Security method for allocation of virtual machines in a cloud computing network
A method for enhancing security in a cloud computing system by allocating virtual machines over hypervisors, in a cloud computing environment, in a security-aware fashion. The invention solves the cloud user risk problem by inducing a state such that, unless there is a change in the conditions under which the present invention operates, the cloud users do not gain by deviating from the allocation induced by the present invention. The invention's methods include grouping virtual machines of similar loss potential on the same hypervisor, creating hypervisor environments of similar total loss, and implementing a risk tiered system of hypervisors based on expense factors.
1. A method for enhancing security in a cloud computing system having a cloud computing provider, cloud computing users, and hypervisors, comprising:
observing interactions between an attacker and said cloud computing users;
identifying the most likely targeted users,
wherein said identifying further comprises an observation of events and equilibrium states;
based on said observation of interactions, minimizing damage caused by said attacker and providing a proactive response to
subsequent attacks;
an ordering of all said users according to their respective loss potential in increasing order;
a determination that a first condition
L 1 ( q N −q I π) <e <L n ( q N −q I )
is satisfied;
wherein when said first condition is not satisfied, said cost associated with investing in security, e, is set by said cloud provider to satisfy said first condition; and
wherein when said first condition is satisfied, said met od further comprises
an allocation of a first said user to a first said hypervisor; and
an allocation of an n th said user to an n th said hypervisor,
wherein
L 1 is the loss potential for a first said user;
L n is the loss potential for an n th said user;
q N is the probability of compromise when not having invested in security;
q I is the probability of compromise when having invested in security;
e is the cost associated with investing in security; and
π is the probability that a hypervisor is compromised given a successful attack on a user.
2. The security enhancement of claim 1 , wherein said n th said hypervisor is a second said hypervisor.
3. The security enhancement method of claim 2 , further comprising a determination that a second condition
L
n
>
q
N
q
l
(
L
n
-
1
+
π
L
n
-
2
+
…
+
π
L
2
+
π
L
1
)
and a third condition
L
n
<
q
N
q
l
(
L
n
-
1
+
π
L
n
-
2
+
…
+
π
L
2
+
π
L
1
)
and
e
<
(
q
N
-
q
l
)
L
n
-
1
are satisfied.
4. The security enhancement method of claim 3 , wherein when either said second or said third conditions are satisfied, said method further comprises an allocation of all users beginning with a second said user through a (n−1) th said user to said first said hypervisor.
5. The security enhancement method of claim 3 , wherein when neither said second nor said third conditions are satisfied, said method further comprises an iterative determination for a Nash Equilibrium solution.
6. The security enhancement method of claim 5 , wherein said determination for a Nash Equilibrium solution further comprises:
an iteration value setting of i=n−2, where i is said iteration value, and n is the number of said users; and
an allocation of current and lower iteration users to said first hypervisor and
an allocation of all higher iteration users to said second hypervisor when all the following conditions are satisfied
e >( q N −q I ) L i ;
e <( q N π−q I ) L i ; and
LHS<RHS
where LHS is a left hand side of said equilibrium; and
RHS is a right hand side of said equilibrium.
7. The security enhancement method of claim 6 , wherein said determination for a Nash Equilibrium solution further comprises
a new equilibrium condition when the following conditions are not all satisfied
e >( q N −q I ) L i ;
e <( q N π−q I ) L i ; and
LHS <RHS;
wherein said new equilibrium condition comprises
updating a left hand side (LHS) of said equilibrium according to
LHS=LHS +q I πL i ;
updating a right hand side (RHS) of said equilibrium according to
RHS =RHS −q N L i +q N L i−1 −q N πL i−1 ; and
decrementing said iteration value by 1.
8. The security enhancement method of claim 7 , further comprises
a determination whether said iteration value i is equal to zero; and when said iteration value i is not zero; and
a continuation of said iterative determination for a Nash Equilibrium solution when said iteration value i is not zero.
9. The security enhancement method of claim 8 , further comprises a determination that a mixed Nash Equilibrium exists when said iteration value i is zero.
