Resource allocation using restore credits
Systems and methods for allocating resources are disclosed. Resources such as streams are allocated using restore credits. Credits are issued to the clients in a manner that ensure the system is operating in a safe allocation state. The credits can be used not only to allocate resources but also to throttle clients where necessary. Credits can be granted fully, partially, and in a number greater than requested. Zero or negative credits can also be issued to throttle clients. Restore credits are associated with reads and may be allocated by determining how many credits a CPU/cores can support. This maximum number may be divided amongst clients connected with the server.
1. A method for allocating resources of a server to clients connected to the server and restoring data from the server, the method comprising:
receiving a request for restore credits from a client, wherein the restore credits each correspond to an amount of data to read from the server;
determining a number of credits available to the client, wherein the number of credits available to the client accounts for restore credits already issued to the client;
issuing credits based on the request and the number of credits available to the client and
dynamically adapting a size of a client cache based on the number of restore credits issued to the client.
2. The method of claim 1 , further comprising at least one of:
issuing restore credits equal to a number of restore credits requested by the client;
issuing restore credits greater than the number of restore credits requested by the client;
issuing restore credits less than the number of restore credits requested by the client;
issuing zero restore credits to the client; or
issuing negative restore credits to the client.
3. The method of claim 1 , further comprising including a prefetch flag in the request, wherein the prefetch flag indicates that the data at the server is accessed sequentially by the client.
4. The method of claim 1 , further comprising determining a number of reads that consumes 1% of a CPU on average.
5. The method of claim 4 , further comprising determining an average per core reads allowed based on the number of reads that consume 1% of the CPU and the average free percentage of the CUP and determining a maximum credits per client based on the number of client connections.
6. The method of claim 5 , further comprising determining a calculated number of credits per client based on a tuning factor that is applied to the maximum credits per client.
7. The method of claim 6 , further comprising issuing the restore credits in a number equal to a minimum of the request or the calculated number of credits.
8. The method of claim 6 , further comprising issuing the restore credits in a number equal to a maximum of the request or the calculated number of credits.
9. The method of claim 1 , further comprising throttling all clients when an average free percentage per core of a processor is less than a predetermined threshold.
10. A non-transitory computer readable medium including computer executable instructions for implementing a method, when executed, for allocating resources of a server to clients connected to the server and restoring data from the server, the method comprising:
receiving a request for restore credits from a client, wherein the restore credits each correspond to an amount of data to read from the server;
determining a number of credits available to the client, wherein the number of credits available to the client accounts for restore credits already issued to the client;
issuing credits based on the request and the number of credits available to the client and
dynamically adapting a size of a client cache based on the number of restore credits issued to the client.
11. The non-transitory computer readable medium of claim 10 , further comprising at least one of:
issuing restore credits equal to a number of restore credits requested by the client;
issuing restore credits greater than the number of restore credits requested by the client;
issuing restore credits less than the number of restore credits requested by the client;
issuing zero restore credits to the client; or
issuing negative restore credits to the client.
12. The non-transitory computer readable medium of claim 10 , further comprising including a prefetch flag in the request, wherein the prefetch flag indicates that the data at the server is accessed sequentially by the client.
13. The non-transitory computer readable medium of claim 10 , further comprising determining a number of reads that consume 1% of a CPU on average.
14. The non-transitory computer readable medium of claim 13 , further comprising determining an average per core reads allowed based on the number of reads that consume 1% of the CPU and the average free percentage of the CUP and determining a maximum credits per client based on the number of client connections.
15. The non-transitory computer readable medium of claim 14 , further comprising determining a calculated number of credits per client based on a tuning factor that is applied to the maximum credits per client.
16. The non-transitory computer readable medium of claim 14 , further comprising issuing the restore credits in a number equal to a minimum of the request or the calculated number of credits.
17. The non-transitory computer readable medium of claim 14 , further comprising issuing the restore credits in a number equal to a maximum of the request or the calculated number of credits.