SEARCH · Searcharxiv

Results for “cs.OS”

Search indexed arXiv papers on artificial intelligence, large language models, computer vision and robotics. Read source abstracts and follow links to arXiv.

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 739 records · Page 41Linked to original sources

arXiv2024

FlexBSO: Flexible Block Storage Offload for Datacenters

Efficient virtualization of CPU and memory is standardized and mature. Capabilities such as Intel VT-x [3] have been added by manufacturers for efficient hypervisor support. In contrast, virtualization of a block device and its presentation to the virtual machines on the host can be done in multiple ways. Indeed, hyperscalers develop in-house solutions to improve performance and cost-efficiency of their storage solutions for datacenters. Unfortunately, these storage solutions are based on specialized hardware and software which are not publicly available. The traditional solution is to expose virtual block device to the VM through a paravirtualized driver like virtio [2]. virtio provides significantly better performance than real block device driver emulation because of host OS and guest OS cooperation. The IO requests are then fulfilled by the host OS either with a local block device such as an SSD drive or with some form of disaggregated storage over the network like NVMe-oF or iSCSI. There are three main problems to the traditional solution. 1) Cost. IO operations consume host CPU cycles due to host OS involvement. These CPU cycles are doing useless work from the application point of view. 2) Inflexibility. Any change of the virtualized storage stack requires host OS and/or guest OS cooperation and cannot be done silently in production. 3) Performance. IO operations are causing recurring VM EXITs to do the transition from non-root mode to root mode on the host CPU. This results into excessive IO performance impact. We propose FlexBSO, a hardware-assisted solution, which solves all the mentioned issues. Our prototype is based on the publicly available Bluefield-2 SmartNIC with NVIDIA SNAP support, hence can be deployed without any obstacles.

cs.NI↗

arXiv2024

Head-First Memory Allocation on Best-Fit with Space-Fitting

Although best-fit is known to be slow, it excels at optimizing memory space utilization. Interestingly, by keeping the free memory region at the top of the memory, the process of memory allocation and deallocation becomes approximately 34.86% faster while also maintaining external fragmentation at minimum.

cs.OS↗

arXiv2024

SafeBPF: Hardware-assisted Defense-in-depth for eBPF Kernel Extensions

The eBPF framework enables execution of user-provided code in the Linux kernel. In the last few years, a large ecosystem of cloud services has leveraged eBPF to enhance container security, system observability, and network management. Meanwhile, incessant discoveries of memory safety vulnerabilities have left the systems community with no choice but to disallow unprivileged eBPF programs, which unfortunately limits eBPF use to only privileged users. To improve run-time safety of the framework, we introduce SafeBPF, a general design that isolates eBPF programs from the rest of the kernel to prevent memory safety vulnerabilities from being exploited. We present a pure software implementation using a Software-based Fault Isolation (SFI) approach and a hardware-assisted implementation that leverages ARM's Memory Tagging Extension (MTE). We show that SafeBPF incurs up to 4% overhead on macrobenchmarks while achieving desired security properties.

cs.CR↗

arXiv2025

Rethinking Programmed I/O for Fast Devices, Cheap Cores, and Coherent Interconnects

Conventional wisdom holds that an efficient interface between an OS running on a CPU and a high-bandwidth I/O device should use Direct Memory Access (DMA) to offload data transfer, descriptor rings for buffering and queuing, and interrupts for asynchrony between cores and device. In this paper we question this wisdom in the light of two trends: modern and emerging cache-coherent interconnects like CXL3.0, and workloads, particularly microservices and serverless computing. Like some others before us, we argue that the assumptions of the DMA-based model are obsolete, and in many use-cases programmed I/O, where the CPU explicitly transfers data and control information to and from a device via loads and stores, delivers a more efficient system. However, we push this idea much further. We show, in a real hardware implementation, the gains in latency for fine-grained communication achievable using an open cache-coherence protocol which exposes cache transitions to a smart device, and that throughput is competitive with DMA over modern interconnects. We also demonstrate three use-cases: fine-grained RPC-style invocation of functions on an accelerator, offloading of operators in a streaming dataflow engine, and a network interface targeting serverless functions, comparing our use of coherence with both traditional DMA-style interaction and a highly-optimized implementation using memory-mapped programmed I/O over PCIe.

