Systems and methods for managing the processing of relatively large data objects in a communications stack

• US 5,920,703 A
• Filed: 02/19/1997
• Issued: 07/06/1999
• Est. Priority Date: 02/19/1997
• Status: Expired due to Term

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: Rajaraman, Bala, Overby, Linwood Hugh Jr., Miyake, Masashi E., Long, Lynn Douglas, Campbell, Scott Allan, Gugliotta, Dale C.
• Primary Examiner(s): Harrell, Robert B.
• Application Number:
  US08/802,597
• Time in Patent Office: 867 Days
• Field of Search: 364/DIG. 1 MS File, 364/DIG. 2 MS File, 395/200.3, 395/200.43, 395/200.44, 395/200.47, 395/200.48, 395/200.53, 395/200.61, 395/200.62, 395/200.66, 395/200.8
• US Class Current: 709/236
• CPC Class Codes: H04L 49/90 Buffering arrangements
• CPC Class Codes: H04L 69/22 Parsing or analysis of headers
• CPC Class Codes: H04L 69/32 Architecture of open systems interconnection [OSI] 7-layer type protocol stacks, e.g. the interfaces between the data link level and the physical level
• CPC Class Codes: H04L 9/40 Network security protocols

Abstract

Management of the processing of relatively large data objects in a communications stack having multiple layers improves the performance in a communications system in preparing relatively large data objects for transmission across a communications network. This reduces or eliminates data movement and copying during segmentation of the relatively large data objects into relatively small data objects, and appendage of headers to the relatively small data object segments during processing in the communications stack. A shared storage manager creates and controls multiple tokens representing multiple images of portions of the relatively large data objects to enable separate scheduling of the multiple images from the same storage unit or buffer to be passed from one layer in a communications stack to the next lower layer in the communications stack. The large data object is segmented into a plurality of relatively small data object segments at one or more of the layers in the communications stack. When dictated by the communications stack, header segments are also created for each relatively small data object segment. Each header segment contains specific information for the relatively small data object segment to which the header corresponds. A separate buffer list is generated by the communications stack for each relatively small data object segment and for each corresponding header segment.

Claims

1. A method for managing processing of relatively large data objects in a communications stack having multiple layers, said communications stack generally requiring segmentation of the relatively large data objects as the large data objects proceed through the layers of said communications stack and appendage of headers containing protocol specific information to each segment of data, said method comprising the steps of:

receiving a relatively large data object from an application for processing in said communications stack;

segmenting said relatively large data object into a plurality of relatively small data objects without physically moving the relatively large data object and without creating a copy of the large data object;

generating a header containing protocol specific information for each segment of relatively small data objects such that each header is uniquely associated with one segment of small data objects, without physically moving the small data object, without creating a copy of the large data object, and without creating a copy of the associated small data object; and

discontiguously passing said plurality of relatively small data objects and said plurality of associated headers from one layer in said communications stack to a second layer in said communications stack, whereby each of said plurality of small data objects is passed to said second layer in said communications stack discontiguously from its associated header and without physically moving any of the small data objects and its associated header and without creating a copy of any of the small data objects and its associated header;

whereby said headers are added at different layers in the communications stack without physically moving any of the large data objects, the small data objects and the associated headers and without creating a copy of any of the large data objects, small data objects and associated headers.

2. A method for managing processing of relatively large data objects in a communications stack according to claim 1 wherein the discontiguously passing step comprises the steps of:

generating a buffer list for each small data object segment created by said segmenting step, said buffer list uniquely associated with a small data object segment and the header corresponding to the small data object segment; and

passing each buffer list from one layer in said communications stack to a second layer in said communications stack, whereby each of said plurality of small data objects is passed to said second layer in said communications stack discontiguously from its associated header and without physically moving any of the small data objects and its associated header and without creating a copy of any of the small data objects and its associated header.

3. A method for managing processing of relatively large data objects in a communications stack according to claim 2 wherein said generating a buffer list step comprises the steps of:

generating a primary token for representing the relatively large data object, said primary token being uniquely associated with the corresponding large data object;

generating a secondary token for representing a logical image of each small data object segment, each secondary token being uniquely associated with a corresponding small data object segment;

generating a header pointer for referencing a header associated with a particular small data object segment, each header pointer uniquely referencing the header corresponding to the particular small data object segment;

storing the primary token in each separate buffer list;

storing each secondary token in a separate buffer list, each buffer list corresponding to the same small data object segment to which the secondary token corresponds; and

storing each header pointer in a separate buffer list having the secondary token stored therein which corresponds to the same small data object segment to which the header uniquely referenced by the header pointer corresponds.

