Location information recovery and management for mobile networks
First Claim
1. In a distributed database system having n database servers, D1, . . . , Dn, arranged in a logical ring, a method of updating k of the database servers, Dγ
- 1, . . . , Dγ
k, referred to by a placement vector Γ
=(γ
1, . . . , γ
k), wherein γ
i ε
{1, . . . , n} is the index of the ith database server updated by the method, the method comprising;
selecting γ
1 from the set (1, . . . , n);
for i=1, . . . , k−
1, selecting γ
i+1 according to γ
i⊕
└
n/k┘
+ai, wherein;
⊕
is modulo addition defined over the set (1, 2, . . . , n);
displacement vector â
=(a1, . . . , ak) is a binary vector having a Hamming weight of β
; and
β
=n−
└
n/k┘
*k; and
updating the k database servers, Dγ
1, . . . , Dγ
k, referred to by the placement vector Γ
=(γ
1, . . . , γ
k), with updated information.
0 Assignments
0 Petitions
Accused Products
Abstract
Systems and methods for recovering and managing location information in mobile communication networks using a fast recovery protocol and load balanced query and update processes. According to the fast recovery protocol, if a location update processor does not receive a message from a global database server acknowledging receipt by the global database server of a location update message after a predetermined retry interval has elapsed since the location update message was sent by the location update processor, the location update processor sends a location update retry message after each predetermined retry interval elapses until the location update processor receives an acknowledgement message from the global database server. The global database server can use the location update retry messages and the predetermined retry interval to recover from a database or link failure. The recovery period using the fast recovery protocol is bounded by the predetermined retry interval. The fast recovery protocol can also be used in a system having a distributed location information database architecture. The load balanced query and update processes can be used to query and update, respectively, the databases of the distributed location information database architecture.
-
Citations
10 Claims
-
1. In a distributed database system having n database servers, D1, . . . , Dn, arranged in a logical ring, a method of updating k of the database servers, Dγ
- 1, . . . , Dγ
k, referred to by a placement vector Γ
=(γ
1, . . . , γ
k), wherein γ
i ε
{1, . . . , n} is the index of the ith database server updated by the method, the method comprising;selecting γ
1 from the set (1, . . . , n);for i=1, . . . , k−
1, selecting γ
i+1 according to γ
i⊕
└
n/k┘
+ai, wherein;⊕
is modulo addition defined over the set (1, 2, . . . , n);displacement vector â
=(a1, . . . , ak) is a binary vector having a Hamming weight of β
; andβ
=n−
└
n/k┘
*k; andupdating the k database servers, Dγ
1, . . . , Dγ
k, referred to by the placement vector Γ
=(γ
1, . . . , γ
k), with updated information. - View Dependent Claims (2, 3, 4, 5, 6, 7, 8)
- 1, . . . , Dγ
-
9. In a distributed database system having n database servers, D1, . . . . Dn, arranged in a logical ring for storing information about a first group of items in k1, of the database servers and for storing information about a second group of items in k2 of the database servers, wherein k1>
- k2, a method of querying the database servers for information about a given item, the method comprising;
selecting a first database server Di; determining in parallel if any of ┌
n/k1┐
successive database servers, Di, Di+1, . . . , Di⊕
┌
n/k1┐
, on the logical ring contain information about the given item;if at least one of the ┌
n/k1┐
successive database servers, Di, . . . , Di⊕
┌
n/k1┐
, does not contain information about the given item then determining in parallel if any of thenext (┌
n/k1┐
−
┌
n/k2┐
) successive database servers, Di⊕
┌
n/k1┐
⊕
1, . . . , Di⊕
┌
n/k2┐
, on the logical ring contain information about the given device; andif at least one of the next (┌
n/k2┐
−
┌
n/k1┐
) successive database servers,Di⊕
┌
n/k1┐
⊕
1, . . . , Di⊕
┌
n/k2┐
, does not contain information about the given item, then until information for the given item is found or until all the database servers D1, . . . , Dn, have been checked, determining in parallel if successive groups of ┌
n/k2┐
database servers, Di⊕
(j*┌
n/k2┐
)⊕
1, . . . , Di⊕
(j+1)*┌
n/k2┐
, on the logical ring contain information for the given item wherein j represents the jth group of ┌
n/k2┐
database servers that are checked. - View Dependent Claims (10)
- k2, a method of querying the database servers for information about a given item, the method comprising;
Specification