cs.AR↗

arXiv2024

Skip TLB flushes for reused pages within mmap's

Memory access efficiency is significantly enhanced by caching recent address translations in the CPUs' Translation Lookaside Buffers (TLBs). However, since the operating system is not aware of which core is using a particular mapping, it flushes TLB entries across all cores where the application runs whenever addresses are unmapped, ensuring security and consistency. These TLB flushes, known as TLB shootdowns, are costly and create a performance and scalability bottleneck. A key contributor to TLB shootdowns is memory-mapped I/O, particularly during mmap-munmap cycles and page cache evictions. Often, the same physical pages are reassigned to the same process post-eviction, presenting an opportunity for the operating system to reduce the frequency of TLB shootdowns. We demonstrate, that by slightly extending the mmap function, TLB shootdowns for these "recycled pages" can be avoided. Therefore we introduce and implement the "fast page recycling" (FPR) feature within the mmap system call. FPR-mmaps maintain security by only triggering TLB shootdowns when a page exits its recycling cycle and is allocated to a different process. To ensure consistency when FPR-mmap pointers are used, we made minor adjustments to virtual memory management to avoid the ABA problem. Unlike previous methods to mitigate shootdown effects, our approach does not require any hardware modifications and operates transparently within the existing Linux virtual memory framework. Our evaluations across a variety of CPU, memory, and storage setups, including persistent memory and Optane SSDs, demonstrate that FPR delivers notable performance gains, with improvements of up to 28% in real-world applications and 92% in micro-benchmarks. Additionally, we show that TLB shootdowns are a significant source of bottlenecks, previously misattributed to other components of the Linux kernel.

cs.OS↗

arXiv2024

eBPF-mm: Userspace-guided memory management in Linux with eBPF

We leverage eBPF in order to implement custom policies in the Linux memory subsystem. Inspired by CBMM, we create a mechanism that provides the kernel with hints regarding the benefit of promoting a page to a specific size. We introduce a new hook point in Linux page fault handling path for eBPF programs, providing them the necessary context to determine the page size to be used. We then develop a framework that allows users to define profiles for their applications and load them into the kernel. A profile consists of memory regions of interest and their expected benefit from being backed by 4KB, 64KB and 2MB pages. In our evaluation, we profiled our workloads to identify hot memory regions using DAMON.

cs.OS↗

arXiv2024

Analysis of Synchronization Mechanisms in Operating Systems

This research analyzed the performance and consistency of four synchronization mechanisms-reentrant locks, semaphores, synchronized methods, and synchronized blocks-across three operating systems: macOS, Windows, and Linux. Synchronization ensures that concurrent processes or threads access shared resources safely, and efficient synchronization is vital for maintaining system performance and reliability. The study aimed to identify the synchronization mechanism that balances efficiency, measured by execution time, and consistency, assessed by variance and standard deviation, across platforms. The initial hypothesis proposed that mutex-based mechanisms, specifically synchronized methods and blocks, would be the most efficient due to their simplicity. However, empirical results showed that reentrant locks had the lowest average execution time (14.67ms), making them the most efficient mechanism, but with the highest variability (standard deviation of 1.15). In contrast, synchronized methods, blocks, and semaphores exhibited higher average execution times (16.33ms for methods and 16.67ms for blocks) but with greater consistency (variance of 0.33). The findings indicated that while reentrant locks were faster, they were more platform-dependent, whereas mutex-based mechanisms provided more predictable performance across all operating systems. The use of virtual machines for Windows and Linux was a limitation, potentially affecting the results. Future research should include native testing and explore additional synchronization mechanisms and higher concurrency levels. These insights help developers and system designers optimize synchronization strategies for either performance or stability, depending on the application's requirements.

cs.OS↗

arXiv2024

Flexible Swapping for the Cloud

Memory has become the primary cost driver in cloud data centers. Yet, a significant portion of memory allocated to VMs in public clouds remains unused. To optimize this resource, "cold" memory can be reclaimed from VMs and stored on slower storage or compressed, enabling memory overcommit. Current overcommit systems rely on general-purpose OS swap mechanisms, which are not optimized for virtualized workloads, leading to missed memory-saving opportunities and ineffective use of optimizations like prefetchers. This paper introduces a userspace memory management framework designed for VMs. It enables custom policies that have full control over the virtual machines' memory using a simple userspace API, supports huge page-based swapping to satisfy VM performance requirements, is easy to deploy by leveraging Linux/KVM, and supports zero-copy I/O virtualization with shared VM memory. Our evaluation demonstrates that an overcommit system based on our framework outperforms the state-of-the-art solutions on both micro-benchmarks and commonly used cloud workloads. Specifically our implementation outperforms the Linux Kernel baseline implementation by up to 25% while saving a similar amount of memory. We also demonstrate the benefits of custom policies by implementing workload-specific reclaimers and prefetchers that save $10\%$ additional memory, improve performance in a limited memory scenario by 30% over the Linux baseline, and recover faster from hard limit releases.

