Operating Systems Review (ACM)最新文献_第4页

WineFS

Operating Systems Review (ACM) Pub Date : 2021-10-26 DOI: 10.1145/3477132.3483567

Rohan Kadekodi, Saurabh Kadekodi, Soujanya Ponnapalli, Harshad Shirwadkar, G. Ganger, Aasheesh Kolli, Vijay Chidambaram

引用次数: 22

When Idling is Ideal: Optimizing Tail-Latency for Heavy-Tailed Datacenter Workloads with Perséphone 当空闲是理想的:使用pers<s:1> phone优化重尾数据中心工作负载的尾延迟

Operating Systems Review (ACM) Pub Date : 2021-10-26 DOI: 10.1145/3477132.3483571

Henri Maxime Demoulin, Joshua Fried, Isaac Pedisich, Marios Kogias, B. T. Loo, L. T. Phan, Irene Zhang

引用次数: 23

Scale and Performance in a Filesystem Semi-Microkernel 文件系统半微内核中的规模和性能

Operating Systems Review (ACM) Pub Date : 2021-10-26 DOI: 10.1145/3477132.3483581

Jing Liu, Anthony Rebello, Yifan Dai, Chenhao Ye, Sudarsun Kannan, A. Arpaci-Dusseau, Remzi H. Arpaci-Dusseau

引用次数: 8

Mycelium: Large-Scale Distributed Graph Queries with Differential Privacy 菌丝体:具有差分隐私的大规模分布式图查询

Operating Systems Review (ACM) Pub Date : 2021-10-26 DOI: 10.1145/3477132.3483585

Edo Roth, Karan Newatia, Ke Zhong, Sebastian Angel, Andreas Haeberlen

引用次数: 14

Witcher: Systematic Crash Consistency Testing for Non-Volatile Memory Key-Value Stores 非易失性内存键值存储的系统崩溃一致性测试

Operating Systems Review (ACM) Pub Date : 2021-10-26 DOI: 10.1145/3477132.3483556

Xinwei Fu, Wook-Hee Kim, Ajay Paddayuru Shreepathi, Mohannad Ismail, Sunny Wadkar, Dongyoon Lee, Changwoo Min

{"title":"Witcher: Systematic Crash Consistency Testing for Non-Volatile Memory Key-Value Stores","authors":"Xinwei Fu, Wook-Hee Kim, Ajay Paddayuru Shreepathi, Mohannad Ismail, Sunny Wadkar, Dongyoon Lee, Changwoo Min","doi":"10.1145/3477132.3483556","DOIUrl":"https://doi.org/10.1145/3477132.3483556","url":null,"abstract":"The advent of non-volatile main memory (NVM) enables the development of crash-consistent software without paying storage stack overhead. However, building a correct crash-consistent program remains very challenging in the presence of a volatile cache. This paper presents Witcher, a systematic crash consistency testing framework, which detects both correctness and performance bugs in NVM-based persistent key-value stores and underlying NVM libraries, without test space explosion and without manual annotations or crash consistency checkers. To detect correctness bugs, Witcher automatically infers likely correctness conditions by analyzing data and control dependencies between NVM accesses. Then Witcher validates if any violation of them is a true crash consistency bug by checking output equivalence between executions with and without a crash. Moreover, Witcher detects performance bugs by analyzing the execution traces. Evaluation with 20 NVM key-value stores based on Intel's PMDK library shows that Witcher discovers 47 (36 new) correctness consistency bugs and 158 (113 new) performance bugs in both applications and PMDK.","PeriodicalId":38935,"journal":{"name":"Operating Systems Review (ACM)","volume":"50 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2021-10-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"83991805","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 15

Forerunner: Constraint-based Speculative Transaction Execution for Ethereum 先驱:以太坊基于约束的投机交易执行

Operating Systems Review (ACM) Pub Date : 2021-10-26 DOI: 10.1145/3477132.3483564

