US2008295098A1PendingUtilityA1
System Load Based Dynamic Segmentation for Network Interface Cards
Est. expiryJul 29, 2025(expired)· nominal 20-yr term from priority
H04L 69/161H04L 69/16H04L 69/12
53
PatentIndex Score
0
Cited by
0
References
0
Claims
Abstract
The present invention provides a computer implemented method, system, and computer program product of dynamic segmentation based on processor load using various steps. If a network interface card is enabled for segmentation, a determination is made as to whether a processor utilization value is greater than a threshold. If processor utilization value is above the threshold, segments are offloaded to the network interface card, wherein segments are sent to the network interface card.
Claims
exact text as granted — not AI-modified1 . A computer implemented method of dynamic segmentation based on processor load, the computer implemented method comprising:
responsive to a network interface card being enabled for segmentation, determining whether a processor utilization value is greater than a threshold; and responsive to the processor utilization value being above the threshold, offloading segmentation to the network interface card, wherein segments are sent to the network interface card.
2 . The computer implemented method of claim 1 further comprising:
responsive to the processor utilization value being below or equal to the threshold, processing segments normally using a transmission control protocol stack.
3 . The computer implemented method of claim of claim 1 , wherein the offloading step comprises
creating a segment having a size greater than normally processed for segmentation; and sending the segment to the network interface card for segmentation.
4 . The computer implemented method of claim 3 , wherein the segment having a size greater than normally processed for segmentation is greater than a maximum segment size of an interface protocol.
5 . The computer implemented method of claim 4 , wherein the interface protocol is TCP/IP.
6 . The computer implemented method of claim 4 , wherein the interface protocol is user datagram protocol.
7 . The computer implemented method of claim 1 , wherein segments segmented by the network interface card during offloading segmentation are compatible with a standard Ethernet protocol.
8 . The computer implemented method of claim 1 , wherein segments segmented by the network interface card during offloading segmentation are compatible with a jumbo frame Ethernet protocol.
9 . The computer implemented method of claim 1 , wherein the threshold is user specified.
10 . The computer implemented method of claim 1 , wherein a default value of the threshold is between 70 and 100 percent processor utilization.
11 . The computer implemented method of claim 1 , wherein said determining further comprises polling for processor utilization.
12 . The computer implemented method of claim 1 , wherein the processor utilization value is an average to prevent excessive switching between offloading segmentation.
13 . The computer implemented method of claim, 10 wherein the processor utilization value is a timed average.
14 . The computer implemented method of claim 1 , further comprising determining whether the network interface card is a microcode based network interface card or a non-microcode based network interface card.
15 . The computer implemented method of claim 14 , further comprising evaluating performance limitations of the microcode based network interface card, wherein segments are not processed if the microcode based network interface card performs segmentation slower than normal segmentation processing.
16 . A system for dynamically offloading segmentation based on processor workload comprising:
a processor for running an operating system, wherein a processor load is indicated by a processor utilization value, wherein the operating system includes a stack for controlling segmentation; a storage operably connected to the processor for storing the operating system wherein the operating system may be loaded into a main memory for execution by the processor; and a network interface card operably connected to the processor and the storage, wherein the network interface card is enabled for segmentation, wherein if a processor utilization value is above a threshold, segments are sent to the network interface card for segmentation offloading.
17 . The system of claim 16 , wherein the stack for controlling segmentation is a TCP stack.
18 . A computer program product comprising a computer usable medium including computer usable program code for dynamic segmentation based on processor load, said computer program product including:
responsive to a network interface card being enabled for segmentation, computer usable program code for determining whether a processor utilization value is greater than a threshold; and responsive to the processor utilization value being above the threshold, computer usable program code for offloading segmentation to the network interface card, wherein segments are sent to the network interface card.
19 . The computer program product of claim 18 , wherein computer usable program code for offloading segmentation further comprises:
computer usable code for creating a segment having a size greater than normally processed for segmentation; and computer usable code for sending the segment to the network interface card for segmentation.
20 . The computer program product of claim 18 , further comprising:
responsive to the processor utilization value being below or equal to the threshold, computer usable code for processing segments normally using a TCP stack.Join the waitlist — get patent alerts
Track US2008295098A1 — get alerts on status changes and closely related new filings.
We store only your email — no account needed. See our privacy policy.