cs.DC↗

arXiv2024

Dissecting CXL Memory Performance at Scale: Analysis, Modeling, and Optimization

We present SupMario, a characterization framework designed to thoroughly analyze, model, and optimize CXL memory performance. SupMario is based on extensive evaluation of 265 workloads spanning 4 real CXL devices within 7 memory latency configurations across 4 processor platforms. SupMario uncovers many key insights, including detailed workload performance at sub-us memory latencies (140-410 ns), CXL tail latencies, CPU tolerance to CXL latencies, CXL performance root-cause analysis and precise performance prediction models. In particular, SupMario performance models rely solely on 12 CPU performance counters and accurately fit over 99% and 91%-94% workloads with a 10% misprediction target for NUMA and CXL memory, respectively. We demonstrate the practical utility of SupMario characterization findings, models, and insights by applying them to popular CXL memory management schemes, such as page interleaving and tiering policies, to identify system inefficiencies during runtime. We introduce a novel ``bestshot'' page interleaving policy and a regulated page tiering policy (Alto) tailored for memory bandwidth- and latency-sensitive workloads. In bandwidth bound scenarios, our ``best-shot'' interleaving, guided by our novel performance prediction model, achieves close-to optimal scenarios by exploiting the aggregate system and CXL/NUMA memory bandwidth. For latency sensitive workloads, Alto, driven by our key insight of utilizing ``amortized'' memory latency to regulate unnecessary page migrations, achieves up to 177% improvement over state-of-the-art memory tiering systems like TPP, as demonstrated through extensive evaluation with 8 real-world applications.

cs.OS↗

arXiv2024

Assessing FIFO and Round Robin Scheduling:Effects on Data Pipeline Performance and Energy Usage

In the case of compute-intensive machine learning, efficient operating system scheduling is crucial for performance and energy efficiency. This paper conducts a comparative study over FIFO(First-In-First-Out) and RR(Round-Robin) scheduling policies with the application of real-time machine learning training processes and data pipelines on Ubuntu-based systems. Knowing a few patterns of CPU usage and energy consumption, we identify which policy (the exclusive or the shared) provides higher performance and/or lower energy consumption for typical modern workloads. Results of this study would help in providing better operating system schedulers for modern systems like Ubuntu, working to improve performance and reducing energy consumption in compute intensive workloads.

cs.OS↗

arXiv2024

FusionANNS: An Efficient CPU/GPU Cooperative Processing Architecture for Billion-scale Approximate Nearest Neighbor Search

Approximate nearest neighbor search (ANNS) has emerged as a crucial component of database and AI infrastructure. Ever-increasing vector datasets pose significant challenges in terms of performance, cost, and accuracy for ANNS services. None of modern ANNS systems can address these issues simultaneously. We present FusionANNS, a high-throughput, low-latency, cost-efficient, and high-accuracy ANNS system for billion-scale datasets using SSDs and only one entry-level GPU. The key idea of FusionANNS lies in CPU/GPU collaborative filtering and re-ranking mechanisms, which significantly reduce I/O operations across CPUs, GPU, and SSDs to break through the I/O performance bottleneck. Specifically, we propose three novel designs: (1) multi-tiered indexing to avoid data swapping between CPUs and GPU, (2) heuristic re-ranking to eliminate unnecessary I/Os and computations while guaranteeing high accuracy, and (3) redundant-aware I/O deduplication to further improve I/O efficiency. We implement FusionANNS and compare it with the state-of-the-art SSD-based ANNS system -- SPANN and GPU-accelerated in-memory ANNS system -- RUMMY. Experimental results show that FusionANNS achieves 1) 9.4-13.1X higher query per second (QPS) and 5.7-8.8X higher cost efficiency compared with SPANN; 2) and 2-4.9X higher QPS and 2.3-6.8X higher cost efficiency compared with RUMMY, while guaranteeing low latency and high accuracy.

