检索结果-内蒙古大学图书馆

Efficient Remote Memory Paging for Disaggregated Memory systems 1

22nd International Conference on Algorithms and Architectures for Parallel Processing, ICA3PP 2022

作者： Wang, Tao Liu, Haikun Jin, Hai National Engineering Research Center for Big Data Technology and System Services Computing Technology and System Lab Cluster and Grid Computing Lab School of Computer Science and Technology Huazhong University of Science and Technology Wuhan430074 China

ISBN: (数字)9783031226779

ISBN: (纸本)9783031226762

Memory disaggregation has attracted increasing attention in recent years because it is a cost-efficient approach to scale memory capacity for applications in a data center. However, the latency of remote memory access is a major concern in disaggregated memory systems. This paper presents VANDI, a virtual memory paging mechanism that allows applications to use remote memory pools transparently. VANDI enables effective data caching and prefetching mechanisms to address the problem of high access latency in disaggregated memory systems. VANDI exploits a low-complexity cache replacement strategy to optimize the asynchronous staging queue so that the remote write latency can be significantly reduced. VANDI can also prefetch data in multi-granularity from a remote memory pool in an adaptive manner, and thus further improves the hit rate of the local cache to reduce the read latency of remote memory. Our extensive experiments using micro-benchmarks show that VANDI can improve the performance of typical remote paging system–Infiniswap by up to 15 × –102 ×. VANDI can also improve the performance of state-of-the-art disaggregated memory system–Valet by 1.2 × –2.7 ×. For typical machine learning workloads, VANDI can achieve 20% to 80% performance improvement compared with the state-of-the-art Valet. © 2023, Springer Nature Switzerland AG.

关键词： Paging systems

来源：评论

学校读者我要写书评

暂无评论

VMckpt:lightweight and live virtual machine checkpointing

引用

Science China(Information Sciences) 2012年第12期55卷 2865-2880页

作者： LIU HaiKun 1,JIN Hai 1,LIAO XiaoFei 1,MA Bo 1 & XU ChengZhong 2 1 services computing technology and system lab,Cluster and Grid computing lab,School of Computer Science and technology,Huazhong University of Science and technology,Wuhan 430074,China 2 Department of Electrical and Computer Engineering,Wayne State University,Detroit,MI 48202,USA 1. Services Computing Technology and System Lab Cluster and Grid Computing Lab School of Computer Science and Technology Huazhong University of Science and Technology Wuhan 430074 China2. Department of Electrical and Computer Engineering Wayne State University Detroit MI [48202 USA

Recent advance of virtualization technology provides a new approach to check-point/restart at the virtual machine(VM) *** contrast to traditional process-level checkpointing,checkpointing at the virtualization layer brings up several advantages,such as compatibility,transparence,flexibility and ***,because the virtualization layer has little semantic knowledge about the operation system and the applications running atop,VM-layer checkpointing requires saving the entire operating system state rather than a single *** overhead may render the approach *** reduce the size of VM checkpoint,in this paper we propose a page eviction scheme and an incremental checkpointing mechanism to avoid saving unnecessary VM pages in the *** keep the system online transparently,we propose a live checkpointing mechanism by saving the memory image in a copy-on-write(COW) *** implement the performance optimization mechanisms in a prototype system,called *** results with a group of representative applications show that our page eviction scheme and incremental checkpointing can significantly reduce the checkpoint file size by up to 87% and shorten the total checkpointing/restart time by a factor of up to 71%,in comparison with the Xens default checkpointing *** observed application downtimes due to checkpointing can be reduced to as small as 300 ms.

