Dynamic optimization relies on runtime profile information to improve the performance of program execution. Traditional profiling techniques incur significant overhead and are not suitable for dynamic optimization. In...Dynamic optimization relies on runtime profile information to improve the performance of program execution. Traditional profiling techniques incur significant overhead and are not suitable for dynamic optimization. In this paper, a new profiling technique is proposed, that incorporates the strength of both software and hardware to achieve near-zero overhead profiling. The compiler passes profiling requests as a few bits of information in branch instructions to the hardware, and the processor executes profiling operations asynchronously in available free slots or on dedicated hardware. The compiler instrumentation of this technique is implemented using an Itanium research compiler. The result shows that the accurate block profiling incurs very little overhead to the user program in terms of the program scheduling cycles. For example, the average overhead is 0.6% for the SPECint95 benchmarks. The hardware support required for the new profiling is practical. The technique is extended to collect edge profiles for continuous phase transition detection. It is believed that the hardware-software collaborative scheme will enable many profile-driven dynamic optimizations for EPIC processors such as the Itanium processors.展开更多
We present a study of the dynamic behavior of a two-sublattice spin-5/2 Ising model with bilinear and crystal-field interactions in the presence of a time-dependent oscillating external magnetic field on alternating l...We present a study of the dynamic behavior of a two-sublattice spin-5/2 Ising model with bilinear and crystal-field interactions in the presence of a time-dependent oscillating external magnetic field on alternating layers of a hexagonal lattice by using the Glauber-type stochastic dynamics.The lattice is formed by alternate layers of spins σ=5/2 and S=5/2.We employ the Glauber transition rates to construct the mean-field dynamic equations.First,we investigate the time variations of the average sublattice magnetizations to find the phases in the system and then the thermal behavior of the dynamic sublattice magnetizations to characterize the nature（first-or second-order） of the phase transitions and to obtain the dynamic phase transition（DPT） points.We also study the thermal behavior of the dynamic total magnetization to find the dynamic compensation temperature and to determine the type of the dynamic compensation behavior.We present the dynamic phase diagrams,including the dynamic compensation temperatures,in nine different planes.The phase diagrams contain seven different fundamental phases,thirteen different mixed phases,in which the binary and ternary combination of fundamental phases and the compensation temperature or the L-type behavior strongly depend on the interaction parameters.展开更多
文摘Dynamic optimization relies on runtime profile information to improve the performance of program execution. Traditional profiling techniques incur significant overhead and are not suitable for dynamic optimization. In this paper, a new profiling technique is proposed, that incorporates the strength of both software and hardware to achieve near-zero overhead profiling. The compiler passes profiling requests as a few bits of information in branch instructions to the hardware, and the processor executes profiling operations asynchronously in available free slots or on dedicated hardware. The compiler instrumentation of this technique is implemented using an Itanium research compiler. The result shows that the accurate block profiling incurs very little overhead to the user program in terms of the program scheduling cycles. For example, the average overhead is 0.6% for the SPECint95 benchmarks. The hardware support required for the new profiling is practical. The technique is extended to collect edge profiles for continuous phase transition detection. It is believed that the hardware-software collaborative scheme will enable many profile-driven dynamic optimizations for EPIC processors such as the Itanium processors.
文摘We present a study of the dynamic behavior of a two-sublattice spin-5/2 Ising model with bilinear and crystal-field interactions in the presence of a time-dependent oscillating external magnetic field on alternating layers of a hexagonal lattice by using the Glauber-type stochastic dynamics.The lattice is formed by alternate layers of spins σ=5/2 and S=5/2.We employ the Glauber transition rates to construct the mean-field dynamic equations.First,we investigate the time variations of the average sublattice magnetizations to find the phases in the system and then the thermal behavior of the dynamic sublattice magnetizations to characterize the nature（first-or second-order） of the phase transitions and to obtain the dynamic phase transition（DPT） points.We also study the thermal behavior of the dynamic total magnetization to find the dynamic compensation temperature and to determine the type of the dynamic compensation behavior.We present the dynamic phase diagrams,including the dynamic compensation temperatures,in nine different planes.The phase diagrams contain seven different fundamental phases,thirteen different mixed phases,in which the binary and ternary combination of fundamental phases and the compensation temperature or the L-type behavior strongly depend on the interaction parameters.
