Back to results
US 10,761,752 B1Granted
Memory pool configuration for allocating memory in a distributed network
OwnerKumar Narayanan, Saravanan Purushothaman +1inventors
InventorsKumar Narayanan, Saravanan Purushothaman, Ramanand Narayanan
Life of this patent
Estimating the term…
Abstract
An example operation may include one or more of allocating a memory node to a computing memory allocated to one or more computing nodes, allocating a software process dedicated to controlling the memory node, and modifying an existing file system of the one or more computing nodes to replace the existing file system with a new file system that includes the memory node.
Claims16
A method, comprising:
- allocating a memory node to a computing memory allocated to one or more computing nodes;
- allocating a software process dedicated to controlling the memory node;
- modifying an existing file system of the one or more computing nodes to replace the existing file system with a new file system that comprises the memory node;
- assigning, via a memory file system, file-system proxies to identified file systems operating in the computing nodes;
- assigning two or more of the computing nodes to the computing memory, and wherein the memory node comprises a plurality of solid state disks;
- identifying a data block request and an acknowledgement;
- writing the acknowledgment to the memory file system as metadata;
- identifying an aggregated set of write transactions;
- updating the memory node with the aggregated set of write transactions;
- and
- closing an active session prior to updating the memory node.
8 more elements
An apparatus, comprising:
- a processor configured to
- allocate a memory node to a computing memory allocated to one or more computing nodes;
- allocate a software process dedicated to controlling the memory node;
- modify an existing file system of the one or more computing nodes to replace the existing file system with a new file system that comprises the memory node;
- identify a data set requiring data analysis;
- identify the one or more computing nodes to perform the data analysis;
- forward metadata to each of the one or more computing nodes, the metadata including an indicator that identifies the data set as being stored in each of the computing nodes;
- identify file input/output (I/O) memory access commands intended for each of the computing nodes;
- and
- forward the I/O memory access commands to the memory node.
7 more elements
A non-transitory computer readable storage medium configured to store instructions that when executed cause a processor to perform:
- allocating a memory node to a computing memory allocated to one or more computing nodes;
- allocating a software process dedicated to controlling the memory node;
- modifying an existing file system of the one or more computing nodes to replace the existing file system with a new file system that comprises the memory node;
- assigning, via a memory file system, file-system proxies to identified file systems operating in the computing nodes;
- assigning two or more of the computing nodes to the computing memory, and wherein the memory node comprises a plurality of solid state disks;
- identifying a data block request and an acknowledgement;
- writing the acknowledgment to the memory file system as metadata;
- identifying an aggregated set of write transactions;
- updating the memory node with the aggregated set of write transactions;
- and
- closing an active session prior to updating the memory node.
8 more elements
Dependencies are inferred from the claim text.