关键词： checkpoint/restart copy-on-write page eviction virtual machine

来源：评论

学校读者我要写书评

暂无评论

Data Anonymization for Big Crowdsourcing Data

Data Anonymization for Big Crowdsourcing Data

引用

2019 INFOCOM IEEE Conference on Computer Communications Workshops, INFOCOM WKSHPS 2019

作者： Deng, Xiaofeng Zhang, Fan Jin, Hai National Engineering Research Center for Big Data Technology and System Services Computing Technology and System Lab Cluster and Grid Computing Lab School of Computer Science and Technology Huazhong University of Science and Technology Wuhan430074 China

ISBN: (纸本)9781728118789

In traditional database systems, data anonymization has been extensively studied, it provides an effective solution for data privacy preservation, and multidimensional anonymization scheme among them is widely used. However, without delicate parameter settings, these technologies may cause uncontrollable information loss and decrease the accuracy of data analytic tasks. Furthermore, crowdsourcing data is usually huge in amount and must be distributed stored in clouds, which makes the conventional data anonymization technologies not applicable. In this paper, we propose a framework that uses MapReduce to anonymize large-scale data before disseminating them to human workers. In order to guarantee the number and distribution of data records to be similar in all nodes, our framework first redistributes the original data to all participating nodes. Then a heuristic two-phase anonymization schema, which can be seamlessly integrated into the framework, is proposed. Experimental results show that with the same objective of privacy, our approach is scalable for large-scale data and can improve the average accuracy of human worker's analytic tasks. © 2019 IEEE.

关键词： Crowdsourcing

来源：评论

学校读者我要写书评

暂无评论

BED: A Block-Level Deduplication-Based Container Deployment Framework 15th

BED: A Block-Level Deduplication-Based Container Deployment ...

引用

15th International Conference on Green, Pervasive, and Cloud computing, GPC 2020

作者： Zhang, Shiqiang Wu, Song Fan, Hao Zou, Deqing Jin, Hai National Engineering Research Center for Big Data Technology and System Services Computing Technology and System Lab Cluster and Grid Computing Lab School of Computer Science and Technology Huazhong University of Scinece and Technology Wuhan430074 China

ISBN: (纸本)9783030642426

Container technology has gained great popularity in cloud environment since containers provide near-native performance and are lighter and less expensive than traditional virtual machines. However, starting up a non-local container, whose image is unavailable locally, is time-consuming. The reason is that pulling an image from a remote registry requires a long time. In this paper, we experimentally find that a lot of redundant data exists among image layers when we pull images from the registry. This redundant data causes additional pull time and makes the startup of a non-local container slower. To minimize the amount of pulled data while providing intact images, we propose a container deployment framework, BED, based on block-level deduplication. To be specific, BED stores an image layer as numerous data blocks and a fingerprint list which is generated based on these data blocks. When BED needs to pull an image, it pulls the fingerprint lists of the image, deduplicates these fingerprint lists, and pulls the data blocks non-existing locally from the registry. Based on the local and pulled blocks, BED reconstructs the image layers for a container. Experiments show that compared with original Docker, BED reduces the time of pulling images by 35% on average and saves about 48% data transmission in the network. © 2020, Springer Nature Switzerland AG.

关键词： Containers

来源：评论

学校读者我要写书评

暂无评论

Affinity-aware proportional share scheduling for virtual machine system

Affinity-aware proportional share scheduling for virtual mac...

引用

International Conference on Grid and Cloud computing

作者： Chen, Huacai Jin, Hai Hu, Kan Services Computing Technology and System Lab. Cluster and Grid Computing Lab. School of Computer Science and Technology Huazhong University of Science and Technology Wuhan 430074 China

ISBN: (纸本)9780769543130

VM (virtual machine) scheduling is a fundamental topic of virtualization, and fairness is its important design goal. Most VMMs (virtual machine monitor) provide PS (proportional share) schedulers. A PS scheduler assign a weight to every VM to declare the computational resource requirement, and VM is allocated CPU cycles proportional to weight. CPU-affinity is a property of VCPU (virtual CPU) to describe which PCPUs (physical CPU) it can run on. However, current definition of weight does not collaborate well with CPU-affinity. It behaves as if all PCPUs are available to all VMs. We expose the issue by using Xen's Credit scheduler, and show that CPU cycles can be fairly allocated to VMs in the default free-mapping case (no affinity restriction), but not so in restricted-mapping cases (when affinities are configured). This fairness issue makes it necessary to extend the meaning of weight, so as to reflect the resource requirements in all cases. In this paper we reconcile the meaning of a domain's weight with CPU-affinity, and improve Credit scheduler as an affinity-aware one. Experimental results show that our affinity-aware proportional share scheduler (named Credit-APS scheduler) brings a good fairness both in free-mapping and restricted-mapping cases. © 2010 IEEE.

