assume that
is either equal to or equal to , which gives us slightly different
network topologies. If we assume , we call the network a ¡®0-Möbius cube¡¯; and if we assume
, we call the network a
¡®1-Möbius cube¡¯, denoted by and , respectively. The graphs shown in
Fig.1 are and , respectivily.
According to the above definition, it is not difficult to see that
(respectively, ) can be recursively constructed
from and by adding edges. is constructed by connecting all
pairs of vertices that differ only in the -th bit, and is constructed by connecting all
pairs of vertices that differ in the -th through the -th bits. The superscript of notation , , can be omitted if there is no ambiguity arise.
For convenience, we say that and are two sub-Möbius cubes of
, where (respectively, ) is an ()-dimensional -Möbius cube (respectively, -Möbius cube) which includes all
vertices
(respectively, ),
. More simply, let
and . In addition, we define the
set of crossing edges of to be
. For any edge , vertices and are crossing vertices of each other.
Indeed, there are crossing
edges and pairs of crossing
vertices in .
The Möbius cube was first
proposed by Cull and Larson [1]. Like ,
is an -regular -connected graph with vertices and edges. Moreover, has a diameter of for and for . However, for , is neither vertex-transitive nor
edge-transitive.
Cull and Larson[1] first
proved the existence of hamiltonian cycles in by proving that in or , there are disjoint cycles of length for any .
Figure 1 (a) , (b)
We need the following two definitions.
Definition 2.1.
For any edge , let be
the smallest positive integer such that , then is called a
-dimensional edge.
According to Definition 2.1, if we use to denote a vertex in , then always denotes its unique neighbor in
, that is, is an -dimensional edge. Let be a -dimensional edge, then denote . Let
for . Let be a path of length , then for . We use to denote for short. If , then we use to denote cycle
of length containing edge for short.
For example, for
in , we use to denote the cycle
.
Definition 2.2.
If and there exists a vertex with , then is called a weak 2-degree vertex and
is called a -weak vertex pair(or a weak vertex pair,
for short).
Since there is no triangle in , we can obtain the Proposition 2.3 as follows.
Proposition 2.3.
If
, then is not a weak vertex-pair in with .
Lemma 2.4.
( Xu
et al. [9]) If
for any vertex (), () is a neighbor of () in ()
then, the distance between () and () is 1 or 2.
( Xu
et al. [10]) For
any two different vertices and
with distance in , there exists an -path of every length from to except for .
Let be a set of faulty
elements in . We need the
following lemmas:
Lemma 2.7.
If any edge in for , there exists a fault-free
-cycle or -cycle containing the edge .
Proof. Let , we show the lemma according to the following two
cases.
Case 1..
We can find () disjoint -cycles and a -cycle except the common edge as follows.
Since , there exists
a fault-free -cycle or -cycle containing the edge . The lemma holds.
Case 2..
There exist at most disjoint
-cycles as
for except the common
edge .
If one of -cycle () is fault-free, the lemma
holds.
If each -cycles () contains a faulty
vertices. Consider -cycles .
If , then
. We can find a fault
free -cycle containing the edge .
If , then . We can find a
fault-free -cycle containing the edge .
The proof is completed.
Lemma 2.8. If any edge in for , there exists a fault-free
-cycle containing the edge .
Proof. Let . There exist () disjoint -cycles and a -cycle except the common edge as follows.
Since , there exists
a fault-free -cycle or -cycle containing the edge .
By Lemma 2.7 and 2.8, we can
obtain the following result.
Corollary 2.9.
If any edge in for , there exists a fault-free
-cycle or -cycle containing the edge .
Lemma 2.10.
(Xu
et al. [9]) If and then, for
any fault-free edge in and any integer with , there is a
fault-free cycle of length
containing the edge in .
Lemma 2.11.
(Xu
et al. [12]) If and , then for any two distinct fault-free vertices and , there exists a fault-free path of every length with ,
where if vertices and is not a weak vertex-pair in and if vertices and form a weak vertex-pair in ().
Figure 2 Four disjoint -cycles in except the common edge
Lemma 2.12.
