Method and system for libfabric atomics-based lockless cluster-wide shared memory access API in a distributed system

A computer system, comprising:

• a processor;
• and
• a memory coupled to the processor and storing instructions which, when executed by the processor, cause the processor to perform a method, the method comprising:
• allocating, in a distributed system comprising a plurality of nodes, a plurality of portions of memory which comprise shared remote memory content;
• registering the allocated portions with an operating system to be accessed via remote direct memory access (RDMA);
• accessing, by a first node, the allocated portions to obtain a local copy of the shared remote memory content;
• and
• performing an atomic operation on one or more bits of the shared remote memory content via libfabric atomic (LFA) application programming interface (API) calls, which comprises:
• updating the one or more bits of the shared remote memory content based on a new value and an offset;
• retrieving, from the shared remote memory content based on the offset, a current value of the one or more bits prior to the updating;
• and
• performing an action on the shared remote memory content based on a comparison of the retrieve current value with an expected value in the local copy;
• wherein the plurality of nodes comprise server nodes and client nodes,
• wherein the server nodes allocate the plurality of portions of memory,
• wherein the first node is a client node or a server node,
• wherein the client nodes perform the atomic operation by attaching, in a first allocated portion based on the new value and the offset, to a shared object which corresponds to the one or more bits;
• wherein the shared remote memory content is accessed by a filesystem associated with the distributed system,
• wherein the filesystem is a fabric attached memory filesystem (FAMfs),
• wherein the server nodes are I/O nodes in the FAMfs,
• wherein allocating the plurality of portions of memory is performed by allocator modules associated with each I/O node in the FAMfs, and
• wherein the remote shared memory content comprises a bitmap data structure which tracks a status, as used or free, of physical extents in the memory of the FAMfs.

A non-transitory computer-readable storage medium storing instructions that when executed by a computer cause the computer to perform a method, the method comprising:

• allocating, in a distributed system comprising a plurality of nodes, a plurality of portions of memory which comprise shared remote memory content;
• registering the allocated portions with an operating system to be accessed via remote direct memory access (RDMA);
• accessing, by a first node, the allocated portions to obtain a local copy of the shared remote memory content;
• and
• performing an atomic operation on one or more bits of the shared remote memory content via libfabric atomic (LFA) application programming interface (API) calls, which comprises:
• updating the one or more bits of the shared remote memory content based on a new value and an offset;
• retrieving, from the shared remote memory content based on the offset, a current value of the one or more bits prior to the updating;
• and
• performing an action on the shared remote memory content based on a comparison of the retrieved current value with an expected value on the local copy;
• wherein the plurality of nodes comprise server nodes and client nodes,
• wherein the server nodes allocate the plurality of portions of memory,
• wherein the first node is a client node or a server node,
• wherein the client nodes perform the atomic operation by attaching, in a first allocated portion based on the new value and the offset, to a shared object which corresponds to the one or more bits;
• wherein the shared remote memory content is accessed by a filesystem associated with the distributed system,
• wherein the filesystem is a fabric attached memory filesystem (FAMfs),
• wherein the server nodes are I/O nodes in the FAMfs,
• wherein allocating the plurality of portions of memory is performed by allocator modules associated with each I/O node in the FAMfs, and
• wherein the remote shared memory content comprises a bitmap data structure which tracks a status, as used or free, of physical extents in the memory of the FAMfs.