Leveraging Adversarial Detection to Enable Scalable and Low Overhead RowHammer Mitigations
- URL: http://arxiv.org/abs/2404.13477v1
- Date: Sat, 20 Apr 2024 22:09:38 GMT
- Title: Leveraging Adversarial Detection to Enable Scalable and Low Overhead RowHammer Mitigations
- Authors: Oğuzhan Canpolat, A. Giray Yağlıkçı, Ataberk Olgun, İsmail Emir Yüksel, Yahya Can Tuğrul, Konstantinos Kanellopoulos, Oğuz Ergin, Onur Mutlu,
- Abstract summary: We tackle the performance overheads of RowHammer solutions by tracking the generators of memory accesses that trigger RowHammer solutions.
BreakHammer limits the number of on-the-fly requests a thread can inject into the memory system based on the thread's RowHammer likelihood.
- Score: 5.767293823380473
- License: http://creativecommons.org/licenses/by/4.0/
- Abstract: RowHammer is a prime example of read disturbance in DRAM where repeatedly accessing (hammering) a row of DRAM cells (DRAM row) induces bitflips in other physically nearby DRAM rows. RowHammer solutions perform preventive actions (e.g., refresh neighbor rows of the hammered row) that mitigate such bitflips to preserve memory isolation, a fundamental building block of security and privacy in modern computing systems. However, preventive actions induce non-negligible memory request latency and system performance overheads as they interfere with memory requests in the memory controller. As shrinking technology node size over DRAM chip generations exacerbates RowHammer, the overheads of RowHammer solutions become prohibitively large. As a result, a malicious program can effectively hog the memory system and deny service to benign applications by causing many RowHammer preventive actions. In this work, we tackle the performance overheads of RowHammer solutions by tracking the generators of memory accesses that trigger RowHammer solutions. To this end, we propose BreakHammer. BreakHammer cooperates with existing RowHammer solutions to identify hardware threads that trigger preventive actions. To do so, BreakHammer estimates the RowHammer likelihood of a thread, based on how frequently it triggers RowHammer preventive actions. BreakHammer limits the number of on-the-fly requests a thread can inject into the memory system based on the thread's RowHammer likelihood. By doing so, BreakHammer significantly reduces the number of performed counter-measures, improves the system performance by an average (maximum) of 48.7% (105.5%), and reduces the maximum slowdown induced on a benign application by 14.6% with near-zero area overhead (e.g., 0.0002% of a highend processor's chip area).
Related papers
- ImPress: Securing DRAM Against Data-Disturbance Errors via Implicit Row-Press Mitigation [1.3921736520874155]
DRAM cells are susceptible to Data-Disturbance Errors (DDE)
Rowhammer is a well-known DDE vulnerability that occurs when a row is repeatedly activated.
Row-Press (RP) is a new DDE vulnerability that occurs when a row is kept open for a long time.
arXiv Detail & Related papers (2024-07-22T19:20:14Z) - CoMeT: Count-Min-Sketch-based Row Tracking to Mitigate RowHammer at Low Cost [6.658509339087993]
CoMeT prevents RowHammer bitflips with low area, performance, and energy costs.
CoMeT uses low-cost and scalable hash-based counters to track DRAM row activations.
CoMeT achieves a good trade-off between performance, energy, and area overheads.
arXiv Detail & Related papers (2024-02-29T00:21:29Z) - RelayAttention for Efficient Large Language Model Serving with Long System Prompts [59.50256661158862]
This paper aims to improve the efficiency of LLM services that involve long system prompts.
handling these system prompts requires heavily redundant memory accesses in existing causal attention algorithms.
We propose RelayAttention, an attention algorithm that allows reading hidden states from DRAM exactly once for a batch of input tokens.
arXiv Detail & Related papers (2024-02-22T18:58:28Z) - HiRE: High Recall Approximate Top-$k$ Estimation for Efficient LLM
Inference [68.59839755875252]
HiRE comprises of two novel components: (i) a compression scheme to cheaply predict top-$k$ rows/columns with high recall, followed by full computation restricted to the predicted subset, and (ii) DA-TOP-$k$: an efficient multi-device approximate top-$k$ operator.