If with (), then for any edge , there exists a fault-free
-cycle containing the edge in .
Proof. By Lemma 2.10, the lemma holds for . We only need to consider the
case of .
Let . We can assume that . Otherwise, if , then . By Lemma 2.10, there
exists a fault-free -cycle
containing the edge in , and so in .
If , by Lemma 2.10 , there exists a fault-free -cycle containing the edge .
If , then . We can prove the result
according to the relationship between the and as follows.
Case 1.().
There exist disjoint -cycles as except
the common edge as follows.
Since , there exists
a fault-free -cycle containing the
edge .
Case 2.().
Case 2.1..
For , there exist
two disjoint -paths as of length in as follows.
For , there exist two
disjoint -paths as of length in as follows.
Since , there exists a
fault-free -path of length in . Then is a
fault-free -cycle containing the
edge .
Case 2.2.. Without loss of
generality, assume that .
Let , then .
For , there exist -cycles as
containing the common edge
in as follows.
For , there
exist -cycles as
containing the common edge
in as follows:
For , there exist -cycles as
containing the common edge
in as follows.
For , there exist -cycles as containing the common
edge in .
Note that -cycles are disjoint in and contain no in . Since , we have that there
exists a fault-free 6-cycles in , and so in .
Take edge
in for an example. We have
,. Assume , then . There exist four disjoint
-cycles in except a common edge as follows(see
Figure 2).
Lemma 2.13.
If with (), then for any edge , there exists a fault-free
-cycle containing the edge in .
Proof. Let be
the adjacent vertices of
respectively in and . Since
, . There exists a fault-free -path of length in as follows.
Then is
a fault-free -cycle containing the
edge in .
Lemma 2.14.
For any edge in ()() with , there exists a fault-free
cycle of length
containing the edge in ().
Proof. Let . We prove the lemma according to edge in or as follows.
Case 1..
Case 1.1..
For , we can find
disjoint -cycles as except the common edge
as follows.
Since , there exists
a fault-free -cycle containing the
edge .
Case 1.2..
For , we can find
disjoint -cycles as except the common edge
as follows.
Since , there exists
a fault-free -cycle containing the
edge .
Take edge in
for an example, there exist
disjoint -cycles except the common edge as follows(see Figure 3).
Figure 3
disjoint -cycles except the common
edge in
Case 2..
Case 2.1..
For , we can find
disjoint -cycles as except the common edge
as follows.
Since , there exists
a fault-free -cycle containing the
edge .
Case 2.2..
For , we can find
disjoint -cycles as except the common edge
as follows.
Since , there exists
a fault-free -cycle containing the
edge .
Case 2.3..
For , we can find
disjoint -cycles as except the common edge
as follows.
Since , there exists
a fault-free -cycle containing the
edge .
Take edge in
for an example, there exist
disjoint -cycles except the common edge as follows (see Figure 4):
Figure 4 5 disjoint 8-cycles except the common edge (010010, 101101) in
3. Fault-tolerant
edge-pancyclicity of
In this section, we investigate the fault-tolerant edge-Pancyclicity
of and show that is -fault-tolerant edge-Pancyclic.
Let be a set of faulty
elements in , , , , ,
and , , .
Theorem 3.1.
If and then, for any fault-free edge
in and any integer with , there is a
fault-free cycle of length
containing the edge .
In this section, we will give the proof of Theorem 3.1. The theorem
follows from Lemma 2.10 if . We only need to
consider . Start with the
following lemma.
Lemma 3.2.
If Theorem 3.1 holds for any
subset with , then Theorem 3.1 also holds for any subset with .
Proof. The lemma holds for by hypothesis of Lemma 3.2. Assume that the lemma holds for
any with and . We prove the lemma holds for . Let be an edge in .
Case 1. or .
Without loss of generality, assume . Let ,
then . By Lemma 2.10, this result has been proved.
Case 2..
Let ,
then , and contains at most edges and vertices. By induction hypothesis,
for every integer with , there is a
fault-free -cycle containing
the edge in (). By Proposition 2.1, for any , is not a weak vertex-pair in . For , by Lemma 2.11, there is
a fault-free -hamiltonian path in
, i.e., there is a
fault-free -cycle containing
the edge in .
