Global interconnects in VLSI complexity single flux quantum systems

Jabbari, Tahereh; Friedman, Eby G.

doi:10.1145/3414622.3431911

Cited by 32 publications

(1 citation statement)

References 53 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…SFQ logic components are connected by active Josephson transmission lines (JTLs) and passive transmission lines (PTLs) [43]. As Figure 2(a) shows, compared to a CMOS wire, JTL and PTL enjoy two orders of magnitude shorter latency, since they have no DC resistance [18,19]. A PTL requires a much smaller delay than a JTL, particularly when the length is large.…”

Section: Background 21 Sfq Technologymentioning

confidence: 99%

SMART: A Heterogeneous Scratchpad Memory Architecture for Superconductor SFQ-based Systolic CNN Accelerators

Zokaee

Jiang

2021

MICRO-54: 54th Annual IEEE/ACM International Symposium on Microarchitecture

View full text Add to dashboard Cite

Ultra-fast & low-power superconductor single-flux-quantum (SFQ)based CNN systolic accelerators are built to enhance the CNN inference throughput. However, shift-register (SHIFT)-based scratchpad memory (SPM) arrays prevent a SFQ CNN accelerator from exceeding 40% of its peak throughput, due to the lack of random access capability. This paper first documents our study of a variety of cryogenic memory technologies, including Vortex Transition Memory (VTM), Josephson-CMOS SRAM, MRAM, and Superconducting Nanowire Memory, during which we found that none of the aforementioned technologies made a SFQ CNN accelerator achieve high throughput, small area, and low power simultaneously. Second, we present a heterogeneous SPM architecture, SMART, composed of SHIFT arrays and a random access array to improve the inference throughput of a SFQ CNN systolic accelerator. Third, we propose a fast, low-power and dense pipelined random access CMOS-SFQ array by building SFQ passive-transmission-line-based H-Trees that connect CMOS sub-banks. Finally, we create an ILP-based compiler to deploy CNN models on SMART. Experimental results show that, with the same chip area overhead, compared to the latest SHIFT-based SFQ CNN accelerator, SMART improves the inference throughput by 3.9× (2.2×), and reduces the inference energy by 86% (71%) when inferring a single image (a batch of images).

show abstract

Section: Background 21 Sfq Technologymentioning

confidence: 99%