IP Library › Granted Patent US 12,386,506
Granted Patent B2
US 12,386,506 · App. 18/488,581 · Granted Aug 12, 2025

Tagged memory operated at lower VMIN in error tolerant system

Inventors: Nitin Chawla (Noida, IN); Giuseppe Desoli (San Fermo della Battaglia, IT); Anuj Grover (New Delhi, IN); Thomas Boesch (Rovio, CH); Surinder Pal Singh (Noida, IN); Manuj Ayodhyawasi (Noida, IN)
Assignees: STMICROELECTRONICS S.r.l.; STMicroelectronics International N.V.
G06F3/0604G06F3/0619G06F3/0655G06F3/0679G06N3/08
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 12,386,506
App. No.
18/488,581
Granted
Aug 12, 2025
Kind
B2
Abstract

A memory array arranged as a plurality of memory cells. The memory cells are configured to operate at a determined voltage. A memory management circuitry coupled to the plurality of memory cells tags a first set of the plurality of memory cells as low-voltage cells and tags a second set of the plurality of memory cells as high-voltage cells. A power source provides a low voltage to the first set of memory cells and provides a high voltage to the second set of memory cells based on the tags.

Claims (58)

1. A method of implementing an artificial neural network, the method comprising:

initiating performance of a process using a first subset of memory cells of a set of a memory cells of a memory array to store data at a first operational voltage that corresponds to a first operational accuracy of the process, wherein the set of memory cells of the memory array are arranged as a plurality of rows of memory cells intersecting a plurality of columns of memory cells and the first subset of memory cells includes a subset of columns in the plurality of columns of memory cells, a subset of rows in the plurality of rows of memory cells, or both a subset of columns in the plurality of columns of memory cells and a subset of rows in the plurality of rows of memory cells;

detecting an event associated with the process, wherein the event is a transition from processing associated with a first layer of the artificial neural network to processing associated with a second layer of the artificial neural network; and

in response to detection of the event, modifying the performance of the process to correspond to a second operational accuracy that is higher than the first operational accuracy, wherein the modifying the performance of the process includes:

providing, to the first subset of memory cells, a second operational voltage that is higher than the first operational voltage; or

initiating performance of the process using a second subset of memory cells of the set of memory cells of the memory array to store data at a third operational voltage that corresponds to the second operational accuracy of the process.

2. The method of claim 1 , wherein modifying the performance of the process includes:

providing, to the first subset of memory cells, the second operational voltage that is higher than the first operational voltage.

3. The method of claim 1 , wherein modifying the performance of the process includes:

initiating performance of the process using the second subset of memory cells of the set of memory cells of the memory array to store data at the third operational voltage that corresponds to the second operational accuracy of the process.

4. The method of claim 1 , wherein second and third operational voltages are higher than the first operational voltage.

5. The method of claim 1 , wherein modifying the performance of the process includes:

initiating performance of a second process using the plurality of memory cells to store data at the first operational accuracy.

6. The method of claim 1 , comprising:

simultaneously providing different operational voltages to the first and second subsets of memory cells of the memory array.

7. A non-transitory computer-readable medium having contents that cause a processing device to perform a method of implementing an artificial neural network, the method comprising:

initiating performance of a process using a first subset of memory cells of set of memory cells of a memory array to store data at a first operational voltage that corresponds to a first operational accuracy of the process, wherein the set of memory cells of the memory array are arranged as a plurality of rows of memory cells intersecting a plurality of columns of memory cells and the first subset of memory cells includes a subset of columns in the plurality of columns of memory cells, a subset of rows in the plurality of rows of memory cells, or both a subset of columns in the plurality of columns of memory cells and a subset of rows in the plurality of rows of memory cells;

detecting an event associated with the process, wherein the event is a transition from processing associated with a first layer of the artificial neural network to processing associated with a second layer of the artificial neural network; and

in response to detection of the event, modifying the performance of the process to correspond to a second operational accuracy that is higher than the first operational accuracy, wherein the modifying the performance of the process includes:

providing, to the first subset of memory cells, a second operational voltage that is higher than the first operational voltage; or

initiating performance of the process using a second subset of memory cells of the set of memory cells of the memory array to store data at a third operational voltage that corresponds to the second operational accuracy of the process.