The proof of lemma is completed.
Proof of Theorem 3.1. By Lemma 3.2, we only
need to prove the theorem with . The proof proceeds by induction on . The result holds for by developing computer program using
depth first searching technique combining with backtracking and branch
and bound algorithm.
Assume that the theorem holds for any with . Then we must show the theorem holds for .
Let be a fault-free edge in
. By Proposition 2.3, is not a weak vertex-pair in . Let and , where
then, by Lemma 2.11, there exists a fault-free -path of length in . Then is a fault-free cycle of length
containing the edge in . Thus, we only need to consider
with in ().
Case 1..
Case 1.1.. Let .
By Lemma 2.12, there exists a fault-free -cycle containing the edge in . Thus, we only need to consider
the length of .
Case 1.1.1.. Then .
Case 1.1.1.1..
By induction hypothesis, there is a fault-free cycle of length containing the edge in , so in .
Case 1.1.1.2..
Write ,
where and . Since for and , by induction hypothesis,
there is a fault-free cycle of
length containing the edge
in . Note that a cycle of length contains a matching with .
Consider the following inequality.
Let . Since
for
, is an
increasing function, which implies that . It follows that, there is such an edge, say , in that and . By
Lemma 2.1, there is a fault-free -path of length in . Then is a fault-free
cycle of length containing (see Figure 5 (a)).
Case 1.1.2.. In this case, .
Let and be neighbors of and in , respectively.
Let , where
. By Lemma 2.4, . Since , by Lemma 2.6, there is
a fault-free -path of length in , and so is a fault-free
cycle of length containing the
edge (see Figure 5
(b)).
Figure 5 The illustrations of Case 1.1
Case 1.2.. Let .
Case 1.2.1.. Then .
Case 1.2.1.1. in
.
By induction hypothesis, there is a fault-free cycle of length containing the edge in , so in .
Case 1.2.1.2..
The proof is similar to Case 1.1.1.2.
Case 1.2.2.. In this case, .
Since , by Lemma 2.5, there exists a fault-free cycle of
length with containing the
edge in , and so in .
Case 1.3.. Let .
By Lemma 2.14, there exists a fault-free cycle
of length containing
the edge in (). Thus, we only need to consider
the length .
Case 1.3.1.. Then .
By Corollary 2.1., there exists a fault-free -cycle (or -cycle) containing the edge .
Case 1.3.1.1..
For . Let , where .
By induction hypothesis, there is a fault-free cycle of length containing the edge in . Then is a
fault-free cycle of length
containing the edge .
For . Let , where and . Since and for , by induction hypothesis, there
exists a fault-free cycle of
length containing the edge
in and a fault-free cycle of length containing the edge in . Then is
a fault-free cycle of length
containing the edge in
(see Figure 6(a)).
Case 1.3.1.2..
Since . Let , then .
For . Let , where .
By induction hypothesis, there is a fault-free cycle of length containing the edge in . Then
is a fault-free cycle of length containing the edge .
For . Let , where and . Since for , by induction hypothesis, there
exists a fault-free cycle of
length containing the edge
in and, by Lemma 2.11, there
exists a fault-free -path
of length in . Then is a
fault-free cycle of length
containing the edge in
(see 6(b)).
Figure 6 The illustration of Case 1.3.1.1. and Case 1.3.1.2.
Case 1.3.1.3..
For . Let , where .
By induction hypothesis, there is a fault-free cycle of length containing the edge in . Then
is a fault-free cycle of length containing the edge .
For . Let , where and . Since for , by induction hypothesis, there
exists a fault-free cycle of
length containing the edge
in and, by Lemma 2.11, there
exists a fault-free -path
of length in . Then is a
fault-free cycle of length
containing the edge in (see Figure 7(a)).
Case 1.3.2.. In this case, .
Since , there exists a
fault-free neighbor of in . Let be neighbors of in .
