1. A method of balancing load of an individual network element operating in a pool in a core network comprising at least one pool of at least two network elements and a plurality of Radio Access Network nodes controlled by said network elements, wherein a user equipment is assigned to a network element from said pool of network elements based on relative weighting values associated with said network elements and the method comprises:
periodically determining load of said network element; and
reducing the relative weighting value if said network element is overloaded for a number of periods exceeding a limit of allowed consecutive overloads.
2. The method according to claim 1 further comprising increasing the relative weighting value if said network element is underloaded for a number of periods exceeding a limit of allowed consecutive underloads.
3. The method according to claim 1, comprising the steps of:
comparing the load with an upper limit;
incrementing a first counter if the upper limit is exceeded and erasing said first counter otherwise;
comparing the first counter with said limit of allowed consecutive overloads; and
reducing the relative weighting value if said limit of allowed consecutive overloads is exceeded.
4. The method according to claim 2, comprising the steps of:
comparing the traffic load with a lower limit;
incrementing a second counter if the traffic load is below the lower limit and erasing said second counter otherwise;
comparing the second counter with said limit of allowed consecutive underloads; and increasing the relative weighting value if said limit of allowed consecutive underloads is exceeded.
5. The method according to claim 2, wherein the relative weighting value of the network element overloaded for a number of periods exceeding said limit of allowed consecutive overloads is decreased if there is at least one network element in the pool, which can accommodate load from said network element and the relative weighting value of the network element underloaded for a number of periods exceeding said limit of allowed consecutive underloads is increased if there is at least one network element in the pool, which is not underloaded.
6. The method according to claim 2, further comprising marking network elements for which the number of consecutive overloads exceeded said limit of allowed consecutive overloads as overloaded network elements and marking network elements for which the number of consecutive underloads exceeded said limit of allowed consecutive underloads as underloaded network elements, wherein the relative weighting of an overloaded network element is decreased if the number of network elements marked as overloaded in said pool is below a third limit and the relative weighting of an underloaded network element is increased if the number of network elements marked as underloaded in said pool is below a fourth limit.
7. The method according to claim 6, comprising the steps of:
comparing the load with an upper limit;
incrementing a first counter if the upper limit is exceeded and erasing said first counter otherwise;
comparing the first counter with said limit of allowed consecutive overloads; and
marking the network element as overloaded if said limit of allowed consecutive overloads is exceeded.
8. The method according to claim 6, comprising the steps of:
comparing the traffic load with a lower limit;
incrementing a second counter if the traffic load is below the lower limit and erasing said second counter otherwise;
comparing the second counter with said limit of allowed consecutive underloads; and
marking the network element as underloaded if said limit of allowed consecutive underloads is exceeded.
9. The method according to claim 1, wherein the network element is a Mobile Switching Centre or a Serving GPRS Support Node.
10. The method according to claim 1, wherein the Radio Access Node is a Radio Base Station or a Radio Network Controller.
11. The method according to claim 6, wherein the third and fourth limits are determined as a percentage of the total number of network elements in the pool.
12. A network element operating in a pool of at least two network elements in a core network, said network element is adapted to control a plurality of Radio Access Network nodes and has associated a relative weighting value; the network element comprises means for periodic determining its traffic load, a memory unit, a control unit for comparing the traffic load with an upper limit stored in said memory, a first counter for recording number of consecutive occasions of exceeding the upper limit, wherein said control unit is adapted to reduce the relative weighting value if said network element is overloaded for a number of periods exceeding a limit of allowed consecutive overloads.
13. The network element according to claim 12, further comprising a second counter and said memory is adapted to store a lower limit, wherein the control unit is adapted to compare the traffic load with said lower limit stored in said memory and said second counter is adapted to record number of consecutive occasions of the traffic load being below the lower limit, wherein said control unit is adapted to increase the relative weighting value if said network element is underloaded for a number of periods exceeding a limit of allowed consecutive underloads.
14. The network element according to claim 12, adapted to transmit its load figure to other network elements in its pool and to receive load figures from other network elements in its pool.
15. The network element according to claim 12, further comprising a third counter for recording number of network elements in the pool which are overloaded for a number of periods exceeding said limit of allowed consecutive overloads and a fourth counter for recording number of network elements in the pool which are underloaded for a number of periods exceeding said limit of allowed consecutive underloads.
16. The network element according to claim 15, wherein the control unit is adapted to decrease the relative weighting value of said network element if the value of the third counter is below a third limit and to increase the relative weighting value of said network element if the value of the fourth counter is below a fourth limit.
17. The network element according to claim 12, wherein said network element is a Mobile Switching Centre or a Serving GPRS Support Node.
18. A core network controlling radio telecommunications network comprising at least one pool of at least two network elements controlling a plurality of Radio Access Network nodes, wherein a user equipment is assigned to a network element from said pool of network elements based on relative weighting values associated with said network elements, and wherein balancing load of an individual network element operating in the pool is carried out by means for performing:
periodically determining load of said network element; and
reducing the relative weighting value if said network element is overloaded for a number of periods exceeding a limit of allowed consecutive overloads.
19. A network management system controlling a core network, said core network comprising a plurality of network elements operating in a pool of at least two network elements, wherein an individual network element is adapted to control a plurality of Radio Access Network nodes and has associated a relative weighting value stored in said network management system; the network management system comprises means for periodic determining traffic load of individual network elements, a memory unit, a control unit for comparing the traffic load of said individual network element with an upper limit stored in said memory, a first counter for recording number of consecutive occasions of exceeding the upper limit, wherein said control unit is adapted to reduce the relative weighting value if said network element is overloaded for a number of periods exceeding a limit of allowed consecutive overloads.
