A 4-Kb 1-to-8-bit Configurable 6T SRAM-Based Computation-in-Memory Unit-Macro for CNN-Based AI Edge Processors
| Autor: | Yen-Cheng Chiu, Ren-Shuo Liu, Yung-Ning Tu, Jing-Hong Wang, Jia-Jing Chen, Meng-Fan Chang, Kea-Tiong Tang, Shyh-Shyuan Sheu, Chih-Cheng Hsieh, Ruhui Liu, Sih-Han Li, Wei-Hsing Huang, Jian-Wei Su, Chih-I Wu, Wei-Chen Wei, Je-Min Hung, Zhixiao Zhang, Xin Si |
|---|---|
| Rok vydání: | 2020 |
| Předmět: |
Random access memory
business.industry Computer science Computation 020208 electrical & electronic engineering 8-bit Binary number 02 engineering and technology 0202 electrical engineering electronic engineering information engineering Static random-access memory Electrical and Electronic Engineering Macro business Computer hardware |
| Zdroj: | IEEE Journal of Solid-State Circuits. 55:2790-2801 |
| ISSN: | 1558-173X 0018-9200 |
| DOI: | 10.1109/jssc.2020.3005754 |
| Popis: | Previous SRAM-based computing-in-memory (SRAM-CIM) macros suffer small read margins for high-precision operations, large cell array area overhead, and limited compatibility with many input and weight configurations. This work presents a 1-to-8-bit configurable SRAM CIM unit-macro using: 1) a hybrid structure combining 6T-SRAM based in-memory binary product-sum (PS) operations with digital near-memory-computing multibit PS accumulation to increase read accuracy and reduce area overhead; 2) column-based place-value-grouped weight mapping and a serial-bit input (SBIN) mapping scheme to facilitate reconfiguration and increase array efficiency under various input and weight configurations; 3) a self-reference multilevel reader (SRMLR) to reduce read-out energy and achieve a sensing margin 2 $\times $ that of the mid-point reference scheme; and 4) an input-aware bitline voltage compensation scheme to ensure successful read operations across various input-weight patterns. A 4-Kb configurable 6T-SRAM CIM unit-macro was fabricated using a 55-nm CMOS process with foundry 6T-SRAM cells. The resulting macro achieved access times of 3.5 ns per cycle (pipeline) and energy efficiency of 0.6–40.2 TOPS/W under binary to 8-b input/8-b weight precision. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|