Write Sharing Method for a Cluster Filesystem

A computer program product (CPP) for use with a computer data storage system including a plurality of client computers and a plurality of collector nodes, the CPP comprising:

• a set of storage device(s);
• and
• computer code stored collectively in the set of storage device(s), with the computer code including data and instructions to cause a processor(s) set to perform at least the following operations:
• receiving, by a first client computer of the plurality of client computers, a plurality of write requests for a file that is to be collectively stored in the plurality of collector nodes,
• storing the data from the plurality of write requests in a set of client side buffer(s) in a volatile memory of the first client computer,
• splitting the file into a plurality of data blocks,
• for each given write request of the plurality of write requests, determining the data blocks to which the data belongs,
• for each given data block, performing location-aware collector node selection by selecting a collector node, from the plurality of collector nodes, by querying a file system allocation subsystem to determine which collector node is assigned to serve the given data block, and
• for each given write request of the plurality of write requests, sending the given data, from the first client computer through a communication network to the selected collector nodes for the given write requests.

A computer system (CS) for use with a computer data storage system including a plurality of client computers and a plurality of collector nodes, the CS comprising:

• a processor(s) set;
• a set of storage device(s);
• and
• computer code stored collectively in the set of storage device(s), with the computer code including data and instructions to cause the processor(s) set to perform at least the following operations:
• receiving, by a first client computer of the plurality of client computers, a plurality of write requests for a file that is to be collectively stored in the plurality of collector nodes,
• storing the data from the plurality of write requests in a set of client side buffer(s) in a volatile memory of the first client computer,
• splitting the file into a plurality of data blocks,
• for each given write request of the plurality of write requests, determining the data blocks to which the data belongs,
• for each given data block, performing location-aware collector node selection by selecting a collector node, from the plurality of collector nodes, by querying a file system allocation subsystem to determine which collector node is assigned to serve the given data block, and
• for each given write request of the plurality of write requests, sending the given data, from the first client computer through a communication network to the selected collector nodes for the given write requests.