Let , where
. By
Lemma 2.4, . By Lemma 2.6, there is a fault-free -path of length in . So is a fault-free
cycle of length containing the
edge .
Case 2..
Case 2.1.. Let .
Case 2.1.1.. Then .
For in . By induction hypothesis,
there is a fault-free cycle of length containing the edge in , so in .
For . The proof is similar to Case 1.1.1.2.
Case 2.1.2.. In this case, .
By Lemma 2.13, there exists a fault-free -cycle containing the edge in . Thus, we only need to consider
the length of .
Let and be neighbors of and in , respectively.
Let , where
. By Lemma 2.4, . Since , by Lemma 2.6, there is
a fault-free -path of length in , and so is a fault-free
cycle of length containing the
edge .
Case 2.2.. Let .
Case 2.2.1.. Then .
For in . Since , by Lemma 2.10, there
exists a fault-free cycle of length containing the edge in .
For . The proof is similar to Case 1.1.1.2.
Case 2.2.2.. In this case, .
Since , by Lemma 2.5, there exists a fault-free cycle of
length with containing the
edge in , and so in .
Case 2.3.. Let .
By Lemma 2.14, there exists a fault-free cycle
of length containing
the edge in (). Thus, we only need to consider
the length .
Case 2.3.1.. Then .
By Corollary 2.1., there exists a fault-free -cycle (or -cycle) containing the edge .
Case 2.3.1.1..
The proof is similar to Case 1.3.1.1.
Case 2.3.1.2..
For . Let , where .
Since , by Lemma 2.10, there is a fault-free cycle of length containing the edge in . Then
is a fault-free cycle of length containing the edge .
For . Let , where and . Since for , by induction hypothesis, there
exists a fault-free cycle of
length containing the edge
in and, by Lemma 2.11, there
exists a fault-free -path
of length in . Then is a
fault-free cycle of length
containing the edge in
(see Figure 7(b)).
Figure 7 The illustration of Case 1.3.1.3. and Case 2.3.1.2.
Case 2.3.1.3..
The proof is similar to Case 1.3.1.3.
Case 2.3.2.. In this case, .
Since , there exists a
fault-free neighbor of in . Let be the neighbor of in . Let , where . By Lemma
2.4, . By Lemma 2.6, there is a fault-free -path of length in . So is a fault-free
cycle of length containing the
edge .
The proof of the theorem is completed.
4. Conclusion and remarks
As one of the most fundamental networks for parallel and distributed
computation, cycles are suitable for developing simple algorithms with
low communication cost. Edge and/or vertex failures are inevitable when
a large parallel computer system is put in use. Therefore, the
fault-tolerant capacity of a network is a critical issue in parallel
computing. The fault-tolerant edge-pancyclicity of an interconnection
network is a measure of its capability of implementing ring-structured
parallel algorithms in a communication-efficient fashion in the presence
of faults.
In view of the fact that the hypercube network contains only even cycles, is superior to in fault-tolerant pancyclicity. This
shows that, when the is used
to model the topological structure of a large-scale parallel processing
system, our result implies that the system has larger capability of
implementing ring-structured parallel algorithms in a
communication-efficient fashion in the hybrid presence of edge and
vertex failures than one of the hypercube network.
We make some remarks on the optimality of our result in the following
sense.
For , in , taking , there are only two
cycles: , of length 4 containing the edge (we calculate it by computer).
If , then there
exists no fault-free cycle of length 4 containing the edge in . In , taking , there exists only one
cycle of
length 4 containing the edge . If
or , then there exists no
fault-free cycle of length 4 containing the edge in .
For , in , taking , the cycles of length 5
containing the edge are as and (we calculate it by
computer):
Table 1 Cycles of length 5 containing the edge in
(11110 11101 11010 11000 11111)
(11110 11101 10010 10000 11111)
(11110 11001 11011 11100 11111)
(11110 10001 10011 11100 11111)
Table 2 Cycles of length 5 containing the edge in
(11110 11101 11010 11000 11111)
(11110 11101 10010 10000 11111)
(11110 11101 00010 00000 11111)
(11110 11001 11011 11100 11111)
(11110 10001 10011 11100 11111)
(11110 00001 00011 11100 11111)
If , then there
exists no fault-free cycle of length 5 containing the edge in .
