VIRTUAL NODE SUBPOOL MANAGEMENT

• US 20130290387A1
• Filed: 06/27/2013
• Published: 10/31/2013
• Est. Priority Date: 11/30/2010
• Status: Active Grant

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: Nagpal, Abhinay R., Patil, Sandeep Ramesh, Ramanathan, Sri, Trevathan, Matthew B.
• Granted Patent: US 9,092,452 B2
• Time in Patent Office: Days
• US Class Current: 707/822
• CPC Class Codes: G06F 16/188 Virtual file systems

Abstract

Virtual file system virtual nodes are grouped in subpools in response to identified resource components and managed as a function of their subpool groupings. Virtual nodes are decomposed into individual components linked to each other within each node. The components that have repetitive accesses by applications within the virtual file system and their respective frequencies of repetitive access are identified. Modules of linked occurrences of the repetitive components within each of the plurality of virtual nodes are formed, and subsets of the virtual nodes sharing common modules are grouped into subpools. Accordingly, in response to an application of the virtual file system requesting a service that is satisfied by a subpool common module, selection of a virtual node for reinitiating for reuse by the application is restricted to virtual nodes within the subpool associated with the common module.

Claims

1. A method for managing virtual nodes in response to identified component access frequencies, the method comprising:

ordering individual components decomposed from each of a plurality of virtual nodes within a virtual file system and linked to each other within the virtual nodes into root components as a function of the root components having at least three accesses by applications within the virtual file system, and of having highest frequencies of accesses by the applications relative to others of the individual components;

progressively depending others of the individual components that have at least three accesses from the roots as a function of having frequencies of the accesses by the applications that are lower than the highest frequencies of accesses;

omitting the individual components that each have less than three accesses by the applications; and

in response to an application of the virtual file system requesting a service that is satisfied by an inactive one of the virtual nodes, selecting the inactive virtual node and reinitiating the selected inactive virtual node for reuse by the application by locating, quiescing and returning the root components decomposed from the selected inactive virtual node and the others of the individual components progressively depending from the root components decomposed from the selected inactive virtual node, without quiescing the omitted components and any others of the components that are not decomposed from the selected inactive virtual node.

2. The method of claim 1, further comprising:

integrating computer-readable program code into a computer system comprising a processing unit, a computer readable memory and a computer readable storage medium, wherein the computer readable program code is embodied on the computer readable storage medium and comprises instructions that, when executed by the processing unit via the computer readable memory, cause the processing unit to perform the steps of ordering the individual components into the root components as the function of having at least three accesses by applications within the virtual file system and of having highest frequencies of accesses by the applications relative to others of the individual components, progressively depending the others of the individual components that have at least three accesses from the roots as the function of having frequencies of the accesses by the applications that are lower than the highest frequencies of accesses, omitting the individual components that each have less than three accesses by the applications, and selecting the inactive virtual node and reinitiating the selected inactive virtual node for reuse by locating, quiescing and returning the root components decomposed from the selected inactive virtual node and the others of the individual components progressively depending from the root components decomposed from the selected inactive virtual node without the omitted components and any others of the components that are not decomposed from the selected inactive virtual node.

3. The method of claim 1, wherein the virtual file system is a Unix architecture file system comprising a file system dependent/independent split just above a kernel inode layer, and the virtual nodes are file system independent inodes.

4. The method of claim 2, wherein the individual components are linked to each other within the virtual nodes in one of a max-heap structure and a min-heap structure;

the method further comprising;

identifying that the decomposed individual components have repetitive accesses by the applications within the virtual file system and their respective frequencies of the repetitive accesses by scanning virtual node logs that capture virtual node file component accesses by the applications;

iteratively rescanning the virtual node logs for each of plurality of samples; and

iteratively repeating the steps of ordering the individual decomposed linked to each other within the virtual nodes into the root components and progressively depending others of the individual components from the roots and omitting the individual components that each have less than three accesses by the application in response to each iterative rescanning of the virtual node logs for each of the plurality of samples.