4. A method for managing processing of relatively large data objects in a communications stack according to claim 1 wherein the receiving step comprises the step of receiving the relatively large data object from an application processing in said communications stack without receiving any information relating to the relationship between the relatively large data object and segmentation requirements of the communications stack.

5. A method for managing processing of relatively large data objects in a communications stack according to claim 1 further comprising the steps of:

discontiguously receiving the plurality of small data object segments at the second layer in said communications stack;

segmenting each of said received small data object segments into a plurality of smaller data object segments without physically moving the received small data object segments and without creating a copy of the received small data object segments, each smaller data object segment being relatively smaller in size than the received small data object segment from which it was segmented;

generating a smaller data object header containing protocol specific information for each smaller data object segment such that each smaller data object header is uniquely associated with one smaller data object segment, without physically moving the smaller data object segment and with creating a copy of the smaller data object segment; and

discontiguously passing said plurality of smaller data object segments and said plurality of associated smaller data object headers from the second layer in said communications stack to a third layer in said communications stack, whereby each of said plurality of smaller data object segments is passed to said third layer in said communications stack discontiguously from its associated smaller data object header and without physically moving any of the smaller data object segments and its associated smaller data object header and without creating a copy of any of the smaller data object segments and its associated smaller data object header.

6. A method for managing processing of relatively large data objects according to claim 1 wherein said communications stack is a communications stack implemented in a virtual telecommunications access method (VTAM) system.

7. A system for managing processing of a relatively large data object in a communications stack having multiple layers, said communications stack generally requiring segmentation of the relatively large data object as the large data object proceeds through the layers of said communications stack and appendage of headers containing protocol specific information to each segment of data, said system comprising:

means for segmenting said relatively large data object into a plurality of relatively small data objects without requiring physical movement of the relatively large data object and without requiring creation of a copy of the large data object;

means for generating a header containing protocol specific information for each segment of relatively small data objects without requiring physical movement of the small data object, without requiring creation of a copy of the large data object, and without requiring creation of a copy of the associated small data object, and for uniquely associating the generated header with one segment of small data objects; and

means for discontiguously passing said plurality of relatively small data objects and said plurality of associated headers from one layer in said communications stack to a second layer in said communications stack, whereby each of said plurality of small data objects may be passed to said second layer in said communications stack discontiguously from its associated header and without requiring physical movement of any of the small data objects and its associated header and without requiring creation of a copy of any of the small data objects and its associated header.

8. A system for managing processing of a relatively large data object in a communications stack according to claim 7 wherein said means for discontiguously passing comprises:

means for generating a buffer list for each small data object segment created by said segmenting means, said buffer list being uniquely associated with a small data object segment and the header corresponding to the small data object segment; and

means for passing each buffer list from one layer in said communications stack to a second layer in said communications stack, whereby each of said plurality of small data objects may be passed to said second layer in said communications stack discontiguously from its associated header and without requiring physical movement of any of the small data objects and its associated header and without requiring creation of a copy of any of the small data objects and its associated header.

9. A system for managing processing of a relatively large data object in a communications stack according to claim 8 wherein said means for generating a buffer list comprises:

means for generating a primary token for representing the relatively large data object, said primary token being uniquely associated with the corresponding large data object;

means for generating a secondary token for representing a logical image of each small data object segment, each secondary token being uniquely associated with a corresponding small data object segment;

means for generating a header pointer for referencing a header associated with a particular small data object segment, each header pointer uniquely referencing the header corresponding to the particular small data object segment;

means for storing the primary token in each separate buffer list;

means for storing each secondary token in a separate buffer list, each buffer list corresponding to the same small data object segment to which the secondary token corresponds; and

means for storing each header pointer in the separate buffer list having the secondary token stored therein which corresponds to the same small data object segment to which the header uniquely referenced by the header pointer corresponds.

10. A system for managing processing of relatively large data objects in a communications stack according to claim 7 further comprising means for receiving the relatively large data object from an application processing in said communications stack without receiving any information relating to the relationship between the relatively large data object and segmentation requirements of the communications stack.

11. A system for managing processing of relatively large data objects in a communications stack according to claim 7 further comprising:

means for discontiguously receiving the plurality of small data object segments at the second layer in said communications stack;

means for segmenting each of said received small data object segments into a plurality of smaller data object segments without requiring physical movement of the received small data object segments and without requiring creation of a copy of the received small data object segments, each smaller data object segment being relatively smaller in size than the received small data object segment from which it was segmented;

means for generating a smaller data object header containing protocol specific information for each smaller data object segment such that each smaller data object header is uniquely associated with one smaller data object segment, without requiring physical movement of the smaller data object segment and without requiring creation of a copy of the smaller data object segment; and

means for discontiguously passing said plurality of smaller data object segments and said plurality of associated smaller data object headers from the second layer in said communications stack to a third layer in said communications stack, whereby each of said plurality of smaller data object segments is passed to said third layer in said communications stack discontiguously from its associated smaller data object header and without requiring physical movement of any of the smaller data object segments and its associated smaller data object header and without requiring creation of a copy of any of the smaller data object segments and its associated smaller data object header.

