Proceedings of the 3rd International Conference on High Performance Compilation, Computing and Communications最新文献_第2页

Multivariable based checkpoints to mitigate the long range attack in proof-of-stake based blockchains 基于多变量的检查点，以减轻基于权益证明的区块链中的远程攻击

Proceedings of the 3rd International Conference on High Performance Compilation, Computing and Communications Pub Date : 2019-03-08 DOI: 10.1145/3318265.3318289

Ibrahim Ahmed I. AlMallohi, A. Alotaibi, Rahaf Alghafees, F. Azam, Z. Khan

引用次数: 2

VPS-SDN: cloud datacenters network resource allocation VPS-SDN:云数据中心网络资源分配

Proceedings of the 3rd International Conference on High Performance Compilation, Computing and Communications Pub Date : 2019-03-08 DOI: 10.1145/3318265.3318290

Saleem Karmoshi, Wan Shuo, Fekri Saleh, Jing Li, Ming Zhu

{"title":"VPS-SDN: cloud datacenters network resource allocation","authors":"Saleem Karmoshi, Wan Shuo, Fekri Saleh, Jing Li, Ming Zhu","doi":"10.1145/3318265.3318290","DOIUrl":"https://doi.org/10.1145/3318265.3318290","url":null,"abstract":"Cloud datacenters resource request development have altered resource allocation and management efficiency, presenting a pertinent research topic. Hence, we innovatively outline Virtual Physical Switch-Software Defined Network (VPS-SDN), an application-aware resource allocation method, an innovative model tenants portraying Virtual Datacenter (VDC) network requirements, plus network framework and uplink and downlink bandwidth. Contrasting hose abstraction, VPS-SDN engenders models resembling physical network, neither the TAG abstraction modeling the actual pattern of communication of applications. The VPS-SDN model leverage per-tenant network structure and bandwidth requirements concisely and flexibly depicting actual networks. VPS-SDN engenders a heuristic switch-centered Virtual Network (vNetwork) to Physical Network (pNetwork) mapping diagram, permitting tenants to run their applications using cloud networking to simulate an enterprise environment cloud datacenters, thereby promoting comparable outcomes. Simulation results show that our proposed scheme achieves high network utilization and enable the substrate network to satisfy much larger virtual networks.","PeriodicalId":241692,"journal":{"name":"Proceedings of the 3rd International Conference on High Performance Compilation, Computing and Communications","volume":"3 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-03-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124317633","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: 3

Research load balancing technology of distributed database based on consistent hash 研究基于一致性哈希的分布式数据库负载均衡技术

Proceedings of the 3rd International Conference on High Performance Compilation, Computing and Communications Pub Date : 2019-03-08 DOI: 10.1145/3318265.3318283

Wenbo Gong

{"title":"Research load balancing technology of distributed database based on consistent hash","authors":"Wenbo Gong","doi":"10.1145/3318265.3318283","DOIUrl":"https://doi.org/10.1145/3318265.3318283","url":null,"abstract":"Consistent hash algorithm is applied to build a distributed database load balancing model, but this model should deal with thousands of user requests, handle 10 billion data information, along with low-latency response scenarios, all of this is a grim challenge. When load balancing control mechanism is built by the consistency hash, some nodes in the database cluster are overloaded, but some are idle. These load imbalances can seriously do great damage to the overall performance of distributed database system. This paper proposes a detailed description of variance mathematics model about dynamic load balancing, the core is to track system load, evaluation, classification and storage of each node in distributed cluster. This algorithm controls mutual feedback between node load states, idle data node allots the item of overload node, overall suppression of single point overload. After by experimental simulation, compared with the auxiliary loop hash model, this algorithm improves load balancing efficiency by 30% and settles disputes about distributed database load imbalance based on consistent hashing.","PeriodicalId":241692,"journal":{"name":"Proceedings of the 3rd International Conference on High Performance Compilation, Computing and Communications","volume":"179 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-03-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"134375498","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: 1

Parallelization of groundwater flow simulation on multiple GPUs 地下水流动模拟在多gpu上的并行化

Proceedings of the 3rd International Conference on High Performance Compilation, Computing and Communications Pub Date : 2019-03-08 DOI: 10.1145/3318265.3318271

Ho-Sung Sun, Xiaohui Ji, Xu-sheng Wang

引用次数: 1

An effective method for operations placement in Tensor Flow 张量流中操作放置的一种有效方法

Proceedings of the 3rd International Conference on High Performance Compilation, Computing and Communications Pub Date : 2019-03-08 DOI: 10.1145/3318265.3318270