cs.IR↗

arXiv2024

Exploring Time-Space trade-offs for synchronized in Lilliput

In the context of Project Lilliput, which attempts to reduce the size of object header in the HotSpot Java Virtual Machine (JVM), we explore a curated set of synchronization algorithms. Each of the algorithms could serve as a potential replacement implementation for the "synchronized" construct in HotSpot. Collectively, the algorithms illuminate trade-offs in space-time properties. The key design decisions are where to locate synchronization metadata (monitor fields), how to map from an object to those fields, and the lifecycle of the monitor information. The reader is assumed to be familiar with current HotSpot implementation of "synchronized" as well as the Compact Java Monitors (CJM) design and Project Lilliput.

cs.OS↗

arXiv2025

Energy-Efficient Computation with DVFS using Deep Reinforcement Learning for Multi-Task Systems in Edge Computing

Finding an optimal energy-efficient policy that is adaptable to underlying edge devices while meeting deadlines for tasks has always been challenging. This research studies generalized systems with multi-task, multi-deadline scenarios with reinforcement learning-based DVFS for energy saving for periodic soft real-time applications on edge devices. This work addresses the limitation of previous work that models a periodic system as a single task and single-deadline scenario, which is too simplified to cope with complex situations. The method encodes time series data in the Linux kernel into information that is easy to interpret for reinforcement learning, allowing the system to generate DVFS policies to adapt system patterns based on the general workload. For encoding, we present two different methods for comparison. Both methods use only one performance counter: system utilization, and the kernel only needs minimal information from the userspace. Our method is implemented on Jetson Nano Board (2GB) and is tested with three fixed multitask workloads, which are three, five, and eight tasks in the workload, respectively. For randomness and generalization, we also designed a random workload generator to build different multitask workloads to test. Based on the test results, our method could save 3%-10% power compared to Linux built-in governors.

cs.OS↗

arXiv2024

The eBPF Runtime in the Linux Kernel

Extended Berkeley Packet Filter (eBPF) is a runtime that enables users to load programs into the operating system (OS) kernel, like Linux or Windows, and execute them safely and efficiently at designated kernel hooks. Each program passes through a verifier that reasons about the safety guarantees for execution. Hosting a safe virtual machine runtime within the kernel makes it dynamically programmable. Unlike the popular approach of bypassing or completely replacing the kernel, eBPF gives users the flexibility to modify the kernel on the fly, rapidly experiment and iterate, and deploy solutions to achieve their workload-specific needs, while working in concert with the kernel. In this paper, we present the first comprehensive description of the design and implementation of the eBPF runtime in the Linux kernel. We argue that eBPF today provides a mature and safe programming environment for the kernel. It has seen wide adoption since its inception and is increasingly being used not just to extend, but program entire components of the kernel, while preserving its runtime integrity. We outline the compelling advantages it offers for real-world production usage, and illustrate current use cases. Finally, we identify its key challenges, and discuss possible future directions.

cs.OS↗

arXiv2024

Global Scheduling of Weakly-Hard Real-Time Tasks using Job-Level Priority Classes

Real-time systems are intrinsic components of many pivotal applications, such as self-driving vehicles, aerospace and defense systems. The trend in these applications is to incorporate multiple tasks onto fewer, more powerful hardware platforms, e.g., multi-core systems, mainly for reducing cost and power consumption. Many real-time tasks, like control tasks, can tolerate occasional deadline misses due to robust algorithms. These tasks can be modeled using the weakly-hard model. Literature shows that leveraging the weakly-hard model can relax the over-provisioning associated with designed real-time systems. However, a wide-range of the research focuses on single-core platforms. Therefore, we strive to extend the state-of-the-art of scheduling weakly-hard real-time tasks to multi-core platforms. We present a global job-level fixed priority scheduling algorithm together with its schedulability analysis. The scheduling algorithm leverages the tolerable continuous deadline misses to assigning priorities to jobs. The proposed analysis extends the Response Time Analysis (RTA) for global scheduling to test the schedulability of tasks. Hence, our analysis scales with the number of tasks and number of cores because, unlike literature, it depends neither on Integer Linear Programming nor reachability trees. Schedulability analyses show that the schedulability ratio is improved by 40% comparing to the global Rate Monotonic (RM) scheduling and up to 60% more than the global EDF scheduling, which are the state-of-the-art schedulers on the RTEMS real-time operating system. Our evaluation on industrial embedded multi-core platform running RTEMS shows that the scheduling overhead of our proposal does not exceed 60 Nanosecond.

