APEmille is a SIMD parallel processor under development at the Italian National Institute for Nuclear Physics (INFN). It is the third machine of the APE family, following Ape and Ape100. APEmille is well suited for Lattice QCD applications, climate modelling, neural networks, computational chemistry, numerical wind tunnels, seismic and combustion simulation. APEmille is a tridimensional array of custom processors each with peak performance of 528Mflop and 8-32 Mbytes of locally addressable RAM. The 2048-processor machine reaches the performance of 1.081Tflops. APEmille is interfaced to the external world through a PCI interface. A powerful programming language named TAO is provided. A C++ compiler is foreseen.

We report on the progress and status of the APEmille project: a SIMD parallel computer with a peak performance in the TeraFlops range which is now in an advanced development phase. We discuss the hardware and software architecture, and present some performance estimates for Lattice Gauge Theory (LGT) applications.

APEmille is a SPMD parallel processor under development at INFN, Italy, in cooperation with DESY, Germany. APEmille is suited for grand challenges computational problems such as QCD simulations, climate modelling, neural networks, computational chemistry, numerical wind tunnels, seismic and combustion simulations. Its 1 Teraflop/s peak performance and its architecture, together with its language features, allow such applications to execute effectively.APEmille is based on an array of custom arithmetic processors arranged on a tridimensional torus. The processor is optimized for complex computations and has a peak performance of 528 Mflop at 66 MHz. Each processing element has 8 Mbytes of locally addressable RAM.On the software side particular emphasis is devoted to the programming languages that will be available (TAO and C++) and their object oriented, dynamic characteristics: with TAO it is possible to develop language extensions similar to the usual HEP notation; with C++ the portability from and towards different platforms is made possible. (C) 1998 Published by Elsevier Science B.V.

We describe the architecture of the APEmille Parallel Computer, the new generation of the APE family of processors optimized for Lattice Gauge Theory simulations. We emphasize the features of the new machine potentially useful for applications in other areas of computational physics.

We report on the progress and status of the APEmille project: a SIMD parallel computer with a peak performance in the TeraFlops range which is now in an advanced development phase. We discuss the hardware and software architecture, and present some performance estimates for Lattice Gauge Theory (LGT) applications.

We present the results of a numerical calculation of the B → K∗γ form factors. The results have been obtained by studying the relevant correlation functions at β = 6.0, on an 183 × 64 lattice, using the O(a)-improved fermion action, in the quenched approximation. From the study of the matrix element 〈K∗|sσμνb|B〉 we have obtained the form factor T1(0) which controls the exclusive decay rate. The results are compared with the recent results from CLEO. We also discuss the compatibility between the scaling laws predicted by the Heavy Quark Effective Theory (HQET) and pole dominance, by studying the mass- and q2-dependence of the form factors. From our analysis, it appears that the form factors follow a mass behaviour compatible with the predictions of the HQET and that the q2-dependence of T2 is weaker than would be predicted by pole dominance.

We present the results of a high statistics lattice calculation of hadronic form factors relevant for D- and B-meson semi-leptonic decays into light pseudoscalar and vector mesons. The results have been obtained by averaging over 170 gauge field configurations, generated in the quenched approximation, at β = 6.0, on a 183 × 64 lattice, using the O(α)-improved SW-Clover action. From the study of the matrix element 〈 K−|Jμ|D0 〉, we obtain f+(0) = 0.78 ± 0.08 and from the matrix element 〈 K∗0|Jμ|D+ 〉 we obtain V(0) = 1.08 ± 0.22, Ai(0) = 0.67 ± 0.11 and A2(0) = 0.49 ± 0.34. We also obtain the ratios V(0)A1(0) = 1.6 ± 0.3 and A2(0)A1(0) = 0.7 ± 0.4. Our predictions for the different form factors are in good agreement with the experimental data, although, in the case of A2(0), the errors are still too large to draw any firm conclusion. With the help of the Heavy Quark Effective Theory (HQET) we have also extrapolated the lattice results to B-meson decays. The form factors follow a behaviour compatible with the HQET predictions. Our results are in agreement with a previous lattice calculation, performed at β = 6.4, using the standard Wilson action.

We describe APEmille, the latest generation of the APE parallel processors. This machine, an evolution of the APE100 concept, is very efficient for LGT simulations as well as for a broader class of applications requiring massive floating point computations. Several new features characterise this evolution. In particular local addressing capabilities are added to all computing nodes. APEmille also exhibits a higher degree of integration with a network of workstations acting as a global host system. An APEmille system in the Teraflops range will be completed in three-four years. The architecture proposed in this paper is being currently simulated and evaluated.

This proposal describes APEmille, a 3-D SIMD scalable parallel processor in theTeraflop range. This machine is very efficient for LGT simulations as well as for abroader class of numeric applications requiring massive intensive floating pointcomputations. The proposed architecture satisfies the requirements for a massivelyparallel scientific processor described in [1].The APEmille architecture grows from APE100[2]. Several new features characterisethis evolution. In particular local...

We present numerical results obtained in full QCD with 2 flavors of Wilson fermions. We discuss the relation between the phase of Polyakov loops and the sea quarks boundary conditions. We report preliminary results about the HMC autocorrelation of the hadronic masses, on a 16(3) x 32 lattice volume, at beta = 5.55 with kappa(sea) = 0.1570.

The polarization of Polyakov type loops is responsible for the difference between quenched and unquenched finite size effects on the QCD mass spectrum. With a numerical simulation, using different sea quarks boundary conditions, we show that we can align the spatial Polyakov loops in a predefined direction. Starting from these results, we propose a procedure to partially remove the Polyakov type contributions in the meson propagators.

We present results for the kaon B parameter obtained by APE-100 from a sample of 100 configurations using the Wilson action and 200 configurations using the Clover action, on a 18(3) x 64 lattice at beta = 6.0. We have also compared the results for B-K using two different values for the effective coupling constant g. We observe a strong dependence of B-K on the prescription adopted for g in the Wilson case, contrary to the results of the Clover case which are almost unaffected by the choice of g.

We present results for the Kaon B parameter from a sample of 200 configurations using the Wilson action and 460 configurations using the Clover action, on a 18^3 × 64 lattice at β=6.0. A slight improvement of the chiral behaviour of B_K is observed due to the Clover action. We have also compared the results for B_K obtained from two different procedures for the boosting of the coupling constant g. We observe a strong dependence of B_K on the prescription adopted for g in the Wilson case, contrary to the results of the Clover case which are almost unaffected by the choice of g. Combining some recently obtained non perturbative estimates for the renormalisation constants with our Clover matrix element, we observe a significant improvement in the chiral behaviour of B_K.

We present results for the kaon B parameter obtained by APE-100 from a sample of 100 configurations using the Wilson action and 200 configurations using the Clover action, on a 18 × 64 lattice at β = 6.0. We have also compared the results for BK using two different values for the effective coupling constant g. We observe a strong dependence of BK on the prescription adopted for g in the Wilson case, contrary to the results of the Clover case which are almost unaffected by the choice of g.

We present results for the radiative decay B --> K*gamma, obtained by using the Clover action at beta = 6.0 on APE. The compatibility between the scaling laws predicted by the Heavy Quark Effective Theory (HQET) and pole dominance is discussed. The final result depends crucially on the assumed q(2)-dependence of the form factors.

We present results for the radiative decay B → K * γ, obtained by using the Clover action at β = 6.0 on APE. The compatibility between the scaling laws predicted by the Heavy Quark Effective Theory (HQET) and pole dominance is discussed. The final result depends crucially on the assumed q 2-dependence of the form factors.