System and method for address space slicing with strong CPU core affinity
AssigneeEMC IP Holding Company LLC
InventorsRonen Gazit, Vladimir Shveidel
Life of this patent
Estimating the term…
Abstract
A method, computer program product, and computer system for partitioning, by a computing device, resources between a plurality of central processing unit (CPU) cores. A logical block address (LBA) space of a user may be divided into a slice with an affinity to a CPU core of the plurality of CPU cores. IO flow processing may be processed by the CPU core of the plurality of CPU cores associated with the LBA space divided into the slice.
Claims17
A computer-implemented method comprising:
- partitioning, by a computing device, resources between a plurality of central processing unit (CPU) cores;
- dividing a logical block address (LBA) space of a user into a plurality of slices, wherein each slice of the plurality of slices has an affinity to a respective CPU core of the plurality of CPU cores, wherein dividing the LBA space into the plurality of slices includes providing even slice distribution between the plurality of CPU cores, wherein dividing includes ignoring a pre-determined number of bits of at least one LBA address of the LBA addresses of the LBA address space;
- and
- processing IO flow processing by the respective CPU core of the plurality of CPU cores associated with the LBA space divided into the plurality of slices, wherein the resources between the plurality of CPU cores are a function of the LBA space such that when the IO flow processing is processed by the respective CPU core associated with the LBA space, access to the resources between the plurality of CPU cores are processed only from the same respective CPU core associated with the LBA space.
A computer program product residing on a non-transitory computer readable storage medium having a plurality of instructions stored thereon which, when executed across one or more processors, causes at least a portion of the one or more processors to perform operations comprising:
- partitioning resources between a plurality of central processing unit (CPU) cores;
- dividing a logical block address (LBA) space of a user into a plurality of slices, wherein each slice of the plurality of slices has an affinity to a respective CPU core of the plurality of CPU cores, wherein dividing the LBA space into the plurality of slices includes providing even slice distribution between the plurality of CPU cores, wherein dividing includes ignoring a pre-determined number of bits of at least one LBA address of the LBA addresses of the LBA address space;
- and
- processing IO flow processing by the respective CPU core of the plurality of CPU cores associated with the LBA space divided into the plurality of slices, wherein the resources between the plurality of CPU cores are a function of the LBA space such that when the IO flow processing is processed by the respective CPU core associated with the LBA space, access to the resources between the plurality of CPU cores are processed only from the same respective CPU core associated with the LBA space.
A computing system including one or more processors and one or more memories configured to perform operations comprising:
- partitioning resources between a plurality of central processing unit (CPU) cores;
- dividing a logical block address (LBA) space of a user into a plurality of slices, wherein each slice of the plurality of slices has an affinity to a respective CPU core of the plurality of CPU cores, wherein dividing the LBA space into the plurality of slices includes providing even slice distribution between the plurality of CPU cores, wherein dividing includes ignoring a pre-determined number of bits of at least one LBA address of the LBA addresses of the LBA address space;
- and
- processing IO flow processing by the respective CPU core of the plurality of CPU cores associated with the LBA space divided into the plurality of slices, wherein the resources between the plurality of CPU cores are a function of the LBA space such that when the IO flow processing is processed by the respective CPU core associated with the LBA space, access to the resources between the plurality of CPU cores are processed only from the same respective CPU core associated with the LBA space.
Dependencies are inferred from the claim text.