关键词： Mapping

来源：评论

学校读者我要写书评

暂无评论

SpaHet: A Software/Hardware Co-design for Accelerating Heterogeneous-Sparsity based Sparse Matrix Multiplication 24

SpaHet: A Software/Hardware Co-design for Accelerating Heter...

引用

61st ACM/IEEE Design Automation Conference, DAC 2024

作者： Huang, Haoqin Yao, Pengcheng An, Zhaozeng Sun, Yufei Hu, Ao Xu, Peng Zheng, Long Liao, Xiaofei Jin, Hai National Engineering Research Center for Big Data Technology and System Services Computing Technology and System Lab Cluster and Grid Computing Lab Huazhong University of Science and Technology Wuhan430074 China Zhejiang Lab Hangzhou311121 China

ISBN: (纸本)9798400706011

Sparse general matrix-matrix multiplication is widely used in data mining applications. Its irregular memory access patterns limit the performance of general-purpose processors, thus motivating many FPGA-based hardware innovations in recent years. Nevertheless, existing accelerators fail to efficiently support heterogeneous input matrix sparsity, which is universal in various real-world applications. With in-depth experimental analysis, we observe that their performance is bottlenecked by their fixed tiling mechanisms, which only alleviate the irregularity of one input matrix. Based on the observation, we propose SpaHet, a software/hardware co-design to accelerate heterogeneous-sparsity based sparse matrix multiplication. SpaHet adopts a dual-adaptive sliding window mechanism to cover the reuse characteristics of both input matrices simultaneously. With a specialized exploration algorithm, the window-based mechanism can automatically find the optimal tiling strategy instead of applying a fixed one based on empirical experience. A sparsity-aware merge tree is also proposed to maximize the output matrix reuse via accumulating intermediate results thoroughly. Our results on a Xilinx Alveo U280 accelerator card show that SpaHet outperforms state-of-the-art CPU-, GPU- and FPGA-based solutions by 7.71×, 1.1×, and 2.74× in performance, respectively. © 2024 Copyright is held by the owner/author(s). Publication rights licensed to ACM.

关键词： Program processors

来源：评论

学校读者我要写书评

暂无评论

Dynamic switching-frequency scaling: Scheduling overcommitted domains in xen VMM

Dynamic switching-frequency scaling: Scheduling overcommitte...

引用

International Conference on Parallel Processing

作者： Chen, Huacai Jin, Hai Hu, Kan Huang, Jian Services Computing Technology and System Lab School of Computer Science and Technology Huazhong University of Science and Technology Wuhan 430074 China

ISBN: (纸本)9780769541563

Virtualization enables multiple guest operating systems run on a single physical platform. These virtual machines may host any types of application, including concurrent HPC programs. Traditionally, VMM schedulers have focused on fairly sharing the processor resources among domains, rarely consider VCPUs' behaviors. However, this can result in poor application performance to overcommitted domains if there are concurrent programs hosted in them. In this paper we review the properties of both Xen's Credit and SEDF schedulers, and show how these schedulers may seriously impact the performance of the communication-intensive and I/O-intensive concurrent applications in overcommitted domains. We discuss the origination of the problem theoretically, and confirm the derived conclusion on benchmarks. A novel approach, that dynamically scales the context switching-frequency by selecting variable time slices according to VCPUs' behaviors, is then proposed to improve the Credit scheduler more adaptive for concurrent applications. The experimental results show that this extended Credit scheduler can improve the performance of communication-intensive and I/O-intensive concurrent applications in overcommitted domains to the same magnitude as in under-committed domains. © 2010 IEEE.

