Recent research in micro-architectural attacks has uncovered a variety of vulnerabilities on shared compute devices like CPUs and GPUs which pose a substantial thread to cloud service providers and customers alike. Cloud service providers have therefore moved towards flexible systems that prioritize re-arrangeable hardware components that are not shared between users to minimize attack surfaces while retaining scalability. In this work, we show that for the sake of system security it is necessary to consider not only the security of the processors and peripherals of a system but also the security of the subsystems that connect them. In particular, we investigate the side-channel leakage potential of the I/O translation look-aside buffer (IOTLB) used in I/O memory management units (IOMMUs) to cache address translations. To exploit the IOTLB, we design a hardware module deployed to an FPGA to help us perform precise timing measurements. For the first time, we prove that the IOTLB is the source of a timing-based side-channel leakage and use it to create two covert channels from CPU to peripheral and between peripherals. While the first channel easily achieves an error rate of only 30%, the latter proved to be very reliable as nearly no errors occur. We present a close look at web fingerprints collected through this side-channel, and we examine the I/O operation of a GPU-accelerated SQL database. We then discuss several methods to remedy the observed side-channel leakages, including application design techniques, peripheral layout within existing systems, and micro-architectural features that could harden future IOMMUs.
相關內容
The blockchain-based smart contract lacks privacy since the contract state and instruction code are exposed to the public. Combining smart-contract execution with Trusted Execution Environments (TEEs) provides an efficient solution, called TEE-assisted smart contracts, for protecting the confidentiality of contract states. However, the combination approaches are varied, and a systematic study is absent. Newly released systems may fail to draw upon the experience learned from existing protocols, such as repeating known design mistakes or applying TEE technology in insecure ways. In this paper, we first investigate and categorize the existing systems into two types: the layer-one solution and layer-two solution. Then, we establish an analysis framework to capture their common lights, covering the desired properties (for contract services), threat models, and security considerations (for underlying systems). Based on our taxonomy, we identify their ideal functionalities and uncover the fundamental flaws and reasons for the challenges in each specification design. We believe that this work would provide a guide for the development of TEE-assisted smart contracts, as well as a framework to evaluate future TEE-assisted confidential contract systems.
Industrial Control Systems (ICSs) rely on insecure protocols and devices to monitor and operate critical infrastructure. Prior work has demonstrated that powerful attackers with detailed system knowledge can manipulate exchanged sensor data to deteriorate performance of the process, even leading to full shutdowns of plants. Identifying those attacks requires iterating over all possible sensor values, and running detailed system simulation or analysis to identify optimal attacks. That setup allows adversaries to identify attacks that are most impactful when applied on the system for the first time, before the system operators become aware of the manipulations. In this work, we investigate if constrained attackers without detailed system knowledge and simulators can identify comparable attacks. In particular, the attacker only requires abstract knowledge on general information flow in the plant, instead of precise algorithms, operating parameters, process models, or simulators. We propose an approach that allows single-shot attacks, i.e., near-optimal attacks that are reliably shutting down a system on the first try. The approach is applied and validated on two use cases, and demonstrated to achieve comparable results to prior work, which relied on detailed system information and simulations.
In the realm of unsupervised learning, Bayesian nonparametric mixture models, exemplified by the Dirichlet Process Mixture Model (DPMM), provide a principled approach for adapting the complexity of the model to the data. Such models are particularly useful in clustering tasks where the number of clusters is unknown. Despite their potential and mathematical elegance, however, DPMMs have yet to become a mainstream tool widely adopted by practitioners. This is arguably due to a misconception that these models scale poorly as well as the lack of high-performance (and user-friendly) software tools that can handle large datasets efficiently. In this paper we bridge this practical gap by proposing a new, easy-to-use, statistical software package for scalable DPMM inference. More concretely, we provide efficient and easily-modifiable implementations for high-performance distributed sampling-based inference in DPMMs where the user is free to choose between either a multiple-machine, multiple-core, CPU implementation (written in Julia) and a multiple-stream GPU implementation (written in CUDA/C++). Both the CPU and GPU implementations come with a common (and optional) python wrapper, providing the user with a single point of entry with the same interface. On the algorithmic side, our implementations leverage a leading DPMM sampler from (Chang and Fisher III, 2013). While Chang and Fisher III's implementation (written in MATLAB/C++) used only CPU and was designed for a single multi-core machine, the packages we proposed here distribute the computations efficiently across either multiple multi-core machines or across mutiple GPU streams. This leads to speedups, alleviates memory and storage limitations, and lets us fit DPMMs to significantly larger datasets and of higher dimensionality than was possible previously by either (Chang and Fisher III, 2013) or other DPMM methods.
When IP-packet processing is unconditionally carried out on behalf of an operating system kernel thread, processing systems can experience overload in high incoming traffic scenarios. This is especially worrying for embedded real-time devices controlling their physical environment in industrial IoT scenarios and automotive systems. We propose an embedded real-time aware IP stack adaption with an early demultiplexing scheme for incoming packets and subsequent per-flow aperiodic scheduling. By instrumenting existing embedded IP stacks, rigid prioritization with minimal latency is deployed without the need of further task resources. Simple mitigation techniques can be applied to individual flows, causing hardly measurable overhead while at the same time protecting the system from overload conditions. Our IP stack adaption is able to reduce the low-priority packet processing time by over 86% compared to an unmodified stack. The network subsystem can thereby remain active at a 7x higher general traffic load before disabling the receive IRQ as a last resort to assure deadlines.
