The Experts below are selected from a list of 219 Experts worldwide ranked by ideXlab platform
Jianli Li - One of the best experts on this subject based on the ideXlab platform.
-
CLUSTER Workshops - Transient-Error Detection and Recovery via Reverse Computation and Checkpointing
2012 IEEE International Conference on Cluster Computing Workshops, 2012Co-Authors: Jianjun Xu, Jianli LiAbstract:The integration of error detection and recovery mechanisms becomes mandatory as the probability of the occurrence of transient errors increases. The current study proposes a software-based fault tolerant technique that achieves both detection and recovery. The proposed technique is based on two main mechanisms, namely, reverse Computation and check pointing. This study is the first to introduce reverse Computation for error detection by comparing the input data of the Original Computation and the output data of the reverse Computation. Live variable analysis is introduced to reduce the overhead of the check pointing technique. A translation tool is implemented to make the Original source code fault tolerant with automatic error detection and recovery abilities. Fault injection and performance overhead experiments are performed to evaluate the proposed technique. Experimental results show that most errors can be recovered with relatively low performance overhead.
-
Transient-Error Detection and Recovery via Reverse Computation and Checkpointing
2012 IEEE International Conference on Cluster Computing Workshops, 2012Co-Authors: Jianjun Xu, Jianli LiAbstract:The integration of error detection and recovery mechanisms becomes mandatory as the probability of the occurrence of transient errors increases. The current study proposes a software-based fault tolerant technique that achieves both detection and recovery. The proposed technique is based on two main mechanisms, namely, reverse Computation and check pointing. This study is the first to introduce reverse Computation for error detection by comparing the input data of the Original Computation and the output data of the reverse Computation. Live variable analysis is introduced to reduce the overhead of the check pointing technique. A translation tool is implemented to make the Original source code fault tolerant with automatic error detection and recovery abilities. Fault injection and performance overhead experiments are performed to evaluate the proposed technique. Experimental results show that most errors can be recovered with relatively low performance overhead.
Kevin W. Hamlen - One of the best experts on this subject based on the ideXlab platform.
-
Computation Certification as a Service in the Cloud
2013 13th IEEE ACM International Symposium on Cluster Cloud and Grid Computing, 2013Co-Authors: Safwan Mahmud Khan, Kevin W. HamlenAbstract:This paper proposes a new form of Security as a Service (SECaaS) that allows untrusted, mostly serial Computations in untrusted computing environments to be independently and efficiently validated by trusted, commodity clouds. This addresses the longstanding problem of safely executing high assurance Computations on untrusted hosts. Untrusted Computations are instrumented with a check pointing mechanism that yields a proof of Computation integrity as the Computation progresses. This proof can be validated by a trusted cloud to ensure that the Computation was carried out faithfully. Cloud parallelism and replication is leveraged to validate the proof efficiently even when the Original Computation is not parallel zed. This affords a means of high-assurance, serial Computation on cloud-aware, mobile devices that mix resource-rich but untrusted hardware with trusted but comparatively resource-impoverished hardware components. An implementation for Java and Hadoop MapReduce demonstrates that the approach is effective for commodity VMs, clouds, and software.
Jianjun Xu - One of the best experts on this subject based on the ideXlab platform.
-
CLUSTER Workshops - Transient-Error Detection and Recovery via Reverse Computation and Checkpointing
2012 IEEE International Conference on Cluster Computing Workshops, 2012Co-Authors: Jianjun Xu, Jianli LiAbstract:The integration of error detection and recovery mechanisms becomes mandatory as the probability of the occurrence of transient errors increases. The current study proposes a software-based fault tolerant technique that achieves both detection and recovery. The proposed technique is based on two main mechanisms, namely, reverse Computation and check pointing. This study is the first to introduce reverse Computation for error detection by comparing the input data of the Original Computation and the output data of the reverse Computation. Live variable analysis is introduced to reduce the overhead of the check pointing technique. A translation tool is implemented to make the Original source code fault tolerant with automatic error detection and recovery abilities. Fault injection and performance overhead experiments are performed to evaluate the proposed technique. Experimental results show that most errors can be recovered with relatively low performance overhead.
