Optimized resource management in core network elements
The present invention addresses method, apparatus and computer program product for resource management within a distributed system in a core network element, comprising organizing computing resources of the core network element into sets, wherein a first set is always active, setting an upper threshold and a lower threshold for the load of the sets, wherein the sets in operation are loaded as long as the average load of the sets in operation reaches the upper threshold, and when the upper threshold is exceeded, a new set is activated, whereas, when the load thereof falls below the lower threshold, the last activated set is deactivated, assigning a priority number to each set, segmenting an interval of random numbers for randomizing a request distribution in subintervals which are allocated to the computing resources of the active sets, wherein the length of each subinterval is determined based on the priority number of the set of the respective computing resource, and allotting a random number out of the interval to an incoming request, and forwarding the request to such computing resource which belongs to the subinterval that contains the allotted random number.
1. A method for resource management within a distributed system in a core network element, comprising:
organizing computing resources of the core network element into sets, wherein a first set is always active, and wherein a load of incoming traffic is associated with each set;
controlling distribution of incoming traffic by setting an upper threshold and a lower threshold for the load of the sets, wherein the sets in operation are loaded as long as the average load of the sets in operation reaches the upper threshold, and when the upper threshold is exceeded, a new set is activated, whereas, when the load of the last activated set falls below the lower threshold, the last activated set is deactivated;
assigning a priority number to each set which provides an indication of an amount of traffic manageable by each set;
segmenting an interval of a group of random numbers for randomizing a request distribution into subintervals which are allocated to the computing resources of the active sets, wherein the length of each subinterval is determined based on the priority number of the set of the respective computing resource; and
allotting a random number out of the interval to an incoming request, and forwarding the request to such computing resource which belongs to the subinterval that contains the allotted random number,
wherein, if the load of a single set reaches the upper threshold, its priority number becomes maximal, indicating that the set receives incoming requests with highest priority.
2. The method according to claim 1 , wherein the bigger the priority number is, the longer is the length of the subinterval.
3. The method according to claim 1 , wherein, when a set of computing resources is in an inactive state or already scaled in, its priority number indicating zero priority.
4. The method according to claim 1 , wherein, if a new set is loaded, its priority number is dynamically changeable from a preset starting value up to its maximal value, dependent on the real-time traffic.
5. The method according to claim 1 , wherein the core network element is a Telecommunication Application Server, a Mobile Switching Server, a Mobility Management Entity, a Serving GPRS Support Node, or a call processing node.
6. An apparatus for resource management within a distributed system in a core network element, comprising:
at least one processor, and
at least one memory for storing instructions to be executed by the processor,
wherein the at least one memory and the instructions are configured to, with the at least one processor, cause the apparatus at least to perform
organizing computing resources of the core network element into sets, wherein a first set is always active, and wherein a load of incoming traffic is associated with each set;
controlling distribution of incoming traffic by setting an upper threshold and a lower threshold for the load of the sets, wherein the sets in operation are loaded as long as the average load of the sets in operation reaches the upper threshold, and when the upper threshold is exceeded, a new set is activated, whereas, when the load of the last activated set falls below the lower threshold, the last activated set is deactivated;
assigning a priority number to each set which provides an indication of an amount of traffic manageable by each set;
segmenting an interval of a group of random numbers for randomizing a request distribution into subintervals which are allocated to the computing resources of the active sets, wherein the length of each subinterval is determined based on the priority number of the set of the respective computing resource; and
allotting a random number out of the interval to an incoming request, and forwarding the request to such computing resource which belongs to the subinterval that contains the allotted random number,
wherein, if the load of a single set reaches the upper threshold, its priority number becomes maximal, indicating that the set receives incoming requests with highest priority.
7. The apparatus according to claim 6 , wherein the bigger the priority number is, the longer is the length of the subinterval.
8. The apparatus according to claim 6 , wherein, when a set of computing resources is in an inactive state or already scaled in, its priority number
indicating zero priority.
9. The apparatus according to claim 6 , wherein, if a new set is loaded, its priority number is dynamically changeable from a preset starting value up to its maximal value, dependent on the real-time traffic.
10. The apparatus according to claim 6 , wherein the core network element is a Telecommunication Application Server, a Mobile Switching Server, a Mobility Management Entity, a Serving GPRS Support Node, or a call processing node.
11. A computer program product embodied on a non-transitory computer-readable medium, comprising software code portions for performing the steps, when said product is run on a computer, of
organizing computing resources of the core network element into sets, wherein a first set is always active, and wherein a load of incoming traffic is associated with each set;
controlling distribution of incoming traffic by setting an upper threshold and a lower threshold for the load of the sets, wherein the sets in operation are loaded as long as the average load of the sets in operation reaches the upper threshold, and when the upper threshold is exceeded, a new set is activated, whereas, when the load of the last activated set falls below the lower threshold, the last activated set is deactivated;
assigning a priority number to each set which provides an indication of an amount of traffic manageable by each set;
segmenting an interval of a group of random numbers for randomizing a request distribution into subintervals which are allocated to the computing resources of the active sets, wherein the length of each subinterval is determined based on the priority number of the set of the respective computing resource; and
allotting a random number out of the interval to an incoming request, and forwarding the request to such computing resource which belongs to the subinterval that contains the allotted random number,
wherein, if the load of a single set reaches the upper threshold, its priority number becomes maximal, indicating that the set receives incoming requests with highest priority.
12. The computer program product according to claim 11 , wherein
the computer-readable medium is directly loadable into the internal memory of the computer and/or transmittable via a network by means of at least one of upload, download and push procedures.