关键词： Virtualization

来源：评论

学校读者我要写书评

暂无评论

High-quality sound device virtualization in Xen para-virtualized environment 1

引用

4th International Conference on Human-Centric computing, HumanCom'11 and the 6th International Conference on Embedded and Multimedia computing, EMC'11

作者： Hu, Kan Zhou, Wenbo Jin, Hai Shao, Zhiyuan Chen, Huacai Services Computing Technology and System Lab. School of Computer Science and Technology Huazhong University of Science and Technology Wuhan China

ISBN: (数字)9789400721050

ISBN: (纸本)9789400721043

I/O virtualization plays an important role in client virtualization, and audio performance directly impacts user experience. Solutions for I/O virtualization includes full device emulation, split driver model and direct I/O. Full device emulation supports sound device virtualization but suffers in heavy load environment. Split driver model has better performance but does not support sound devices. Direct I/O has the best I/O performance which is close to native, but it sacrifices fault isolation and device transparency. In order to improve user audio experience in virtual machine, a method is proposed to implement high-quality sound device virtualization based on split driver model. Built on top of Xen, the frontend and backend drivers of sound device are both provided. The test results show that sound device virtualization based on split driver model enhances user audio experience in Xen virtual environment. © 2011 Springer Science+Business Media B.V.

关键词： Virtualization

来源：评论

学校读者我要写书评

暂无评论

Towards Redundancy-Free Recommendation Model Training via Reusable-aware Near-Memory Processing 24

Towards Redundancy-Free Recommendation Model Training via Re...

引用

61st ACM/IEEE Design Automation Conference, DAC 2024

作者： Liu, Haifeng Zheng, Long Huang, Yu Huang, Haoyan Liao, Xiaofei Hai, Jin National Engineering Research Center for Big Data Technology and System Services Computing Technology and System Lab Cluster and Grid Computing Lab Huazhong University of Science and Technology Wuhan430074 China Zhejiang Lab Hangzhou311121 China

ISBN: (纸本)9798400706011

The memory-intensive embedding layer in recommendation model continues to be the performance bottleneck. While prior works have attempted to improve the embedding layer performance by exploiting the data locality to cache the frequently accessed embedding vectors and their partial sums. However, these solutions rely on the static cache, which is inapplicable in the embedding training scenario where the embedding vectors are updated frequently. To this end, this paper proposes ReFree, a redundancy-free near-memory processing (NMP) solution for recommendation model training. Specifically, ReFree identifies the reusable data in realtime for both embedding layer forward and backward stages and leverages a lightweight NMP architecture to enable redundancy-free near-memory acceleration of the entire embedding training process. Evaluation results on real-world datasets show that ReFree outperforms the state-of-the-art solutions by 10.9× and reduces 5.3× energy consumption on average. © 2024 Copyright is held by the owner/author(s). Publication rights licensed to ACM.

关键词： Embeddings

来源：评论

学校读者我要写书评

暂无评论

Measurement study of PKTown: A P2P-based Multiplayer Online Game supporting system

Measurement study of PKTown: A P2P-based Multiplayer Online ...

引用

International Symposium on Parallel Architectures, Algorithms, and Programming

作者： Tu, Xuping Jin, Hai Jiang, Wenbin Liao, Xiaofei Services Computing Technology and System Lab School of Computer Science and Technology Huazhong University of Science and Technology Wuhan 430074 China

ISBN: (纸本)9780769543123

Recently, Multiplayer Online Game has been extremely popular, in which peer-to-peer technology has been introduced to help with the scaling. This paper presents a measurement study on a P2P-based MOG supporting system, called PKTown, which has been fully deployed over the CERNET. By leveraging a large set of traces obtained recently over a 4-month period, we study the benefit of P2P architecture applying to MOG supporting system, workload characteristics, players' behaviors and impact form a number of system parameters. We believe this provides insights in understanding the fundamental nature of MOG, which can be further explored to enhance the system design. © 2010 Crown Copyright.

关键词： Social networking (online)

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：