Apparatus and methods for managing packets in a broadband data stream

• US 6,741,562 B1
• Filed: 12/15/2000
• Issued: 05/25/2004
• Est. Priority Date: 12/15/2000
• Status: Expired due to Term

• Current Assignee: Tellabs Operations Incorporated (Infinera Corp.)
• Original Assignee: Tellabs San Jose, Inc. (Infinera Corp.)
• Inventors: Keirouz, Joe, Liston, Robert Edward, Sy, Simeon Ngo
• Primary Examiner(s): Hsu, Alpus H.
• Assistant Examiner(s): Tran, Thien D
• Application Number:
  US09/737,916
• Time in Patent Office: 1,257 Days
• Field of Search: 370/395.1, 370/395.52, 370/397.7, 370/470, 370/471, 370/472, 370/230, 370/395.6
• US Class Current: 370/230
• CPC Class Codes: H04L 47/125 by balancing the load, e.g. traffic engineering
• CPC Class Codes: H04L 47/36 by determining packet size, e.g. maximum transfer unit [MTU]
• CPC Class Codes: H04L 49/90 Buffering arrangements
• CPC Class Codes: H04L 49/901 using storage descriptor, e.g. read or write pointers
• CPC Class Codes: H04L 49/9047 including multiple buffers, e.g. buffer pools
• CPC Class Codes: H04L 49/9052 with buffers of different sizes
• CPC Class Codes: H04L 49/9094 Arrangements for simultaneous transmit and receive, e.g. simultaneous reading/writing from/to the storage element

Abstract

An exemplary method for writing packets in a data stream comprises the steps of dividing a packet into long packet cells and short packet cells, storing the long packet cells in a set of long cell queues and the short packet cells in a set of short cell queues, selectively pairing a long packet cell in one of the set of long cell queues with a short packet cell in one of the set of short cell queues to obtain an optimized pair of packet cells at each write cycle, and sequentially writing the optimized pair of packet cells at each write cycle to a set of memory banks. An exemplary method for reading packets in a data stream comprises the steps of receiving long cell requests and short cell requests, storing the long cell requests in a set of long cell request queues, storing the short cell requests in a set of short cell request queues, selectively pairing a long cell request in one of the set of long cell request queues and a short cell request in one of the set of short cell request queues to obtain an optimized pair of cell requests at each read cycle, retrieving packet cells corresponding to the optimized pair of cell requests from a set of memory banks at each read cycle, reordering the packet cells in a first data buffer, and sending the packet cells in a sequential order.

Claims

1. A method for reading packets in a data stream, comprising the steps of:

receiving long cell requests and short cell requests;

storing said long cell requests in a set of long cell request queues;

storing said short cell requests in a set of short cell request queues;

selectively pairing a long cell request in one of said set of long cell request queues and a short cell request in one of said set of short cell request queues to obtain an optimized pair of cell requests at each read cycle;

retrieving packet cells corresponding to said optimized pair of cell requests from a set of memory banks at each read cycle;

reordering said packet cells in a first data buffer; and

sending said packet cells in a sequential order.

2. The method of claim 1, wherein said reordering step includes the steps of:

assigning a tag number to each cell request, said tag number associating said cell request to a long cell request queue or a short cell request queue; and

reordering said packet cells based on said tag number.

3. The method of claim 1, wherein said reordering step includes the steps of:

storing said packet cells in said first data buffer; and

organizing said packet cells in a sequential order in said first data buffer.

4. The method of claim 3, wherein said organizing step includes the step of using a tag number index to reorder said packet cells in said sequential order.

5. The method of claim 1, further comprising the steps of:

organizing multiple packets comprising packet cells in a second data buffer; and

sending said multiple packets as a frame.

6. The method of claim 5, wherein said multiple packets include packets of a first size and packets of a second size and said multiple packets are organized and sent as a frame.

7. The method of claim 1, further comprising the steps of:

monitoring said long cell request queues and said short cell request queues; and

retrieving a packet cell corresponding to a stale cell request out of order after said stale cell request remains in a respective cell request queue for a predetermined amount of time.

8. The method of claim 7, further comprising the steps of:

maintaining a cell request index;

calculating an index difference between a top long cell request and a top short cell request based on said cell request index; and

determining a stale cell request based on said index difference.