12. A system for managing processing of relatively large data objects according to claim 7 wherein said communications stack is a communications stack implemented in a virtual telecommunications access method (VTAM) system.

13. A computer program product for managing processing of a relatively large data object in a communications stack having multiple layers, said communications stack generally requiring segmentation of the relatively large data object as the large data object proceeds through the layers of said communications stack and appendage of headers containing protocol specific information to each segment of data, said computer program product comprising:

a computer readable storage medium having computer readable code means embodied in said medium, said computer readable code means comprising;

computer instruction means for segmenting said relatively large data object into a plurality of relatively small data objects without requiring physical movement of the relatively large data object and without requiring creation of a copy of the large data object;

computer instruction means for generating a header containing protocol specific information for each segment of relatively small data objects without requiring physical movement of the small data object, without requiring creation of a copy of the large data object, and without requiring creation of a copy of the associated small data object, and for uniquely associating the generated header with one segment of small data objects; and

computer instruction means for discontiguously passing said plurality of relatively small data objects and said plurality of associated headers from one layer in said communications stack to a second layer in said communications stack, whereby each of said plurality of small data objects may be passed to said second layer in said communications stack discontiguously from its associated header and without requiring physical movement of any of the small data objects and its associated header and without requiring creation of a copy of any of the small data objects and its associated header.

14. A computer program product for managing processing of a relatively large data object in a communications stack according to claim 13 wherein said computer instruction means for discontiguously passing comprises:

computer instruction means for generating a buffer list for each small data object segment created by said computer instruction means for segmenting, said buffer list being uniquely associated with a small data object segment and the header corresponding to the small data object segment; and

computer instruction means for passing each buffer list from one layer in said communications stack to a second layer in said communications stack, whereby each of said plurality of small data objects may be passed to said second layer in said communications stack discontiguously from its associated header and without requiring physical movement of any of the small data objects and its associated header and without requiring creation of a copy of any of the small data objects and its associated header.

15. A computer program product for managing processing of a relatively large data object in a communications stack according to claim 14 wherein computer instruction means for generating a buffer list comprises:

computer instruction means for generating a primary token for representing the relatively large data object, said primary token being uniquely associated with the corresponding large data object;

computer instruction means for generating a secondary token for representing a logical image of each small data object segment, each secondary token being uniquely associated with a corresponding small data object segment;

computer instruction means for generating a header pointer for referencing a header associated with a particular small data object segment, each header pointer uniquely referencing the header corresponding to the particular small data object segment;

computer instruction means for storing the primary token in each separate buffer list;

computer instruction means for storing each secondary token in a separate buffer list, each buffer list corresponding to the same small data object segment to which the secondary token corresponds; and

computer instruction means for storing each header pointer in the separate buffer list having the secondary token stored therein which corresponds to the same small data object segment to which the header uniquely referenced by the header pointer corresponds.

16. A computer program product for managing processing of relatively large data objects in a communications stack according to claim 13 further comprising computer instruction means for receiving the relatively large data object from an application processing in said communications stack without receiving any information relating to the relationship between the relatively large data object and segmentation requirements of the communications stack.

17. A computer program product for managing processing of relatively large data objects in a communications stack according to claim 13 further comprising:

computer instruction means for discontiguously receiving the plurality of small data object segments at the second layer in said communications stack;

computer instruction means for segmenting each of said received small data object segments into a plurality of smaller data object segments without requiring physical movement of the received small data object segments and without requiring creation of a copy of the received small data object segments, each smaller data object segment being relatively smaller in size than the received small data object segment from which it was segmented;

computer instruction means for generating a smaller data object header containing protocol specific information for each smaller data object segment such that each smaller data object header is uniquely associated with one smaller data object segment, without requiring physical movement of the smaller data object segment and without requiring creation of a copy of the smaller data object segment; and

computer instruction means for discontiguously passing said plurality of smaller data object segments and said plurality of associated smaller data object headers from the second layer in said communications stack to a third layer in said communications stack, whereby each of said plurality of smaller data object segments is passed to said third layer in said communications stack discontiguously from its associated smaller data object header and without requiring physical movement of any of the smaller data object segments and its associated smaller data object header and without requiring creation of a copy of any of the smaller data object segments and its associated smaller data object header.

18. A computer program product for managing processing of relatively large data objects according to claim 13 wherein said communications stack is a communications stack implemented in a virtual telecommunications access method (VTAM) system.