CoChat

25.4 A 20nm 6GB Function-In-Memory DRAM, Based on HBM2 with a 1.2TFLOPS Programmable Computing Unit Using Bank-Level Parallelism, for Machine Learning Applications

Kwon, Young-Cheon, Lee, Suk Han, Lee, Jaehoon, Kwon, Sang-Hyuk, Ryu, Je Min, Son, Jong-Pil, Seongil, O, Yu, Hak-Soo, Lee, Haesuk, Kim, Soo Young, Cho, Youngmin, Kim, Jin Guk, Choi, Jongyoon, Shin, Hyun-Sung, Kim, Jin, Phuah, BengSeng, Kim, HyoungMin, Song, Myeong Jun, Choi, Ahn, Kim, Daeho, Kim, SooYoung, Kim, Eun-Bong, Wang, David, Kang, Shinhaeng, Ro, Yuhwan, Seo, Seungwoo, Song, JoonHo, Youn, Jaeyoun, Sohn, Kyomin, Kim, Nam Sung · 2021 · 2021 IEEE International Solid-State Circuits Conference (ISSCC)
Cited by 204

The DOI matches its Crossref record; 1 of the 1 references it cites were found in a registry or an index.

Abstract

In recent years, artificial intelligence (AI) technology has proliferated rapidly and widely into application areas such as speech recognition, health care, and autonomous driving. To increase the capabilities of AI more powerful systems are needed to process a larger amount of data. This requirement has made domain-specific accelerators, such as GPUs and TPUs, popular; as they can provide orders of magnitude higher performance than state-of-the-art CPUs. However, these accelerators can only operate at their peak performance when they get the necessary data from memory as quickly as it is processed: requiring off-chip memory with a high bandwidth and a large capacity [1]. HBM has thus far met the bandwidth and capacity requirement [2-6], but recent AI technologies such as recurrent neural networks require an even higher bandwidth than HBM [7-8]. While a further increase in off-chip bandwidth can be accomplished by various techniques, it is often limited by power constraints at the chip or system level [9]. Hence, it is essential to decrease demand for off-chip bandwidth with unconventional architectures: such as processing-in-memory. In this paper, we present function-Inmemory DRAM (FIMDRAM) that integrates a 16-wide single-instruction multiple-data engine within the memory banks and that exploits bank-level parallelism to provide 4× higher processing bandwidth than an off-chip memory solution. Second, we show techniques that do not require any modification to conventional memory controllers and their command protocols, which make FIMDRAM more practical for quick industry adoption. Finally, we conclude this paper with circuitand system-level evaluations of our fabricated FIMDRAM.

Overview

The claims and statistics of this paper haven’t been checked yet.

Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.

A few free checks a day · papers already verified open instantly · results are public

Integrity signals

Registry

Crossref · 10.1109/isscc42613.2021.9365862

Cites

1 of 1 references verified0 citation errors · 0 not checkable

Statistics

Verification

Claims the paper makes

The claims and statistics of this paper haven’t been checked yet.

Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.

A few free checks a day · papers already verified open instantly · results are public

Self-check

Percentages, sums and reported differences re-computed from the paper’s own numbers.

The self-check needs the paper’s full text.

Citations

Cites

  • In-Datacenter Performance Analysis of a Tensor Processing Unit

    Jouppi, N. P., Young, C., Patil, N., Patterson, D., Agrawal, G., Bajwa, R., Bates, S., Bhatia, S., Boden, N., Borchers, A., et al. (2017) · Proceedings of the 44th Annual International Symposium on Computer Architecture · doi:10.1145/3079856.3080246 · Cited by 4,537 · cited 1×

    “However, these accelerators can only operate at their peak performance when they get the necessary data from memory as quickly as it is processed: requiring off-chip memory with a high bandwidth and a large capacity [1].”
    Background

Show all 1 references

Metadata

Authors
Kwon, Y.-C., Lee, S. H., Lee, J., Kwon, S.-H., Ryu, J. M., Son, J.-P., Seongil, O., Yu, H.-S., Lee, H., Kim, S. Y., Cho, Y., Kim, J. G., Choi, J., Shin, H.-S., Kim, J., Phuah, B., Kim, H., Song, M. J., Choi, A., Kim, D., Kim, S., Kim, E.-B., Wang, D., Kang, S., Ro, Y., Seo, S., Song, J., Youn, J., Sohn, K., & Kim, N. S.
Venue
2021 IEEE International Solid-State Circuits Conference (ISSCC)
Publisher
IEEE
Published
2021-02-13
Pages
350-352
ISSN
0193-6530 (ISSN-L)
OpenAlex
W3134274954
Cited by
204
References
1
FWCI
24.60
Type
conference paper
Language
en
Open access
closed

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review

How to cite

Cite CoChat for the checks on this page, and the paper for its own findings.

This page
CoChat. (2026). Citation check: 25.4 A 20nm 6GB Function-In-Memory DRAM, Based on HBM2 with a 1.2TFLOPS Programmable Computing Unit Using Bank-Level Parallelism, for Machine Learning Applications. https://app.cochat.ai/papers/10.1109/isscc42613.2021.9365862
The paper
Kwon, Y.-C., Lee, S. H., Lee, J., Kwon, S.-H., Ryu, J. M., Son, J.-P., Seongil, O., Yu, H.-S., Lee, H., Kim, S. Y., Cho, Y., Kim, J. G., Choi, J., Shin, H.-S., Kim, J., Phuah, B., Kim, H., Song, M. J., Choi, A., . . . Kim, N. S. (2021). 25.4 A 20nm 6GB Function-In-Memory DRAM, Based on HBM2 with a 1.2TFLOPS Programmable Computing Unit Using Bank-Level Parallelism, for Machine Learning Applications. 2021 IEEE International Solid-State Circuits Conference (ISSCC). https://doi.org/10.1109/isscc42613.2021.9365862

Data sources

Bibliographic data: Semantic Scholar, OpenAlex, Crossref.

Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).

Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.