5. A computer system, comprising:

a central processing unit, computer readable memory and a computer readable storage medium;

wherein the central processing unit, when executing program instructions stored on the tangible computer-readable storage medium via the computer readable memory;

orders individual components decomposed from each of a plurality of virtual nodes within a virtual file system and linked to each other within the virtual nodes into root components as a function of the root components having at least three accesses by applications within the virtual file system, and of having highest frequencies of accesses by the applications relative to others of the individual components;

progressively depends others of the individual components that have at least three accesses from the roots as a function of having frequencies of the accesses by the applications that are lower than the highest frequencies of accesses;

omits the individual components that each have less than three accesses by the applications; and

in response to an application of the virtual file system requesting a service that is satisfied by an inactive one of the virtual nodes, selects the inactive virtual node and reinitiates the selected inactive virtual node for reuse by the application by locating, quiescing and returning the root components decomposed from the selected inactive virtual node and the others of the individual components progressively depending from the root components decomposed from the selected inactive virtual node, without quiescing the omitted components and any others of the components that are not decomposed from the selected inactive virtual node.

6. The computer system of claim 5, wherein the virtual file system is a Unix architecture file system comprising a file system dependent/independent split just above a kernel inode layer, and the virtual nodes are file system independent inodes.

7. The computer system of claim 6, wherein the individual components are linked to each other within the virtual nodes in one of a max-heap structure and a min-heap structure;

and

wherein the central processing unit, when executing program instructions stored on the computer-readable storage medium via the computer readable memory, further;

identifies that the decomposed individual components have repetitive accesses by the applications within the virtual file system and their respective frequencies of the repetitive accesses by scanning virtual node logs that capture virtual node file component accesses by the applications;

iteratively rescans the virtual node logs for each of plurality of samples; and

iteratively repeats ordering the individual decomposed linked to each other within the virtual nodes into the root components and progressively depending others of the individual components from the roots and omitting the individual components that each have less than three accesses by the application in response to each iterative rescanning of the virtual node logs for each of the plurality of samples.

8. A computer program product for managing virtual nodes in response to identified component access frequencies, the computer program product comprising:

a computer readable storage medium having computer readable program code embodied therewith, the computer readable program code comprising instructions that, when executed by a computer processing unit, cause the computer processing unit to;

order individual components decomposed from each of a plurality of virtual nodes within a virtual file system and linked to each other within the virtual nodes into root components as a function of the root components having at least three accesses by applications within the virtual file system, and of having highest frequencies of accesses by the applications relative to others of the individual components;

progressively depend others of the individual components that have at least three accesses from the roots as a function of having frequencies of the accesses by the applications that are lower than the highest frequencies of accesses;

omit the individual components that each have less than three accesses by the applications; and

in response to an application of the virtual file system requesting a service that is satisfied by an inactive one of the virtual nodes, select the inactive virtual node and reinitiate the selected inactive virtual node for reuse by the application by locating, quiescing and returning the root components decomposed from the selected inactive virtual node and the others of the individual components progressively depending from the root components decomposed from the selected inactive virtual node, without quiescing the omitted components and any others of the components that are not decomposed from the selected inactive virtual node.

9. The computer program product of claim 8, wherein the virtual file system is a Unix architecture file system comprising a file system dependent/independent split just above a kernel inode layer, and the virtual nodes are file system independent inodes.

10. The computer program product of claim 9, wherein the individual components are linked to each other within the virtual nodes in one of a max-heap structure and a min-heap structure;

and

wherein the computer readable program code instructions, when executed by the computer processing unit, further cause the computer processing unit to;

identify that the decomposed individual components have repetitive accesses by the applications within the virtual file system and their respective frequencies of the repetitive accesses by scanning virtual node logs that capture virtual node file component accesses by the applications;

iteratively rescan the virtual node logs for each of plurality of samples; and

iteratively repeat ordering the individual decomposed linked to each other within the virtual nodes into the root components and progressively depending others of the individual components from the roots and omitting the individual components that each have less than three accesses by the application in response to each iterative rescanning of the virtual node logs for each of the plurality of samples.