We demonstrate that on a one billion parameter model, HiRE applied to both the softmax as well as feedforward layers, achieves almost matching pretraining and downstream accuracy, and speeds up inference latency by $1.47times$ on a single TPUv5e device.
arXiv Detail & Related papers (2024-02-14T18:04:36Z) - ABACuS: All-Bank Activation Counters for Scalable and Low Overhead RowHammer Mitigation [6.570851573752742]
We introduce ABACuS, a new low-cost hardware-counter-based RowHammer mitigation technique.
ABACuS uses a single shared row activation counter to track activations to the rows with the same row address in all DRAM banks.
Our evaluations show that ABACuS securely prevents RowHammer bitflip/energy overhead and low area cost.
arXiv Detail & Related papers (2023-10-15T23:04:30Z) - Randomized Line-to-Row Mapping for Low-Overhead Rowhammer Mitigations [0.716879432974126]
Recently proposed secure Rowhammer mitigations rely on performing mitigative action on the aggressor rather than the victims.
We propose Rubix, which breaks the spatial correlation in the line-to-row mapping by using an encrypted address to access the memory.
We also propose Rubix-D, which dynamically changes the line-to-row mapping.
arXiv Detail & Related papers (2023-08-28T21:22:15Z) - RowPress: Amplifying Read Disturbance in Modern DRAM Chips [7.046976177695823]
RowPress breaks memory isolation by keeping a DRAM row open for a long period of time.
In extreme cases, RowPress induces bitflips in a DRAM row when an adjacent row is activated only once.
Our detailed characterization of 164 real DDR4 DRAM chips shows that RowPress affects chips from all three major DRAM manufacturers.
arXiv Detail & Related papers (2023-06-29T16:09:56Z) - Recurrent Dynamic Embedding for Video Object Segmentation [54.52527157232795]
We propose a Recurrent Dynamic Embedding (RDE) to build a memory bank of constant size.
We propose an unbiased guidance loss during the training stage, which makes SAM more robust in long videos.
We also design a novel self-correction strategy so that the network can repair the embeddings of masks with different qualities in the memory bank.
arXiv Detail & Related papers (2022-05-08T02:24:43Z) - Hashing-based Non-Maximum Suppression for Crowded Object Detection [63.761451382081844]
We propose an algorithm, named hashing-based non-maximum suppression (HNMS) to efficiently suppress the non-maximum boxes for object detection.
For two-stage detectors, we replace NMS in region proposal network with HNMS, and observe significant speed-up with comparable accuracy.
Experiments are conducted on CARPK, SKU-110K, CrowdHuman datasets to demonstrate the efficiency and effectiveness of HNMS.
arXiv Detail & Related papers (2020-05-22T23:45:59Z) - Projection & Probability-Driven Black-Box Attack [205.9923346080908]
Existing black-box attacks suffer from the need for excessive queries in the high-dimensional space.
We propose Projection & Probability-driven Black-box Attack (PPBA) to tackle this problem.
Our method requires at most 24% fewer queries with a higher attack success rate compared with state-of-the-art approaches.
arXiv Detail & Related papers (2020-05-08T03:37:50Z) - Parallelising the Queries in Bucket Brigade Quantum RAM [69.43216268165402]
Quantum algorithms often use quantum RAMs (QRAM) for accessing information stored in a database-like manner.
We show a systematic method to significantly reduce the effective query time by using Clifford+T gate parallelism.
We conclude that, in theory, fault-tolerant bucket brigade quantum RAM queries can be performed approximately with the speed of classical RAM.
arXiv Detail & Related papers (2020-02-21T14:50:03Z)
This list is automatically generated from the titles and abstracts of the papers in this site.
This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.