IP Library › Granted Patent US 12,436,679
Granted Patent B2
US 12,436,679 · App. 18/439,092 · Granted Oct 7, 2025

Method and device of accessing memory with near memory accelerator

Inventors: Arnab Roy (Bengaluru, IN); Saptarsi Das (Bengaluru, IN); Kiran Kolar Chandrasekharan (Bengaluru, IN); Yeongon Cho (Suwon-si, KR)
Assignee: Samsung Electronics Co., Ltd.
G06F3/061G06F3/0629G06F3/0673
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 12,436,679
App. No.
18/439,092
Granted
Oct 7, 2025
Kind
B2
Abstract

Disclosed are a method of accessing a memory and an electronic device for performing the method. The electronic device includes a processor, and a memory electrically connected to the processor, wherein the processor may be configured to select a rank including bank groups of the memory, select a bank corresponding to a memory address to be accessed from among banks included in the selected rank, select a row and one or more columns from rows and columns of the selected bank corresponding to the memory address, and generate the memory address to access the memory based on an address mapping scheme according to the selected rank, the selected bank, the selected row, and the selected one or more columns.

Claims (31)

1. An electronic device comprising:

a processor; and

a memory electrically connected to the processor,

wherein the processor is configured to:

select, between a first rank and a second rank, a rank comprising bank groups of the memory,

select a bank corresponding to a memory address to be accessed from among banks included in the selected rank,

select a row and one or more columns from among rows and columns of the selected bank, and

generate the memory address to access the memory based on an address mapping scheme according to the selected rank, the selected bank, the selected row, and the selected one or more columns, wherein when the first rank is selected, reshuffling of an order of the generated address is prevented, and wherein when the second rank is selected, reshuffling of the order of the generated address is not prevented.

2. The electronic device of claim 1 , wherein

the processor is further configured to select a channel to access a near memory accelerator (NMA) included in the memory, the channel selected from among channels connected to the memory.

3. The electronic device of claim 1 , wherein

the processor is further configured to generate a column identity corresponding to the selected one or more columns, and wherein

a width of the column identity is greater than a width of a set byte offset, and the column identity is set such that it is not to be repeated within a set wrap around window.

4. The electronic device of claim 1 , wherein

the selected row corresponds to a row hit for all write addresses.

5. The electronic device of claim 1 , wherein

the processor is further configured to select banks from different bank groups among the bank groups included in the selected rank.

6. A method of accessing a memory, the method comprising:

selecting, between a first rank and a second rank, a rank comprising bank groups of a memory;

selecting a bank corresponding to a memory address to be accessed, the bank selected from among banks included in the selected rank;

selecting a row and one or more columns from rows and columns of the selected bank; and

generating the memory address to access the memory based on an address mapping scheme according to the selected rank, the selected bank, the selected row, and the selected one or more columns, wherein when the first rank is selected, reshuffling of an order of generated addresses, including the generated address, is prevented, and wherein when the second rank is selected, reshuffling of the order of the generated addresses is not prevented.

7. The method of claim 6 , further comprising:

selecting a channel to access a near memory accelerator (NMA) included in the memory, from among channels connected to the memory.

8. The method of claim 6 , further comprising:

generating a column identity corresponding to the selected one or more columns, wherein

a width of the column identity is greater than a width of a set byte offset, and the column identity is set not to be repeated within a set wrap around window.

9. The method of claim 6 , wherein

the selected row corresponds to a row hit for all write addresses.

10. The method of claim 6 , wherein

the selecting of the bank comprises selecting banks from different bank groups among the bank groups included in the selected rank.

Assignments (1)
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Apr 30, 2024
From: ROY, ARNAB; DAS, SAPTARSI; CHANDRASEKHARAN, KIRAN KOLAR; CHO, YEONGON
To: SAMSUNG ELECTRONICS CO., LTD.
Reel/Frame 067270/0435 →
Priority Claims (2)
IN 202341009525 · Feb 13, 2023 · national
KR 10-2023-0071830 · Jun 2, 2023 · national
Continuity (1)
Related Publication 20240311009A1 · Sep 19, 2024
References Cited (14)
US 6070227A · Rokicki · 2000 [cited by applicant]
US 7844785B2 · Skerlj et al. · 2010 [cited by applicant]
US 8108596B2 · Aldworth · 2012 [cited by examiner]
US 10203878B2 · Haller et al. · 2019 [cited by applicant]
US 11249651B2 · Salamat et al. · 2022 [cited by applicant]
US 20150089183A1 · Bains · 2015 [cited by examiner]
US 20180284994A1 · Haller et al. · 2018 [cited by applicant]
US 20190212918A1 · Wang et al. · 2019 [cited by applicant]
US 20210124500A1 · Salamat et al. · 2021 [cited by applicant]
US 20210271680A1 · Lee et al. · 2021 [cited by applicant]
KR 1020160040289A · 2016 [cited by applicant]
Rixner, Scott, et al. “Memory Access Scheduling.” ISCA-27, Computer Systems Laboratory, Stanford University, 2000, (11 pages). [cited by applicant]
Ipek, Engin, et al. “Self-Optimizing Memory Controllers: A Reinforcement Learning Approach.” ACM Sigarch Computer Architecture News 36.3, 2008, (12 pages). [cited by applicant]
Zhang, Jialiang, et al., “Software-Defined Address Mapping: A Case on 3d Memory.” Proceedings of the 27th ACM International Conference on Architectural Support for Programming Languages and Operating Systems. 2022., (14… [cited by applicant]