Recent years,the deep learning algorithm has been widely deployed from cloud servers to terminal units.And researchers proposed various neural network accelerators and software development environments.In this article...Recent years,the deep learning algorithm has been widely deployed from cloud servers to terminal units.And researchers proposed various neural network accelerators and software development environments.In this article,we have reviewed the representative neural network accelerators.As an entirety,the corresponding software stack must consider the hardware architecture of the specific accelerator to enhance the end-to-end performance.And we summarize the programming environments of neural network accelerators and optimizations in software stack.Finally,we comment the future trend of neural network accelerator and programming environments.展开更多
Cloud applications are implemented on top of different distributed systems to provide online service.A service request is decomposed into multiple sub-tasks,which are dispatched to different distributed systems compon...Cloud applications are implemented on top of different distributed systems to provide online service.A service request is decomposed into multiple sub-tasks,which are dispatched to different distributed systems components.For cloud providers,monitoring the execution of a service request is crucial to promptly find problems that may compromise cloud availability.In this paper,we present AgamottoEye,to automatically construct request flow from existing logs.AgamottoEye addresses the challenges of analyzing interleaved log instances,and can successfully extract request flow spread across multiple distributed systems.Our experiments with Hadoop2/YARN show that AgamottoEye can analyze 25,050 log instances in 57.4s,and the extracted request flow information is helpful with error detection and diagnosis.展开更多
Ideal homomorphic encryption is theoretically achievable but impractical in reality due to tremendous computing overhead. Homomorphically encrypted databases, such as CryptDB, leverage replication with partially homom...Ideal homomorphic encryption is theoretically achievable but impractical in reality due to tremendous computing overhead. Homomorphically encrypted databases, such as CryptDB, leverage replication with partially homomorphic encryption schemes to support different SQL queries over encrypted data directly. These databases reach a balance between security and efficiency, but incur considerable storage overhead, especially when making backups. Unfortunately, general data compression techniques relying on data similarity exhibit inefficiency on encrypted data. We present CryptZip, a backup and recovery system that could highly reduce the backup storage cost of encrypted databases. The key idea is to leverage the metadata information of encryption schemes and selectively backup one or several columns among semantically redundant columns. The experimental results show that CryptZip could reduce up to 90.5% backup storage cost on TPC-C benchmark.展开更多
基金partially supported by the National Key Research and Development Program of China (under Grant 2017YFB1003101, 2018AAA0103300, 2017YFA0700900, 2017YFA0700902, 2017YFA0700901)the National Natural Science Foundation of China (under Grant 61732007, 61432016, 61532016, 61672491, 61602441, 61602446, 61732002, 61702478, and 61732020)+6 种基金Beijing Natural Science Foundation (JQ18013)National Science and Technology Major Project (2018ZX01031102)the Transformation and Transferof Scientific and Technological Achievements of Chinese Academy of Sciences (KFJ-HGZX-013)Key Research Projects in Frontier Science of Chinese Academy of Sciences (QYZDBSSW-JSC001)Strategic Priority Research Program of Chinese Academy of Science (XDB32050200, XDC01020000)Standardization Research Project of Chinese Academy of Sciences (BZ201800001)Beijing Academy of Artificial Intelligence (BAAI) and Beijing Nova Program of Science and Technology (Z191100001119093)
文摘Recent years,the deep learning algorithm has been widely deployed from cloud servers to terminal units.And researchers proposed various neural network accelerators and software development environments.In this article,we have reviewed the representative neural network accelerators.As an entirety,the corresponding software stack must consider the hardware architecture of the specific accelerator to enhance the end-to-end performance.And we summarize the programming environments of neural network accelerators and optimizations in software stack.Finally,we comment the future trend of neural network accelerator and programming environments.
文摘Cloud applications are implemented on top of different distributed systems to provide online service.A service request is decomposed into multiple sub-tasks,which are dispatched to different distributed systems components.For cloud providers,monitoring the execution of a service request is crucial to promptly find problems that may compromise cloud availability.In this paper,we present AgamottoEye,to automatically construct request flow from existing logs.AgamottoEye addresses the challenges of analyzing interleaved log instances,and can successfully extract request flow spread across multiple distributed systems.Our experiments with Hadoop2/YARN show that AgamottoEye can analyze 25,050 log instances in 57.4s,and the extracted request flow information is helpful with error detection and diagnosis.
基金National Key R&D Program of China (2016YFB1000201)the National Natural Science Foundation of China (Grant Nos. 61420106013 and 61702480).
文摘Ideal homomorphic encryption is theoretically achievable but impractical in reality due to tremendous computing overhead. Homomorphically encrypted databases, such as CryptDB, leverage replication with partially homomorphic encryption schemes to support different SQL queries over encrypted data directly. These databases reach a balance between security and efficiency, but incur considerable storage overhead, especially when making backups. Unfortunately, general data compression techniques relying on data similarity exhibit inefficiency on encrypted data. We present CryptZip, a backup and recovery system that could highly reduce the backup storage cost of encrypted databases. The key idea is to leverage the metadata information of encryption schemes and selectively backup one or several columns among semantically redundant columns. The experimental results show that CryptZip could reduce up to 90.5% backup storage cost on TPC-C benchmark.