Yang Chen, Zhongxin Guo, Runhuai Li, Shuo Chen, Lidong Zhou, Yajin Zhou, Xian Zhang

{"title":"Forerunner: Constraint-based Speculative Transaction Execution for Ethereum","authors":"Yang Chen, Zhongxin Guo, Runhuai Li, Shuo Chen, Lidong Zhou, Yajin Zhou, Xian Zhang","doi":"10.1145/3477132.3483564","DOIUrl":"https://doi.org/10.1145/3477132.3483564","url":null,"abstract":"Ethereum is an emerging distributed computing platform that supports a decentralized replicated virtual machine at a large scale. Transactions in Ethereum are specified in smart contracts, disseminated through broadcast, accepted into the chain of blocks, and then executed on each node. In this new Dissemination-Consensus-Execution (DiCE) paradigm, the time interval between when a transaction is known (during the dissemination phase) to when the transaction is executed (after the consensus phase) offers a window of opportunity to accelerate transaction processing through speculative execution. However, the traditional speculative execution, which hinges on the ability to predict the future accurately, is inadequate because of DiCE's many-future nature. Forerunner proposes a novel constraint-based approach for speculative execution on Ethereum. In contrast to the traditional approach of predicting a single future and demanding it to be perfectly accurate, Forerunner speculates on multiple futures and can leverage speculative results based on imperfect predictions whenever certain constraints are satisfied. Under these constraints, a transaction execution is substantially accelerated through a novel multi-trace program specialization enhanced by a new form of memoization. The fully implemented Forerunner is evaluated as a node connected to the worldwide Ethereum network. When processing 13 million transactions live in real time, Forerunner achieves an effective average speedup of 8.39x on the transactions that it hears during the dissemination phase, which accounts for 95.71% of all the transactions. The end-to-end speedup over all the transactions is 6.06x. The code and data sets are publicly available.","PeriodicalId":38935,"journal":{"name":"Operating Systems Review (ACM)","volume":"22 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2021-10-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"78376070","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 18

HEALER: Relation Learning Guided Kernel Fuzzing 治疗师:关系学习引导核模糊

Operating Systems Review (ACM) Pub Date : 2021-10-26 DOI: 10.1145/3477132.3483547

Hao Sun, Yuheng Shen, Cong Wang, Jianzhong Liu, Yu Jiang, Ting Chen, Aiguo Cui

{"title":"HEALER: Relation Learning Guided Kernel Fuzzing","authors":"Hao Sun, Yuheng Shen, Cong Wang, Jianzhong Liu, Yu Jiang, Ting Chen, Aiguo Cui","doi":"10.1145/3477132.3483547","DOIUrl":"https://doi.org/10.1145/3477132.3483547","url":null,"abstract":"Modern operating system kernels are too complex to be free of bugs. Fuzzing is a promising approach for vulnerability detection and has been applied to kernel testing. However, existing work does not consider the influence relations between system calls when generating and mutating inputs, resulting in difficulties when trying to reach into the kernel's deeper logic effectively. In this paper, we propose HEALER, a kernel fuzzer that improves fuzzing's effectiveness by utilizing system call relation learning. HEALER learns the influence relations between system calls by dynamically analyzing minimized test cases. Then, HEALER utilizes the learned relations to guide input generation and mutation, which improves the quality of test cases and the effectiveness of fuzzing. We implemented HEALER and evaluated its performance on recent versions of the Linux kernel. Compared to state-of-the-art kernel fuzzers such as Syzkaller and Moonshine, HEALER improves branch coverage by 28% and 21%, while achieving a speedup of 2.2x and 1.8x, respectively. In addition, HEALER detected 218 vulnerabilities, 33 of which are previously unknown and have been confirmed by the corresponding kernel maintainers.","PeriodicalId":38935,"journal":{"name":"Operating Systems Review (ACM)","volume":"34 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2021-10-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"79029410","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 25

TwinVisor: Hardware-isolated Confidential Virtual Machines for ARM TwinVisor:用于ARM的硬件隔离机密虚拟机

