IP Library Granted Patent US 12,237,831
Granted Patent B2
US 12,237,831 · App. 18/198,122 · Granted Feb 25, 2025

Network-on-chip (NOC) with flexible data width

Inventors: Sharath Raghava (Los Gatos, CA); Ankireddy Nalamalpu (Portland, OR); Dheeraj Subbareddy (Portland, OR); Harsha Gupta (Sunnyvale, CA); James Ball (San Jose, CA); Kavitha Prasad (San Jose, CA); Sean R. Atsatt (Santa Cruz, CA)
Assignee: Intel Corporation
H03K19/17736H03K19/17796H04L41/5019H04L41/5003
View Patent ↗
Loading inventors, assignments & file history…
Monitor This Case
Get email alerts when status or documents change.
Order Certified Copies
Most orders are placed with the USPTO same day — all within 24 business hours.
Order via The Patent Place →
Pre-filled with this patent's details
Quick Facts
Patent No.
US 12,237,831
App. No.
18/198,122
Granted
Feb 25, 2025
Kind
B2
Abstract

Techniques described herein may relate to providing a programmable interconnect network (e.g., a programmable network-on-chip (NOC)). A method may include determining a transmission parameter, bonding one or more channels of an interconnect network based at least in part on the transmission parameter, and power-gating any unused channels after the bonding.

Claims (23)

1. A multi-die system, comprising:

a processing circuit disposed on a first die; and

a network-on-chip disposed at least in part on the first die, wherein the network-on-chip is configurable to facilitate memory transactions between the first die and a second die, wherein the network-on-chip comprises a plurality of data lanes, wherein the network-on-chip is configurable to transmit data between the first die and the second die via a bus formed from the plurality of data lanes, and wherein the network-on-chip is configurable to reduce power consumed at least in part by reducing a width of the bus based on traffic latency and bandwidth.

2. The multi-die system of claim 1 , wherein the network-on-chip is configurable to couple to memory disposed external to the network-on-chip.

3. The multi-die system of claim 1 , wherein reducing the width of the bus comprises reducing the width of the bus from 16 lanes to 8 lanes.

4. The multi-die system of claim 1 , wherein the network-on-chip comprises a first set of data lanes of the plurality of data lanes and a second set of data lanes of the plurality of data lanes.

5. The multi-die system of claim 4 , wherein the network-on-chip is configurable to reduce the width of the bus by using the first set of data lanes to transmit the data while the second set of data lanes is not used.

6. A method, comprising:

operating a programmable interconnect network at a first data width, wherein the programmable interconnect network is configurable to span multiple die of a same package;

determining, based on memory latency, to reduce a power state of the programmable interconnect network; and

reducing the power state of the programmable interconnect network at least in part by operating the programmable interconnect network at a second data width less than the first data width.

7. The method of claim 6 , wherein operating the programmable interconnect network at the first data width comprises operating a plurality of data lanes of the programmable interconnect network to transmit data having the first data width.

8. The method of claim 6 , comprising determining to control the power state of the programmable interconnect network based on usage of the programmable interconnect network based on a setting.

9. The method of claim 6 , wherein the first data width comprises 16 lanes, and wherein the second data width comprises 8 lanes.

10. The method of claim 6 , wherein the second data width is half that of the first data width.

11. The method of claim 6 , comprising transmitting, via the programmable interconnect network, a memory transaction between a first die and a second die.

12. An integrated circuit, comprising:

a first set of lanes of a plurality of lanes corresponding to a network-on-chip; and

a second set of lanes of the plurality of lanes corresponding to the network-on-chip, wherein the second set of lanes and the first set of lanes are configurable to facilitate memory transactions between a first die and a second die, and wherein using the first set of lanes to transmit data while the second set of lanes is inactive is based on memory latency and reduces power consumed by the network-on-chip.

13. The integrated circuit of claim 12 , wherein the network-on-chip is configurable to reduce the power consumed based on usage of the network-on-chip.

14. The integrated circuit of claim 13 , wherein the first die corresponds to an external system.

15. The integrated circuit of claim 12 , wherein the first set of lanes comprises 8 lanes, and wherein the second set of lanes comprises 8 lanes.

16. The integrated circuit of claim 12 , wherein the first set of lanes and the second set of lanes are both used to transmit the data in response to traffic of the network-on-chip.

Assignments (2)
SECURITY INTEREST Recorded Sep 12, 2025
From: ALTERA CORPORATION
To: BARCLAYS BANK PLC, AS COLLATERAL AGENT
Reel/Frame 073431/0309 →
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Jan 8, 2024
From: INTEL CORPORATION
To: ALTERA CORPORATION
Reel/Frame 066055/0412 →
Continuity (4)
Continuation 17556917 · Dec 20, 2021
Continuation 17033524 · Sep 25, 2020
Continuation 16234212 · Dec 27, 2018
Related Publication 20230370068A1 · Nov 16, 2023
References Cited (22)
US 7106760B1 · Perumal et al. · 2006 [cited by applicant]
US 8411696B1 · Ko et al. · 2013 [cited by applicant]
US 8464088B1 · Nguyen et al. · 2013 [cited by applicant]
US 8693314B1 · Horton et al. · 2014 [cited by applicant]
US 9105316B2 · Jones · 2015 [cited by examiner]
US 9432298B1 · Smith · 2016 [cited by examiner]
US 10503690B2 · Swarbrick · 2019 [cited by examiner]
US 20060104315A1 · Volkening et al. · 2006 [cited by applicant]
US 20060182139A1 · Bugajski et al. · 2006 [cited by applicant]
US 20070140286A1 · Kraus · 2007 [cited by applicant]
US 20100080561A1 · Paranjape et al. · 2010 [cited by applicant]
US 20110135046A1 · Jones · 2011 [cited by examiner]
US 20120027026A1 · Duwel et al. · 2012 [cited by applicant]
US 20120128045A1 · Ling et al. · 2012 [cited by applicant]
US 20130010585A1 · Chu et al. · 2013 [cited by applicant]
US 20130235739A1 · Mamidwar et al. · 2013 [cited by applicant]
US 20170170153A1 · Khare · 2017 [cited by examiner]
US 20190103872A1 · Clark · 2019 [cited by examiner]
US 20190363717A1 · Swarbrick · 2019 [cited by examiner]
Burd, Thomas, et al.; “Zeppelin': An SoC for Multichip Architectures,” IEEE Journal of Solid State Circuits, vol. 54, No. Oct. 26, 2018, 11 pgs. (Year: 2018). [cited by examiner]
Beck, Noah, et al.; “Zeppelin': An SoC for Multichip Architectures,” 2018 IEEE International Solid State Circuits Conference, Digest of Technical Papers, Feb. 12, 2018, pp. 40-42. [cited by applicant]
Burd, Thomas, et al.; “Zepplin': An SoC for Multichip Architectures,” IEEE Journal of Solid State Circuits, vol. 54, No. Oct. 26, 2018, 11 pgs. [cited by applicant]