-
Transient-Error Detection and Recovery via Reverse Computation and Checkpointing
2012 IEEE International Conference on Cluster Computing Workshops, 2012Co-Authors: Jianjun Xu, Jianli LiAbstract:The integration of error detection and recovery mechanisms becomes mandatory as the probability of the occurrence of transient errors increases. The current study proposes a software-based fault tolerant technique that achieves both detection and recovery. The proposed technique is based on two main mechanisms, namely, reverse Computation and check pointing. This study is the first to introduce reverse Computation for error detection by comparing the input data of the Original Computation and the output data of the reverse Computation. Live variable analysis is introduced to reduce the overhead of the check pointing technique. A translation tool is implemented to make the Original source code fault tolerant with automatic error detection and recovery abilities. Fault injection and performance overhead experiments are performed to evaluate the proposed technique. Experimental results show that most errors can be recovered with relatively low performance overhead.
Yi Hong - One of the best experts on this subject based on the ideXlab platform.
-
Phase Precoding for the Compute-and-Forward Protocol.
arXiv: Information Theory, 2014Co-Authors: Amin Sakzad, Emanuele Viterbo, Joseph J. Boutros, Yi HongAbstract:The compute-and-forward (CoF) is a relaying protocol, which uses algebraic structured codes to harness the interference and remove the noise in wireless networks. We propose the use of phase precoders at the transmitters of a network, where relays apply CoF strategy. We define the {\em phase precoded Computation rate} and show that it is greater than the Original Computation rate of CoF protocol. We further give a new low-complexity method for finding network equations. We finally show that the proposed precoding scheme increases the degrees-of-freedom (DoF) of CoF protocol. This overcomes the limitations on the DoF of the CoF protocol, recently presented by Niesen and Whiting. Using tools from Diophantine approximation and algebraic geometry, we prove the existence of a phase precoder that approaches the maximum DoF when the number of transmitters tends to infinity.
-
Phase precoded compute-and-forward with partial feedback
2014 IEEE International Symposium on Information Theory, 2014Co-Authors: Amin Sakzad, Emanuele Viterbo, Joseph Boutros, Yi HongAbstract:In this work, we propose phase precoding for the compute-and-forward (CoF) protocol. We derive the phase precoded Computation rate and show that it is greater than the Original Computation rate of CoF protocol without precoder. To maximize the phase precoded Computation rate, we need to `jointly' find the optimum phase precoding matrix and the corresponding network equation coefficients. This is a mixed integer programming problem where the optimum precoders should be obtained at the transmitters and the network equation coefficients have to be computed at the relays. To solve this problem, we introduce phase precoded CoF with partial feedback. It is a quantized precoding system where the relay jointly computes both a quasi-optimal precoder from a finite codebook and the corresponding network equations. The index of the obtained phase precoder within the codebook will then be fedback to the transmitters. A “deep hole phase precoder” is presented as an example of such a scheme. We further simulate our scheme with a lattice code carved out of the Gosset lattice and show that significant coding gains can be obtained in terms of equation error performance.
Safwan Mahmud Khan - One of the best experts on this subject based on the ideXlab platform.
-
Computation Certification as a Service in the Cloud
2013 13th IEEE ACM International Symposium on Cluster Cloud and Grid Computing, 2013Co-Authors: Safwan Mahmud Khan, Kevin W. HamlenAbstract:This paper proposes a new form of Security as a Service (SECaaS) that allows untrusted, mostly serial Computations in untrusted computing environments to be independently and efficiently validated by trusted, commodity clouds. This addresses the longstanding problem of safely executing high assurance Computations on untrusted hosts. Untrusted Computations are instrumented with a check pointing mechanism that yields a proof of Computation integrity as the Computation progresses. This proof can be validated by a trusted cloud to ensure that the Computation was carried out faithfully. Cloud parallelism and replication is leveraged to validate the proof efficiently even when the Original Computation is not parallel zed. This affords a means of high-assurance, serial Computation on cloud-aware, mobile devices that mix resource-rich but untrusted hardware with trusted but comparatively resource-impoverished hardware components. An implementation for Java and Hadoop MapReduce demonstrates that the approach is effective for commodity VMs, clouds, and software.