20. The network management system according to claim 19, further comprising a second counter and said memory is adapted to store a lower limit, wherein the control unit is adapted to compare the traffic load with said lower limit stored in said memory and said second counter is adapted to record number of consecutive occasions of the traffic load being below the lower limit, wherein said control unit is adapted to increase the relative weighting value if said network element is underloaded for a number of periods exceeding a limit of allowed consecutive underloads.
21. The network management system according to claim 19 further comprising a third counter for recording number of overloaded network elements in the pool and a fourth counter for recording number of underloaded network elements in the pool.
22. The network management system according to claim 21, wherein the control unit is adapted to decrease the relative weighting value of said network element if the value of the third counter is below a third limit and to increase the relative weighting value of said network element if the value of the fourth counter is below a fourth limit.
The claims below are in addition to those above.
All refrences to claim(s) which appear below refer to the numbering after this setence.
1. A semiconductor integrated circuit comprising:
a fixed frequency-division clock generation unit configured to generate a fixed frequency-division clock with a fixed frequency based on an output clock of a clock source;
a variable frequency-division clock generation unit configured to generate a variable frequency-division clock with a variable frequency based on the output clock of the clock source; and
a data path selection unit configured to select a data path as a data path for transferring data between a first functional module operating based on the fixed frequency-division clock and a second functional module operating based on the variable frequency-division clock,
wherein, while the variable frequency-division clock is generated by the variable frequency-division clock generation unit, the data path selection unit selects a data path using a synchronization unit for converting the data into clock-synchronous data on a receiving side, and
wherein, while the variable frequency-division clock is not generated by the variable frequency-division clock generation unit, the data path selection unit selects a data path without using the synchronization unit.
2. The semiconductor integrated circuit according to claim 1, wherein the variable frequency-division clock generation unit generates a thinned-out clock in which toggles are thinned out from the output clock of the clock source, and generates the variable frequency-division clock based on the thinned-out clock.
3. The semiconductor integrated circuit according to claim 1, wherein the first functional module is a memory controller for controlling a DRAM.
4. A semiconductor integrated circuit comprising:
a fixed frequency-division clock generation unit configured to generate a fixed frequency-division clock with a fixed frequency based on an output clock of a first clock source;
a variable frequency-division clock generation unit configured to generate a variable frequency-division clock with a variable frequency by selecting either a clock based on the first clock source or a clock based on a second clock source for outputting a clock with a frequency lower than the frequency of the clock based on the first clock source; and
a data path selection unit configured to select a data path as a data path for transferring data between a first functional module operating based on the fixed frequency-division clock and a second functional module operating based on the variable division clock,
wherein, while the variable frequency-division clock is generated by the variable frequency-division clock generation unit, the data path selection unit selects a data path using a synchronization unit for converting the data into clock-synchronous data on a receiving side, and
wherein, while the variable frequency-division clock is not generated by the variable frequency-division clock generation unit, the data path selection unit selects a data path without using the synchronization unit.
5. The semiconductor integrated circuit according to claim 4, wherein the clock based on the first clock source is a first output clock of the first clock source or a frequency-division clock generated based on the first output clock, and
wherein the clock based on the second clock source is a second output clock of the second clock source or a frequency-division clock generated based on the second output clock.
6. The semiconductor integrated circuit according to claim 4, wherein, in a case where the variable frequency-division clock generation unit selects the clock based on the second clock source, the fixed frequency-division clock generated based on the output clock of the first clock source and the variable frequency-division clock generated based on the clock based on the second clock source are in an asynchronous relation.
7. The semiconductor integrated circuit according to claim 4, wherein the first functional module is a memory controller for controlling a Dynamic Random Access Memory (DRAM).
8. A clock control method in a semiconductor integrated circuit, the method comprising:
generating a fixed frequency-division clock with a fixed frequency based on an output clock of a clock source;
generating a variable frequency-division clock with a variable frequency based on the output clock of the clock source; and
selecting a data path as a data path for transferring data between a first functional module operating based on the fixed frequency-division clock and a second functional module operating based on the variable frequency-division clock,
wherein, while the variable frequency-division clock is generated by the generating a variable frequency-division clock, the selecting selects a data path using a synchronization unit for converting the data into clock-synchronous data on a receiving side, and
wherein, while the variable frequency-division clock is not generated by the generating a variable frequency-division clock, the selecting selects a data path without using the synchronization unit.
9. A clock control method in a semiconductor integrated circuit, the method comprising:
generating a fixed frequency-division clock with a fixed frequency based on an output clock of a first clock source;
generating a variable frequency-division clock with a variable frequency by selecting either a clock based on the first clock source or a clock based on a second clock source for outputting a clock with a frequency lower than the frequency of the clock based on the first clock source; and
selecting a data path as a data path for transferring data between a first functional module operating base on the fixed frequency-division clock and a second functional module operating based on the variable frequency-division clock,
wherein, while the variable frequency-division clock is generated by the generating a variable frequency-division clock, the selecting selects a data path using a synchronization unit for converting the data into clock-synchronous data on a receiving side, and
wherein, while the variable frequency-division clock is not generated by the generating a variable frequency-division clock, the selecting selects a data path without using the synchronization unit.