Active-active scale-out for unified data path architecture

A data storage system comprising a first storage processor and a second storage processor, each of the first storage processor and the second storage processor including a respective set of processing units and memory, the memory coupled to the respective set of processing units and storing executable instructions, which when executed on the set of processing units cause each of the first and second storage processors to:

• operate, on each of a first storage processor and a second storage processor of the data storage system, a clustered storage pool, each clustered storage pool provisioning storage units to data objects and maintaining consistency such that changes to a clustered storage pool on one of the storage processors are reflected in the clustered storage pool on the other of the storage processors;
• operate a clustered file system presenting a single namespace shared between the first storage processor and the second storage processor, the clustered file system provisioned with storage units from the clustered storage pool and storing a data object realized in a form of a file of the clustered file system;
• and
• in response to receiving IO requests, provide read and write access to the data object in an active-active arrangement,
• wherein the first storage processor and the second storage processor are front end storage processors, and wherein the data storage system further comprises a set of back end storage processors selectively coupled to the front end storage processors through a switch fabric for effecting read and write operations of the data object on a set of physical storage devices.

A computer program product having a non-transitory computer-readable medium including instructions which, when executed by control circuitry of a data storage system, cause the control circuitry to perform a method of processing IO requests in a data storage system, the method comprising:

• operating, on each of a first storage processor and a second storage processor of the data storage system, a clustered storage pool, each clustered storage pool provisioning storage units to data objects and maintaining consistency such that changes to a clustered storage pool on one of the storage processors are reflected in the clustered storage pool on the other of the storage processors;
• operating, on each of the first storage processor and the second storage processor, a clustered file system presenting a single namespace shared between the first storage processor and the second storage processor, the clustered file system provisioned with storage units from the clustered storage pool and storing a data object realized in a form of a file of the clustered file system;
• and
• in response to receiving IO requests, providing read and write access to the data object by both the first storage processor and the second storage processor in an active-active arrangement,
• wherein the first storage processor and the second storage processor are front end storage processors, and wherein the data storage system further includes a set of back end storage processors selectively coupled to the front end storage processors through a switch fabric, and wherein the method further comprises:
• processing a set of IO requests directed to the data object by multiple ones of the front end storage processors;
• transmitting processed versions of the set of IO requests through the switch fabric to one of the set of back end storage processors using a block-based protocol;
• and
• further processing the processed versions of the IO requests by the back end storage processor to effect one of a read operation and a write operation of the data object on a physical storage device.