10. The security enhancement method of claim 9 , further comprising
a first left hand side (LHS) placeholder of said equilibrium LHS 1 set to
LHS 1 =q I L n
a first right hand side (LHS) placeholder of said equilibrium RHS 1 set to
RHS 1 =q I L n−1 +q N πL n−2 + . . . +q N πL 1
a second left hand side (LHS) placeholder of said equilibrium LHS 2 set to
LHS 2 =q I L n +q I πL n−1
a second right hand side (LHS) placeholder of said equilibrium RHS 2 set to
RHS 1 =q I L n−2 +q N πL n−3 + . . . +q N πL 1 ; and
a resetting of said iteration value i to i =n−1.
11. The security enhancement method of claim 10 , further comprising
an allocation of second user through (i−1) th users to said first hypervisor;
an allocation of (i+2) th users through (n−1) th users to said second hypervisor;
an allocation of an i th user to said first hypervisor with a probability of α 1 ; and
an allocation of said 1 th user to said second hypervisor with a probability of (1−α 1 )
when all of the following conditions are satisfied
LHS 1 <RHS 1 ;
LHS 2 >RHS 2 ; and
e <( q N −q I ) L i .
12. The security enhancement method of claim 11 further comprises when
LHS 1 <RHS 1 ;
LHS 2 >RHS 2 ; and
e <( q N −q I ) L i
are all not satisfied, the following:
upon a determination that said iteration value i is not equal to 2;
setting
LHS 1 =LHS 1 +q I πL i ;
RHS 1 =RHS 1 −q N L i +q N L i−1 −q N πL i−1 ;
LHS 2 =LHS 2 +q I πL i−1 ;
RHS 2 =RHS 2 −q N L i−1 +q N L i−2 −q N πL i−2 ;
a resetting of said iteration value i to i=i−1; and
a redetermination whether
LHS 1 <RHS 1 ;
LHS 2 >RHS 2 ; and
e <( q N −q I ) L i
are all satisfied.
13. The security enhancement method of claim 12 , further comprises upon a determination that said iteration value i is equal to 2,
a setting of
LHS= q N L 1 ;
RHS= q I πL 2 +q I πL 3 + . . . +q I L n ;
a resetting of said iteration value i to i=2; and
upon a determination that iteration value i equals n,
a setting of said cost associated with investing in security e so as to satisfy pure Nash Equilibrium conditions.
14. The security enhancement method of claim 13 , further comprising when said iteration value i is not equal to n,
an allocation of second user through (i−1) th users to said first hypervisor;
an allocation of (i+2) th users through (n−1) th users to said second hypervisor,
an allocation of an i th user to said first hypervisor with a probability of α 2 ; and
an allocation of said i th user to said second hypervisor with a probability of (1−α 2 ) when all of the following conditions are satisfied
LHS>RHS;
e <( q N −q I ) L i ; and
e>q N πL i .
15. The security enhancement method of claim 14 further comprises when
LHS 1 <RHS 1 ;
LHS 2 >RHS 2 ; and
e <( q N −q I ) L i
are all not satisfied, the following:
setting
RHS=RHS− q I πL i ;
LHS=LHS− q N L i−1 +q N L i +q N πL i−1 ;
a resetting of said iteration value i to i=i+1; and
a redetermination whether said iteration value i equals n.
16. The security enhancement method of claim 14 , wherein said probability α 1 is
α
1
=
q
l
(
1
-
π
)
(
L
n
-
L
i
)
q
N
π
(
L
1
+
L
2
+
…
+
L
i
-
1
)
-
q
l
π
(
L
i
+
1
+
L
i
+
2
+
…
+
L
n
)
+
[
q
N
-
q
l
(
1
-
π
)
]
L
i
and
wherein said probability α 2 is
α
2
=
q
N
π
(
L
1
+
L
2
+
…
+
L
i
-
2
)
-
q
l
π
(
L
i
+
1
+
L
i
+
2
+
…
+
L
n
)
+
q
N
L
i
-
1
-
q
l
L
i
q
N
π
(
L
1
+
L
2
+
…
+
L
i
-
1
)
-
q
l
π
(
L
i
+
1
+
L
i
+
2
+
…
+
L
n
)
+
[
q
N
(
1
-
π
)
-
q
l
]
L
i