8. The non-transitory computer-readable medium of claim 7 , wherein modifying the performance of the process includes:

providing, to the first subset of memory cells, the second operational voltage that is higher than the first operational voltage.

9. The non-transitory computer-readable medium of claim 7 , wherein modifying the performance of the process includes:

initiating performance of the process using the second subset of memory cells of the set of memory cells of the memory array to store data at the third operational voltage that corresponds to the second operational accuracy of the process.

10. The non-transitory computer-readable medium of claim 7 , wherein second and third operational voltages are higher than the first operational voltage.

11. The non-transitory computer-readable medium of claim 7 , wherein modifying the performance of the process includes:

initiating performance of a second process using the plurality of memory cells to store data at the first operational accuracy.

12. The non-transitory computer-readable medium of claim 7 , wherein the contents comprise instructions executed by the processing device.

13. A device, comprising:

a memory array arranged as a plurality of rows of memory cells intersecting a plurality of columns of memory cells, wherein the memory cells, in operation, store data; and

processing circuitry coupled to the memory array, wherein the processing circuitry, in operation, implements an artificial neural network, the implementing the artificial neural network including:

initiating performance of a process using a first subset of memory cells of the memory to store data at a first operational voltage that corresponds to a first operational accuracy of the process, wherein the first subset of memory cells includes a subset of columns in the plurality of columns of memory cells, a subset of rows in the plurality of rows of memory cells, or both a subset of columns in the plurality of columns of memory cells and a subset of rows in the plurality of rows of memory cells;

detecting an event associated with the process, wherein the event is a transition from processing associated with a first layer of the artificial neural network to processing associated with a second layer of the artificial neural network; and

in response to detection of the event, modifying the performance of the process to correspond to a second operational accuracy that is higher than the first operational accuracy, wherein the modifying the performance of the process includes:

providing, to the first subset of memory cells, a second operational voltage that is higher than the first operational voltage; or

initiating performance of the process using a second subset of memory cells of the memory array to store data at a third operational voltage that corresponds to the second operational accuracy of the process.

14. The device of claim 13 , wherein modifying the performance of the process includes:

providing, to the first subset of memory cells, the second operational voltage that is higher than the first operational voltage.

15. The device of claim 13 , wherein modifying the performance of the process includes:

initiating performance of the process using the second subset of memory cells of the memory array to store data at the third operational voltage that corresponds to the second operational accuracy of the process.

16. The device of claim 13 , wherein second and third operational voltages are higher than the first operational voltage.

17. The device of claim 13 , wherein modifying the performance of the process includes:

initiating performance of a second process using the plurality of memory cells to store data at the first operational accuracy.

18. A system, comprising:

memory including a memory array arranged as a plurality of rows of memory cells intersecting a plurality of columns of memory cells, wherein the memory cells of the memory array, in operation, store data; and

processing circuitry including memory management circuitry, wherein the memory management circuitry is coupled to the memory array, and the memory management circuitry, in operation:

initiates performance of a process using a first subset of memory cells of the memory to store data at a first operational voltage that corresponds to a first operational accuracy of the process, wherein the first subset of memory cells includes a subset of columns in the plurality of columns of memory cells, a subset of rows in the plurality of rows of memory cells, or both a subset of columns in the plurality of columns of memory cells and a subset of rows in the plurality of rows of memory cells;

detects an event associated with the process; and

in response to detection of the event, modifies the performance of the process to correspond to a second operational accuracy that is higher than the first operational accuracy, wherein the modifying the performance of the process includes:

providing, to the first subset of memory cells, a second operational voltage that is higher than the first operational voltage; or

initiating performance of the process using a second subset of memory cells of the memory array to store data at a third operational voltage that corresponds to the second operational accuracy of the process, wherein the process implements an artificial neural network and the event is a transition from processing associated with a first layer of the artificial neural network to processing associated with a second layer of the artificial neural network.

19. The system of claim 18 , wherein modifying the performance of the process includes:

providing, to the first subset of memory cells, the second operational voltage that is higher than the first operational voltage.