For , in , taking , the cycles of length 6
containing the edge are as (we calculate it by
computer):
Table 3 Cycles of length 6 containing the vertex in
(01011 01010 01000 00000 00100 01100)
(01011 01010 01101 01110 01111 01100)
(01011 01010 01101 00101 00100 01100)
(01011 01010 01101 11101 11100 01100)
(01011 01010 00010 00000 00100 01100)
(01011 01010 11010 11011 11100 01100)
(01011 01010 11010 11101 11100 01100)
(01011 01010 11010 11101 01101 01100)
(01011 01001 01000 01010 01101 01100)
(01011 01001 01000 00000 00100 01100)
(01011 01001 00001 00000 00100 01100)
(01011 01001 00001 00101 00100 01100)
(01011 01001 00001 00101 01101 01100)
(01011 01001 11001 11011 11100 01100)
(01011 00011 00010 00000 00100 01100)
(01011 00011 00010 01010 01101 01100)
(01011 00011 00001 00000 00100 01100)
(01011 00011 00001 00101 00100 01100)
(01011 00011 00001 00101 01101 01100)
(01011 11011 11010 11101 11100 01100)
(01011 11011 11010 11101 01101 01100)
(01011 11011 11010 01010 01101 01100)
(01011 11011 11100 11101 01101 01100)
(01011 11011 11100 11111 01111 01100)
If , then
there exists no fault-free cycle of length 6 containing the edge in .
Acknowledgements
The research is supported by National Natural Science Foundation of
China under Grant No.U22A2005, No. U21A20491, No. U1936109, No.U1908214,
No. KLSA201906 and Key Research and Development Program of Zhejiang
(2024SSYS0001).
References:
P. Cull and S. M. Larson. The möbius cubes.
IEEE Transactions on Computers, 44(5):647–659, 1995.
J. X. Fan. Diagnosability of the möbius cubes.
IEEE Transactions on Parallel and Distributed Systems, 9(9):923–928, 1998.
S. Y. Hsieh. Fault-tolerant cycle embedding in the hypercube with more both faulty vertices and faulty edges.
Parallel Computing, 32(1):84–91, 2006.
https://doi.org/10.1016/j.parco.2005.09.003.
S. Y. Hsieh and N. W. Chang. Hamiltonian path embedding and pancyclicity on the möbius cube with faulty nodes and faulty edges.
IEEE Transactions on Computers, 55(7):854–863, 2006.
W. T. Huang, Y. C. Chuang, J. J. M. Tan, and L. H. Hsu. Fault-free Hamiltonian cycle in faulty möbius cubes.
Journal of Computing Systems, 4:106–114, 2000.
J. M. Xu and M. J. Ma. Survey on path and cycle embedding in some networks.
Frontiers of Mathematics in China, 4(2):217–252, 2009.
https://doi.org/10.1007/s11464-009-0017-5.
J. M. Xu, M. J. Ma, and M. Lu. Paths in möbius cubes and crossed cubes.
Information Processing Letters, 97(3):94–97, 2006.
https://doi.org/10.1016/j.ipl.2005.09.015.
X. R. Xu, H. F. Zhang, and Y. S. Yang. Fault-tolerant edge-pancyclicity of möbius cubes MQn.
In 2018 IEEE International Conference on Parallel and Distributed Processing with Applications, pages 237–243, Australia, 2018.
https://doi.org/10.1109/BDCloud.2018.00046.
X. Yang and G. M. Megson. Fault tolerance of möbius cubes under two forbidden fault set models.
International Journal of Computer Mathematics, 81(8):909–916, 2004.
https://doi.org/10.1080/00207160410001712332.
H. F. Zhang, X. R. Xu, P. D. Soomro, and Y. S. Yang. Fault-tolerant Hamiltonian connectivity of twisted hypercube-like networks THLNs.
IEEE Access, 6:74081–74090, 2018.
https://doi.org/10.1109/ACCESS.2018.2882520.