Predicting heartbeat message arrival time is crucial for the quality of failure detection service over intemet. However, intemet dynamic characteristics make it very difficult to understand message behavior and accura...Predicting heartbeat message arrival time is crucial for the quality of failure detection service over intemet. However, intemet dynamic characteristics make it very difficult to understand message behavior and accurately predict heartbeat arrival time. To solve this problem, a novel black-box model is proposed to predict the next heartbeat arrival time. Heartbeat arrival time is modeled as auto-regressive process, heartbeat sending time is modeled as exogenous variable, the model＇ s coefficients are estimated based on the sliding window of observations and this result is used to predict the next heartbeat arrival time. Simulation shows that this adaptive auto-regressive exogenous （ARX） model can accurately capture heartbeat arrival dynamics and minimize prediction error in different network environments.展开更多
The potential energy curves （PECs） of three low-lying electronic states （X^3∑, a^1△, and a^3△） of SO radical have been studied by ab initio quantum chemical method. The calcula- tions were carried out with the ...The potential energy curves （PECs） of three low-lying electronic states （X^3∑, a^1△, and a^3△） of SO radical have been studied by ab initio quantum chemical method. The calcula- tions were carried out with the full valence complete active space self-consistent field method followed by the highly accurate valence internally contracted multireference configuration in- teraction （MRCI） approach in combination with correlation-consistent basis sets. Effects of the core-valence correlation and relativistic corrections on the PECs are taken into account. The core-valence correlation correction is carried out with the cc-pCVDZ basis set. The way to consider the relativistic correction is to use the second-order Douglas-Kroll Hamiltonian approximation, and the correction is performed at the level of cc-pV5Z basis set. To obtain more reliable results, the PECs determined by the MRCI calculations are also corrected for size-extensivity errors by means of the Davidson modification （MRCI＋Q）. These PECs are extrapolated to the complete basis set limit by the two-point energy extrapolation scheme. With these PECs, the spectroscopic parameters are determined.展开更多
基金the National Basic Research Program of China(No.2003CB314806)China Next Generation Intemet Project(CNGI-04-6-2T)
文摘Predicting heartbeat message arrival time is crucial for the quality of failure detection service over intemet. However, intemet dynamic characteristics make it very difficult to understand message behavior and accurately predict heartbeat arrival time. To solve this problem, a novel black-box model is proposed to predict the next heartbeat arrival time. Heartbeat arrival time is modeled as auto-regressive process, heartbeat sending time is modeled as exogenous variable, the model＇ s coefficients are estimated based on the sliding window of observations and this result is used to predict the next heartbeat arrival time. Simulation shows that this adaptive auto-regressive exogenous （ARX） model can accurately capture heartbeat arrival dynamics and minimize prediction error in different network environments.
文摘The potential energy curves （PECs） of three low-lying electronic states （X^3∑, a^1△, and a^3△） of SO radical have been studied by ab initio quantum chemical method. The calcula- tions were carried out with the full valence complete active space self-consistent field method followed by the highly accurate valence internally contracted multireference configuration in- teraction （MRCI） approach in combination with correlation-consistent basis sets. Effects of the core-valence correlation and relativistic corrections on the PECs are taken into account. The core-valence correlation correction is carried out with the cc-pCVDZ basis set. The way to consider the relativistic correction is to use the second-order Douglas-Kroll Hamiltonian approximation, and the correction is performed at the level of cc-pV5Z basis set. To obtain more reliable results, the PECs determined by the MRCI calculations are also corrected for size-extensivity errors by means of the Davidson modification （MRCI＋Q）. These PECs are extrapolated to the complete basis set limit by the two-point energy extrapolation scheme. With these PECs, the spectroscopic parameters are determined.