Junnan Liu, Chengfan Jia, Junshi Chen, Han Lin, Xu Jin, Hong An

{"title":"An effective method for operations placement in Tensor Flow","authors":"Junnan Liu, Chengfan Jia, Junshi Chen, Han Lin, Xu Jin, Hong An","doi":"10.1145/3318265.3318270","DOIUrl":"https://doi.org/10.1145/3318265.3318270","url":null,"abstract":"Recent works in deep learning have shown that large neural networks can dramatically improve performance, followed by is the growth of computational requirements for hardware. To address those requirements, a common approach is to train those models on heterogeneous systems with a mixture of hardware devices such as CPUs and GPUs. Normally, the decision of putting parts of neural networks on devices is made by researchers based on heuristics algorithm. In this paper, we introduce an effective method to optimize operations placement for TensorFlow computational graphs on heterogeneous systems by using deep neural networks to predict devices for each operation in a target computational graph. Based on reinforcement learning, our method learns to group operations and assign each group to a corresponding device. To take advantage of the information of operations, we use a fully-connected network to group operations. In addition, we use the actual running time of the predictive placement as rewards to train the predictive network by using policy gradients. By executing the most widely used models in computer vision and machine translation, our method finds an optimized placement which outperforms human experts. When applying our method to the Neural Machine Translation model on the WMT14 German-English dataset, the execution time of per single training step reduces up to 28.41%.","PeriodicalId":241692,"journal":{"name":"Proceedings of the 3rd International Conference on High Performance Compilation, Computing and Communications","volume":"30 4 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-03-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129928801","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: 2

Temperature-aware workload management for sustainable datacenters powered by renewable energy 可再生能源驱动的可持续数据中心的温度感知工作负载管理

Proceedings of the 3rd International Conference on High Performance Compilation, Computing and Communications Pub Date : 2019-03-08 DOI: 10.1145/3318265.3318287

Yuling Li, Xiaoying Wang, Peicong Luo, Xuejiao Yang

引用次数: 1

Design and implementation of an edge computing platform architecture using Docker and Kubernetes for machine learning 使用Docker和Kubernetes进行机器学习的边缘计算平台架构的设计和实现

Proceedings of the 3rd International Conference on High Performance Compilation, Computing and Communications Pub Date : 2019-03-08 DOI: 10.1145/3318265.3318288

Yuzhou Huang, Kaiyu Cai, R. Zong, Yugang Mao

{"title":"Design and implementation of an edge computing platform architecture using Docker and Kubernetes for machine learning","authors":"Yuzhou Huang, Kaiyu Cai, R. Zong, Yugang Mao","doi":"10.1145/3318265.3318288","DOIUrl":"https://doi.org/10.1145/3318265.3318288","url":null,"abstract":"Huge data sets and high resources consumption are the prominent features of machine learning services. At present, machine learning services are often deployed on large-scaled cloud servers. The cloud utilizes its rich resources to perform the model training and prediction tasks, but the performance of this method is often limited by the unstable network conditions. To combine the rich-resources advantage of the cloud server with the stable-network performance of the edge computing technology, this paper proposes a Cloud-training and Edge-predicting framework. By integrating the Docker container technology and Kubernetes container choreography technology, we build an edge computing platform, and deploy a machine learning model (Inception V3) on the platform. With this method, we implemented machine learning services on the edge side. In this paper, we have described the designing and building process of the edge computing platform and the deployment procedure of the machine learning model in detail, and we have taken an experiment to implement the service to prove the feasibility of our ideas.","PeriodicalId":241692,"journal":{"name":"Proceedings of the 3rd International Conference on High Performance Compilation, Computing and Communications","volume":"34 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-03-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114238777","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: 14

Fast sparse kernel summation on cartesian grids: an on-chip algorithm for 3D implicit surface visualization 直角网格上的快速稀疏核求和:一种三维隐式曲面可视化的片上算法

Proceedings of the 3rd International Conference on High Performance Compilation, Computing and Communications Pub Date : 2019-03-08 DOI: 10.1145/3318265.3318278

Shengxin Zhu, A. Wathen

引用次数: 2

A spatial text query scheme based on semantic-aware 一种基于语义感知的空间文本查询方案

Proceedings of the 3rd International Conference on High Performance Compilation, Computing and Communications Pub Date : 2019-03-08 DOI: 10.1145/3318265.3319613

Hongbo Li, Hong Zhu, Zongmin Cui

引用次数: 1

An energy-efficient and dynamic beaconing strategy for DTN 一种高效动态DTN信标策略

Proceedings of the 3rd International Conference on High Performance Compilation, Computing and Communications Pub Date : 2019-03-08 DOI: 10.1145/3318265.3318273

Lei Zhang, Ruxin Zhi, Jiayin Zhang

引用次数: 0