Dynamic packet training

Abstract

A packet control mechanism for a computer data system that dynamically adjusts packet training depending on the utilization load on the processor. The dynamic adjustment of packet training can be to enable and disable packet training, or adjust the number of packets in the packet train. In preferred embodiments, the computer data system includes a processor utilization mechanism that indicates a load on a processor. When the packet control mechanism determines the load on the processor is above a threshold limit, the packet control mechanism reduces the processor load by compressing the packets into the packet train. The compressing of the packets is stopped or reduced when the processor load is below a threshold in order to increase the data throughput on the network interface.

Description

CROSS-REFERENCE TO PARENT APPLICATION[0001]This patent application is a continuation of U.S. Ser. No. 11 / 106,011 filed on Apr. 14, 2005, which is incorporated herein by reference.BACKGROUND OF THE INVENTION[0002]1. Technical Field[0003]This invention generally relates to data processing and communications, and more specifically relates to dynamically transmitting data packets in a packet train on a computer network or computer communication link.[0004]2. Background Art[0005]Computer systems communicate with each other over computer networks. Such networks include multiple nodes, which are typically computers, that may be distributed over vast distances and connected by communications links. Nodes in the computer network communicate with each other using data packets sent over the communication links. The data packets are the basic units of information transfer. A data packet contains data surrounded by control and routing information supplied by the various nodes.[0006]Sending, rece...

AI Technical Summary

Benefits of technology

[0010]According to the preferred embodiments, a computer data system includes a packet control mechanism that dynamically adjusts packet training depending on the utilization load on the processor. The dynamic adjustment of packet training can be to enable and disable packet training, or adjust the number of packets in the packet train. In preferred embodiments, the computer data system includes a processor utilization mechanism that indicates a load on a processor. When the packet control mechanism determines the load on the processor is above a threshold limit, the packet control mechanism reduces the processor load by processing the packets into a packet train. The training of the packets is stopped or reduced when the processor load is below a threshold in order to increase the data throughput on the network interface.

Problems solved by technology

Sending, receiving, and processing of packets have an overhead, or associated cost.

But in times of light packet-traffic, the packets that the node accumulates experience poor performance while waiting in vain for additional packets to arrive, and the ultimate timer expiration introduces additional processing overhead.

This method provides dynamic training of packets but does not efficiently handle message latency, particularly for burst mode communication traffic in a low CPU utilization environment.

Packet training can save a significant amount of CPU load in a heavy communications workload environment.

However, packet training can have a detrimental affect on the latency of messages sent over the network.

Thus there is a tradeoff between CPU load and communication latency when using packet training.

Packet training decreases the load on the CPU but may increase the time for a message to be sent over the network due to the delay in building a train of packets.

Without a way to optimize the tradeoff between CPU loading and network latency, the computer industry will continue to suffer from sub-optimum performance from a packet data network.

Images