20. The system of claim 18 , wherein modifying the performance of the process includes:

initiating performance of the process using the second subset of memory cells of the memory array to store data at the third operational voltage that corresponds to the second operational accuracy of the process.

21. The system of claim 18 , comprising a plurality of power supply circuits, wherein the memory management circuitry comprises voltage selection circuitry coupled to the plurality of power supply circuits.

22. The system of claim 18 , wherein the memory management circuitry, in operation, controls operational voltages provided to the first and second subsets of memory cells of the memory array to simultaneously provide different operational voltages to the first and second subsets of memory cells of the memory array.

Continuity (4)
Continuation 17742987 · May 12, 2022
Continuation 17012501 · Sep 4, 2020
Provisional Application 62897937 · Sep 9, 2019
Related Publication 20240045589A1 · Feb 8, 2024
References Cited (42)
US 5559956A · Sukegawa · 1996 [cited by applicant]
US 6205063B1 · Aipperspach et al. · 2001 [cited by applicant]
US 7647536B2 · Dempsey et al. · 2010 [cited by applicant]
US 8605527B2 · Shvydun et al. · 2013 [cited by applicant]
US 9432298B1 · Smith · 2016 [cited by applicant]
US 9558107B2 · Camp et al. · 2017 [cited by applicant]
US 9786386B2 · Strauss · 2017 [cited by examiner]
US 9912352B1 · Qin et al. · 2018 [cited by applicant]
US 10622088B2 · Choi · 2020 [cited by applicant]
US 10650902B2 · Kannan et al. · 2020 [cited by applicant]
US 20070033449A1 · Hwang et al. · 2007 [cited by applicant]
US 20100306598A1 · Ackaret et al. · 2010 [cited by applicant]
US 20120069626A1 · Nakano et al. · 2012 [cited by applicant]
US 20120271982A1 · Callaghan · 2012 [cited by applicant]
US 20140095799A1 · Fang et al. · 2014 [cited by applicant]
US 20150128007A1 · Kim · 2015 [cited by applicant]
US 20160092115A1 · Babu · 2016 [cited by examiner]
US 20160225436A1 · Wang et al. · 2016 [cited by applicant]
US 20170110194A1 · Tiwari et al. · 2017 [cited by applicant]
US 20170110206A1 · Ryu et al. · 2017 [cited by applicant]
US 20170185478A1 · Sakai et al. · 2017 [cited by applicant]
US 20180285732A1 · Kurian et al. · 2018 [cited by applicant]
US 20180308561A1 · Koker et al. · 2018 [cited by applicant]
US 20180358110A1 · Buyuktosunoglu et al. · 2018 [cited by applicant]
US 20190042966A1 · Hogaboam et al. · 2019 [cited by applicant]
US 20190056987A1 · Busch et al. · 2019 [cited by applicant]
US 20190068220A1 · Kumar et al. · 2019 [cited by applicant]
US 20190073259A1 · Qin et al. · 2019 [cited by applicant]
US 20190172542A1 · Miladinovic · 2019 [cited by applicant]
US 20190294368A1 · Hiraishi · 2019 [cited by applicant]
US 20190318799A1 · Zorn · 2019 [cited by examiner]
US 20200020390A1 · Fujiwara et al. · 2020 [cited by applicant]
US 20200043531A1 · Earle · 2020 [cited by examiner]
US 20220269410A1 · Chawla et al. · 2022 [cited by applicant]
CN 102301348A · 2011 [cited by applicant]
CN 103247347A · 2013 [cited by applicant]
CN 104572517A · 2015 [cited by applicant]
CN 207440765U · 2018 [cited by applicant]
Azizimazreah et al., “Tolerating Soft Errors in Deep Learning Accelerators with Reliable On-Chip Memory Designs,” [cited by applicant]
Maliuk et al., “Analog Neural Network Design for RF Built-In Self-Test” [cited by applicant]
Reagen et al., “Minerva: Enabling Low-Power, Highly-Accurate Deep Neural Network Accelerators,” [cited by applicant]
Chuanbin, et al., “A PowerPC high-performance processing unit design with SDRAM fault-tolerant and error detectable SDRAM,” China Academic Journal Electronic Publishing House, Aerospace Control, vol. 23, No. 5, Oct. 200… [cited by applicant]