RECORD
A Method for Hiding the Increased Non-Volatile Cache Read Latency
Hardware Architecture · Submitted 20 Dec 2021
abstract compiled · 2026-09-23 · no language-model review
The authors' main claim We find that, on average, Cloak outperforms an SRAM LLC by 23.8% and an NVM-only LLC by 8.9% -- in both cases, with negligible additional area.
The compiled record
The paper's own sentences, arranged. Each label is drawn from a closed vocabulary; the number beside it is the probability the classifier gave that label. Numbers in the tables are the authors', checked by code. No sentence in this record was written by a model.
Open the compiled record
The abstract, sentence by sentence
The increased memory demands of workloads is putting high pressure on Last Level Caches (LLCs). Unfortunately, there is limited opportunity to increase the capacity of LLCs due to the area and power requirements of the underlying SRAM technology. Interestingly, emerging Non-Volatile Memory (NVM) technologies promise a feasible alternative to SRAM for LLCs due to their higher area density. However, NVMs have substantially higher read and write latencies, which offset their area density benefit. Although researchers have proposed methods to tolerate NVM's increased write latency, little emphasis has been placed on reducing the critical NVM read latency. To address this problem, this paper proposes Cloak. Cloak exploits data reuse in the LLC at the page level, to hide NVM read latency. Specifically, on certain L1 TLB misses to a page, Cloak transfers LLC-resident data belonging to the page from the LLC NVM array to a set of small SRAM Page Buffers that will service subsequent requests to this page. Further, to enable the high-bandwidth, low-latency transfer of lines of a page to the page buffers, Cloak uses an LLC layout that accelerates the discovery of LLC-resident cache lines from the page. We evaluate Cloak with full-system simulations of a 4-core processor across 14 workloads. We find that, on average, Cloak outperforms an SRAM LLC by 23.8% and an NVM-only LLC by 8.9% -- in both cases, with negligible additional area. Further, Cloak's ED^2 is 39.9% and 17.5% lower, respectively, than these designs.
background contribution method result interpretation main claim
| # | sentence | role | the authors' own work | states the main conclusion | strength |
|---|---|---|---|---|---|
| s0 | The increased memory demands of workloads is putting high pressure on Last Level Caches (LLCs). | background 1.00 | 0.05 | 0.01 | asserted 1.00 |
| s1 | Unfortunately, there is limited opportunity to increase the capacity of LLCs due to the area and power requirements of the underlying SRAM technology. | background 0.98 | 0.06 | 0.01 | asserted 0.99 |
| s2 | Interestingly, emerging Non-Volatile Memory (NVM) technologies promise a feasible alternative to SRAM for LLCs due to their higher area density. | background 1.00 | 0.05 | 0.02 | hedged possibility 0.71 |
| s3 | However, NVMs have substantially higher read and write latencies, which offset their area density benefit. | background 0.97 | 0.05 | 0.02 | asserted 1.00 |
| s4 | Although researchers have proposed methods to tolerate NVM's increased write latency, little emphasis has been placed on reducing the critical NVM read latency. | gap 1.00 | 0.05 | 0.02 | asserted 0.99 |
| s5 | To address this problem, this paper proposes Cloak. | contribution 1.00 | 0.76 | 0.06 | asserted 0.99 |
| s6 | Cloak exploits data reuse in the LLC at the page level, to hide NVM read latency. | method 0.72 | 0.80 | 0.13 | asserted 1.00 |
| s7 | Specifically, on certain L1 TLB misses to a page, Cloak transfers LLC-resident data belonging to the page from the LLC NVM array to a set of small SRAM Page Buffers that will service subsequent requests to this page. | method 0.98 | 0.71 | 0.04 | asserted 1.00 |
| s8 | Further, to enable the high-bandwidth, low-latency transfer of lines of a page to the page buffers, Cloak uses an LLC layout that accelerates the discovery of LLC-resident cache lines from the page. | method 0.97 | 0.82 | 0.03 | asserted 1.00 |
| s9 | We evaluate Cloak with full-system simulations of a 4-core processor across 14 workloads. | method 0.97 | 0.96 | 0.02 | asserted 0.95 |
| s10 | We find that, on average, Cloak outperforms an SRAM LLC by 23.8% and an NVM-only LLC by 8.9% -- in both cases, with negligible additional area. | result 1.00 | 0.97 | 0.74 | asserted 1.00 |
| s11 | Further, Cloak's ED^2 is 39.9% and 17.5% lower, respectively, than these designs. | result 1.00 | 0.96 | 0.38 | asserted 1.00 |
What was examined, and what was not
- arXiv listing abstract v1
- full text
- figures and tables
- references and citing sentences
- relations to other papers
- P abstract
- S sections
- T body sentences
- C citations
- R cross-paper relations
The paper
Original abstract
The increased memory demands of workloads is putting high pressure on Last Level Caches (LLCs). Unfortunately, there is limited opportunity to increase the capacity of LLCs due to the area and power requirements of the underlying SRAM technology. Interestingly, emerging Non-Volatile Memory (NVM) technologies promise a feasible alternative to SRAM for LLCs due to their higher area density. However, NVMs have substantially higher read and write latencies, which offset their area density benefit. Although researchers have proposed methods to tolerate NVM's increased write latency, little emphasis has been placed on reducing the critical NVM read latency. To address this problem, this paper proposes Cloak. Cloak exploits data reuse in the LLC at the page level, to hide NVM read latency. Specifically, on certain L1 TLB misses to a page, Cloak transfers LLC-resident data belonging to the page from the LLC NVM array to a set of small SRAM Page Buffers that will service subsequent requests to this page. Further, to enable the high-bandwidth, low-latency transfer of lines of a page to the page buffers, Cloak uses an LLC layout that accelerates the discovery of LLC-resident cache lines from the page. We evaluate Cloak with full-system simulations of a 4-core processor across 14 workloads. We find that, on average, Cloak outperforms an SRAM LLC by 23.8% and an NVM-only LLC by 8.9% -- in both cases, with negligible additional area. Further, Cloak's ED^2 is 39.9% and 17.5% lower, respectively, than these designs.
Human review
Discussion
Know something about this paper? Leave a comment, or highlight any passage above to comment on it.