Operating Systems Review (ACM) Pub Date : 2021-10-26 DOI: 10.1145/3477132.3483554

Ding Li, Zeyu Mi, Yubin Xia, B. Zang, Haibo Chen, Haibing Guan

{"title":"TwinVisor: Hardware-isolated Confidential Virtual Machines for ARM","authors":"Ding Li, Zeyu Mi, Yubin Xia, B. Zang, Haibo Chen, Haibing Guan","doi":"10.1145/3477132.3483554","DOIUrl":"https://doi.org/10.1145/3477132.3483554","url":null,"abstract":"Confidential VM, which offers an isolated execution environment for cloud tenants with limited trust in the cloud provider, has recently been deployed in major clouds such as AWS and Azure. However, while ARM has become increasingly popular in cloud data centers, existing confidential VM designs mainly leverage specialized x86 hardware extensions (e.g., AMD SEV and Intel TDX) to isolate VMs upon a shared hypervisor. This paper proposes TwinVisor, the first system that enables the hardware-enforced isolation of confidential VMs on ARM platforms. TwinVisor takes advantage of the mature ARM TrustZone to run two isolated hypervisors, one in the secure world (called S-visor in this paper) and the other in the normal world (called N-visor), to support normal VMs and confidential VMs respectively. Instead of building a new S-visor from scratch, our design decouples protection from resource management, and reuses most functionalities of a full-fledged N-visor to minimize the size of S-visor. We have built two prototypes of TwinVisor: one on an official ARM simulator with S-EL2 enabled to validate functional correctness and the other on an ARM development board to evaluate performance. The S-visor comprises 5.8K LoCs while the N-visor introduces 906 LoC changes to KVM. According to our evaluation, TwinVisor can run unmodified VM images as confidential VMs while incurring less than 5% performance overhead for various real-world workloads on SMP VMs.","PeriodicalId":38935,"journal":{"name":"Operating Systems Review (ACM)","volume":"107 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2021-10-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"73598867","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 19

Snoopy 史努比

Operating Systems Review (ACM) Pub Date : 2021-10-26 DOI: 10.1145/3477132.3483562

Emma Dauterman, Vivian Fang, I. Demertzis, Natacha Crooks, R. A. Popa

引用次数: 19

Syrup: User-Defined Scheduling Across the Stack 糖浆:用户自定义的跨堆栈调度

Operating Systems Review (ACM) Pub Date : 2021-10-26 DOI: 10.1145/3477132.3483548

Kostis Kaffes, J. Humphries, David Mazières, C. Kozyrakis

{"title":"Syrup: User-Defined Scheduling Across the Stack","authors":"Kostis Kaffes, J. Humphries, David Mazières, C. Kozyrakis","doi":"10.1145/3477132.3483548","DOIUrl":"https://doi.org/10.1145/3477132.3483548","url":null,"abstract":"Suboptimal scheduling decisions in operating systems, networking stacks, and application runtimes are often responsible for poor application performance, including higher latency and lower throughput. These poor decisions stem from a lack of insight into the applications and requests the scheduler is handling and a lack of coherence and coordination between the various layers of the stack, including NICs, kernels, and applications. We propose Syrup, a framework for user-defined scheduling. Syrup enables untrusted application developers to express application-specific scheduling policies across these system layers without being burdened with the low-level system mechanisms that implement them. Application developers write a scheduling policy with Syrup as a set of matching functions between inputs (threads, network packets, network connections) and executors (cores, network sockets, NIC queues) and then deploy it across system layers without modifying their code. Syrup supports multi-tenancy as multiple co-located applications can each safely and securely specify a custom policy. We present several examples of uses of Syrup to define application and workload-specific scheduling policies in a few lines of code, deploy them across the stack, and improve performance up to 8x compared with default policies.","PeriodicalId":38935,"journal":{"name":"Operating Systems Review (ACM)","volume":"19 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2021-10-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"88959078","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 23