cs.OS↗

arXiv2024

Serverless Cold Starts and Where to Find Them

This paper releases and analyzes a month-long trace of 85 billion user requests and 11.9 million cold starts from Huawei's serverless cloud platform. Our analysis spans workloads from five data centers. We focus on cold starts and provide a comprehensive examination of the underlying factors influencing the number and duration of cold starts. These factors include trigger types, request synchronicity, runtime languages, and function resource allocations. We investigate components of cold starts, including pod allocation time, code and dependency deployment time, and scheduling delays, and examine their relationships with runtime languages, trigger types, and resource allocation. We introduce pod utility ratio to measure the pod's useful lifetime relative to its cold start time, giving a more complete picture of cold starts, and see that some pods with long cold start times have longer useful lifetimes. Our findings reveal the complexity and multifaceted origins of the number, duration, and characteristics of cold starts, driven by differences in trigger types, runtime languages, and function resource allocations. For example, cold starts in Region 1 take up to 7 seconds, dominated by dependency deployment time and scheduling. In Region 2, cold starts take up to 3 seconds and are dominated by pod allocation time. Based on this, we identify opportunities to reduce the number and duration of cold starts using strategies for multi-region scheduling. Finally, we suggest directions for future research to address these challenges and enhance the performance of serverless cloud platforms. Our datasets and code are available here https://github.com/sir-lab/data-release

cs.DC↗

arXiv2025

A Practical-Driven Framework for Transitioning Drive-by-Wire to Autonomous Driving Systems: A Case Study with a Chrysler Pacifica Hybrid Vehicle

Transitioning from a Drive-by-Wire (DBW) system to a fully autonomous driving system (ADS) involves multiple stages of development and demands robust positioning and sensing capabilities. This paper presents a practice-driven framework for facilitating the DBW-to-ADS transition using a 2022 Chrysler Pacifica Hybrid Minivan equipped with cameras, LiDAR, GNSS, and onboard computing hardware configured with the Robot Operating System (ROS) and Autoware.AI. The implementation showcases offline autonomous operations utilizing pre-recorded LiDAR and camera data, point clouds, and vector maps, enabling effective localization and path planning within a structured test environment. The study addresses key challenges encountered during the transition, particularly those related to wireless-network-assisted sensing and positioning. It offers practical solutions for overcoming software incompatibility constraints, sensor synchronization issues, and limitations in real-time perception. Furthermore, the integration of sensing, data fusion, and automation is emphasized as a critical factor in supporting autonomous driving systems in map generation, simulation, and training. Overall, the transition process outlined in this work aims to provide actionable strategies for researchers pursuing DBW-to-ADS conversion. It offers direction for incorporating real-time perception, GNSS-LiDAR-camera integration, and fully ADS-equipped autonomous vehicle operations, thus contributing to the advancement of robust autonomous vehicle technologies.

cs.RO↗

arXiv2024

SoK: Software Compartmentalization

Decomposing large systems into smaller components with limited privileges has long been recognized as an effective means to minimize the impact of exploits. Despite historical roots, demonstrated benefits, and a plethora of research efforts in academia and industry, the compartmentalization of software is still not a mainstream practice. This paper investigates why, and how this status quo can be improved. Noting that existing approaches are fraught with inconsistencies in terminology and analytical methods, we propose a unified model for the systematic analysis, comparison, and directing of compartmentalization approaches. We use this model to review 211 research efforts and analyze 61 mainstream compartmentalized systems, confronting them to understand the limitations of both research and production works. Among others, our findings reveal that mainstream efforts largely rely on manual methods, custom abstractions, and legacy mechanisms, poles apart from recent research. We conclude with recommendations: compartmentalization should be solved holistically; progress is needed towards simplifying the definition of compartmentalization policies; towards better challenging our threat models in the light of confused deputies and hardware limitations; as well as towards bridging the gaps we pinpoint between research and mainstream needs. This paper not only maps the historical and current landscape of compartmentalization, but also sets forth a framework to foster their evolution and adoption.

cs.CR↗