Region to host affinity for block allocation in clustered file system volume

1. A method, performed by a host computer having a processor, to allocate a resource from a clustered file system (CFS) volume stored on one or more physical storage devices to a file, the CFS volume comprising resources organized into resource clusters, and the resource clusters make up regions, the method comprising:

for each region of the resource clusters:

determining a first count of resources allocated to the host computer and a second count of resources allocated to all other host computers; and

calculating a region weight based on the first count and the second count;

sorting a list of the regions based on their region weights that starts from the greatest region weight to the least region weight;

selecting a region at or near the start of the list; and

allocating the resource from a resource cluster in the selected region to the file.

2. The method of claim 1 , wherein the region weight is based on magnifying the second count and determining a difference between the first count and the magnified second count.

3. The method of claim 1 , wherein:

the region weight is calculated based on

generating a first product by multiplying a number of resource clusters in each region with a multiplier factor,

generating a first sum by adding to the first product to the first count of resources allocated to the host computer,

generating a second product by multiplying the second count of resources allocated to other host computers with a magnifier factor, and

subtracting the second sum from the first sum.

4. The method of claim 1 , further comprising selecting the resource cluster in the selected region to evenly distribute files in all the resource clusters.

5. The method of claim 4 , wherein selecting the resource cluster comprises:

for each resource cluster, determining an affinity count of files allocated from the resource cluster;

determining an average affinity count of files allocated from all the resource clusters in the CFS volume; and

selecting the resource cluster in the region whose affinity count is less than or equal to the average affinity count.

6. A non-transitory, computer-readable storage medium encoded with instructions executable by a processor to implement a hypervisor on a host computer, to allocate a resource from a clustered file system (CFS) volume stored on one or more physical storage devices to a file, the CFS volume comprising resources organized into resource clusters, and the resource clusters make up regions, the instructions comprising:

for each region of the resource clusters:

determining a first count of resources allocated to the host computer and a second count of resources allocated to all other host computers; and

calculating a region weight based on the first count and the second count;

sorting a list of the regions based on their region weights that starts from the greatest region weight to the least region weight;

selecting a region at or near the start of the list; and allocating the resource from a resource cluster in the selected region to the file.

7. The non-transitory, computer-readable storage medium of claim 6 , wherein the region weight is based on magnifying the second count and determining a difference between the first count and the magnified second count.

8. The non-transitory, computer-readable storage medium of claim 7 , wherein:

the region weight is calculated based on

generating a first product by multiplying a number of resource clusters in each region with a multiplier factor,

generating a first sum by adding to the first product to the first count of resources allocated to the host computer,

generating a second product by multiplying the second count of resources allocated to other host computers with a magnifier factor, and

subtracting the second sum from the first sum.

9. The non-transitory, computer-readable storage medium of claim 7 , wherein the instructions further comprise selecting the resource cluster in the selected region to evenly distribute files in all the resource clusters.

10. The non-transitory, computer-readable storage medium of claim 9 , wherein selecting the resource cluster comprises:

for each resource cluster, determining an affinity count of files allocated from the resource cluster;

determining an average affinity count of files allocated from all the resource clusters in the CFS volume; and

selecting the resource cluster in the region whose affinity count is less than or equal to the average affinity count.

11. A computer having a processor to allocate a resource from a clustered file system (CFS) volume stored on one or more physical storage devices to a file, the CFS volume comprising resources organized into resource clusters, and the resource clusters make up regions, wherein the computer is configured to:

for each region of the resource clusters:

determine a first count of resources allocated to the computer and a second count of resources allocated to all other computers; and

calculate a region weight based on the first count and the second count;

sort a list of the regions based on their region weights that starts from the greatest region weight to the least region weight;

select a region at or near the start of the list; and

allocate the resource from a resource cluster in the selected region to the file.

12. The computer of claim 11 , wherein the region weight is based on magnifying the second count and determining a difference between the first count and the magnified second count.

13. The computer of claim 12 , wherein:

the region weight is calculated based on

generating a first product by multiplying a number of resource clusters in each region with a multiplier factor, and

generating a first sum adding to the first product to the first count of resources allocated to the computer,

generating a second product by multiplying the second count of resources allocated to other computers with a magnifier factor, and

subtracting the second sum from the first sum.

14. The computer of claim 11 , wherein the computer is further configured to select the resource cluster in the selected region to evenly distribute files in all the resource clusters.

15. The computer of claim 14 , wherein the computer selects the resource cluster by:

for each resource cluster, determine an affinity count of files allocated from the resource cluster;

determine an average affinity count of files allocated from all the resource clusters in the CFS volume; and

select the resource cluster in the region whose affinity count is less than or equal to the average affinity count.