The COVID-19 pandemic is accompanied by a massive "infodemic" that makes it hard to identify concise and credible information for COVID-19-related questions, like incubation time, infection rates, or the effectiveness of vaccines. As a novel solution, our paper is concerned with designing a question-answering system based on modern technologies from natural language processing to overcome information overload and misinformation in pandemic situations. To carry out our research, we followed a design science research approach and applied Ingwersen's cognitive model of information retrieval interaction to inform our design process from a socio-technical lens. On this basis, we derived prescriptive design knowledge in terms of design requirements and design principles, which we translated into the construction of a prototypical instantiation. Our implementation is based on the comprehensive CORD-19 dataset, and we demonstrate our artifact's usefulness by evaluating its answer quality based on a sample of COVID-19 questions labeled by biomedical experts.
The real-world use cases of Machine Learning (ML) have exploded over the past few years. However, the current computing infrastructure is insufficient to support all real-world applications and scenarios. Apart from high efficiency requirements, modern ML systems are expected to be highly reliable against hardware failures as well as secure against adversarial and IP stealing attacks. Privacy concerns are also becoming a first-order issue. This article summarizes the main challenges in agile development of efficient, reliable and secure ML systems, and then presents an outline of an agile design methodology to generate efficient, reliable and secure ML systems based on user-defined constraints and objectives.
The success of large-scale models in recent years has increased the importance of statistical models with numerous parameters. Several studies have analyzed over-parameterized linear models with high-dimensional data that may not be sparse; however, existing results depend on the independent setting of samples. In this study, we analyze a linear regression model with dependent time series data under over-parameterization settings. We consider an estimator via interpolation and developed a theory for excess risk of the estimator under multiple dependence types. This theory can treat infinite-dimensional data without sparsity and handle long-memory processes in a unified manner. Moreover, we bound the risk in our theory via the integrated covariance and nondegeneracy of autocorrelation matrices. The results show that the convergence rate of risks with short-memory processes is identical to that of cases with independent data, while long-memory processes slow the convergence rate. We also present several examples of specific dependent processes that can be applied to our setting.
The concept of federated learning (FL) was first proposed by Google in 2016. Thereafter, FL has been widely studied for the feasibility of application in various fields due to its potential to make full use of data without compromising the privacy. However, limited by the capacity of wireless data transmission, the employment of federated learning on mobile devices has been making slow progress in practical. The development and commercialization of the 5th generation (5G) mobile networks has shed some light on this. In this paper, we analyze the challenges of existing federated learning schemes for mobile devices and propose a novel cross-device federated learning framework, which utilizes the anonymous communication technology and ring signature to protect the privacy of participants while reducing the computation overhead of mobile devices participating in FL. In addition, our scheme implements a contribution-based incentive mechanism to encourage mobile users to participate in FL. We also give a case study of autonomous driving. Finally, we present the performance evaluation of the proposed scheme and discuss some open issues in federated learning.
Designers reportedly struggle with design optimization tasks where they are asked to find a combination of design parameters that maximizes a given set of objectives. In HCI, design optimization problems are often exceedingly complex, involving multiple objectives and expensive empirical evaluations. Model-based computational design algorithms assist designers by generating design examples during design, however they assume a model of the interaction domain. Black box methods for assistance, on the other hand, can work with any design problem. However, virtually all empirical studies of this human-in-the-loop approach have been carried out by either researchers or end-users. The question stands out if such methods can help designers in realistic tasks. In this paper, we study Bayesian optimization as an algorithmic method to guide the design optimization process. It operates by proposing to a designer which design candidate to try next, given previous observations. We report observations from a comparative study with 40 novice designers who were tasked to optimize a complex 3D touch interaction technique. The optimizer helped designers explore larger proportions of the design space and arrive at a better solution, however they reported lower agency and expressiveness. Designers guided by an optimizer reported lower mental effort but also felt less creative and less in charge of the progress. We conclude that human-in-the-loop optimization can support novice designers in cases where agency is not critical.
Upcoming HEP experiments, e.g. at the HL-LHC, are expected to increase the volume of generated data by at least one order of magnitude. In order to retain the ability to analyze the influx of data, full exploitation of modern storage hardware and systems, such as low-latency high-bandwidth NVMe devices and distributed object stores, becomes critical. To this end, the ROOT RNTuple I/O subsystem has been designed to address performance bottlenecks and shortcomings of ROOT's current state of the art TTree I/O subsystem. RNTuple provides a backwards-incompatible redesign of the TTree binary format and access API that evolves the ROOT event data I/O for the challenges of the upcoming decades. It focuses on a compact data format, on performance engineering for modern storage hardware, for instance through making parallel and asynchronous I/O calls by default, and on robust interfaces that are easy to use correctly. In this contribution, we evaluate the RNTuple performance for typical HEP analysis tasks. We compare the throughput delivered by RNTuple to popular I/O libraries outside HEP, such as HDF5 and Apache Parquet. We demonstrate the advantages of RNTuple for HEP analysis workflows and provide an outlook on the road to its use in production.
小貼士
登錄享
注冊地址： 北京市海淀區羊坊店路18號2幢3層301-191