# US 2008/0107105 A1 · System and method for communicating on a richly connected multi-processor computer system using a pool of buffers for dynamic association with a virtual channel

> Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt

- Number: US 2008/0107105 A1
- Document ID: US20080107105A1
- Status: Published application
- Kind: A1
- Application number: 11/594,405
- Filed: 2006-11-08
- Published: 2008-05-08
- Earliest claimed: 2006-11-08
- Assignees: SiCortex, Inc.
- Inventors: Matthew H. Reilly; Nitin Godiwala; Judson S. Leonard
- Web page: https://patentgrep.com/patent/US20080107105A1

## CPC

- H04L 49/9047: including multiple buffers, e.g. buffer pools
- H04L 45/10: {Routing in connection-oriented networks, e.g. X.25 or ATM}
- H04L 49/90: Buffering arrangements
- H04L: TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION (arrangements common to telegraphic and telephonic communication H04M)

## Abstract

Systems and methods for communicating on a richly-connected multiprocessor computer system using a pool of buffers for dynamic association with a virtual channel. Packets are communicated in a multiprocessor computer system having a large plurality of processing nodes interconnected by a defined interconnection topology, in which a communication from a source processing node to a target processing node may pass through one or more intermediate nodes en route to the target processing node. A set of virtual channels is associated for each link in the interconnection topology. A first subset of buffers is dedicated for fixed correspondence to virtual channel identifiers, and a second subset of buffers is dedicated for dynamic allocation and assignment to virtual channels.

## Claims

### Claim 1

A method of communicating packets in a multiprocessor computer system having a large plurality of processing nodes interconnected by a defined interconnection topology, in which a communication from a source processing node to a target processing node may pass through one or more intermediate nodes en route to the target processing node, the method comprising:
for each link in the interconnection topology associating a set of virtual channels, each virtual channel having an associated virtual channel identifier;
for each communication between a source processing node and a target processing node, assigning an virtual channel to convey the communication;
at a processing node providing a set of buffers to associate with virtual channels, a first subset of buffers being dedicated for fixed correspondence to virtual channel identifiers, and a second subset of buffers being dedicated for dynamic allocation and assignment to virtual channels;
in response to a communication request with an associated virtual channel, providing one of the fixed buffers or dynamically allocated buffers to store packet data for the communication.

### Claim 2 (depends on 1)

The method of claim 1 wherein the processing node records the order of delivery for the communication utilizing the buffer from the second subset and records the virtual channel for the communication.

### Claim 3 (depends on 1)

The method of claim 1 wherein the size of the first subset and the size of the second subset are dynamically configurable.

### Claim 4 (depends on 1)

The method of claim 1 wherein the number of buffers allocated to a virtual channel is dynamically allocated and varies depending on traffic load.

### Claim 5

A large scale multiprocessor computer system, comprising:
a large plurality of processing nodes;
a defined interconnection topology having links interconnecting the processing nodes;
wherein each processing node includes
link logic to transmit data on, and receive data from, links connected to the processing node,
buffers to store communications, including buffers dedicated for fixed correspondence to virtual channel identifiers, and buffers for dynamic allocation and assignment to virtual channels; and
wherein, in response to a communication request with an associated virtual channel, one of the fixed buffers or dynamically allocatable buffers stores communication data.

### Claim 6 (depends on 5)

The system of claim 5 wherein the processing node includes logic to record the order of receipt for all communications in the fixed or allocatable buffers associated with a particular virtual channel.

### Claim 7 (depends on 5)

The system of claim 5 wherein the size of the first subset and the size of the second subset are dynamically configurable.
