Hello, A change in allocation has been requested: Requester: sinclair (Donald Sinclair) Project: Lattice-QCD Title: Lattice simulations of Conformal and Walking Technicolor. Description: We perform simulations to evaluate the functional integrals of QCD-like theories formulated on a discrete space-time lattice, to enable determination of the non-perturbative aspects these theories. These include the properties of these theories at non-zero temperature, including the scales of confinement and chiral symmetry breaking, and such zero temperature properties as spectra (including the Higgs mass), decay constants and the running of the gauge coupling constant. We are particularly interested in those theories where the coupling constant evolves very slowly, since these are candidate 'Walking Technicolor' theories. Related to these are theories with an infrared fixed point (conformal field theories). Our first goal is to differentiate between these two different types of behaviour for candidate theories. We are performing simulations of QCD-like theories which are models for Walking or Conformal Technicolor. We have been studying theories which are essentially QCD but with colour-sextet rather than colour-triplet quarks. We hope to measure the running of the QCD coupling constant. For 2 or 3 flavours, 2-loop perturbation theory predicts an infrared fixed point. For 2 flavours, it is possible that a chiral condensate forms before this fixed point is reached. If so, the fixed point is avoided, the theory is confining, and chiral symmetry breaks spontaneously. However, there is a region where the coupling constant evolves very slowly. These are the properties required for a walking technicolor theory. Simulations we have performed so far at finite temperature suggest that this theory might walk. This is an ongoing program at NERSC and NICS. We have been performing simulations of the 2-flavour theory on Fusion close to the chiral transition on 24^3*8 lattices. This supplements the work performed on Carver at NERSC on 16^3*8 lattices, by testing for finite spatial volume effects. Preliminary results suggest that these finite volume efffects are small (see our progress report). We will complete these simulations on Blues in FY2014. In FY2014 we will start simulations of another candidate Walking-Technicolor model, a gauge theory in which the 'colour' group is SU(2) [rather than SU(3)], and the 'quarks' are 3 'flavours' of Majorana (real) fermions in the adjoint representation of the 'colour' group. We will start by simulating the finite temperature behaviour of this theory on 8^3*4 and 16^3*8 lattices on Blues. This will be supplemented by simulations on 12^3*6 lattices on Carver at NERSC. Our sextet quark codes are based on our earlier triplet quark codes and use the RHMC simulation method. The Rational Hybrid Monte Carlo (RHMC) is a stochastic molecular dynamics algorithm. The functional integral of QCD is written as a partition function of a classical field theory evolving in a fictitious time. The determinant of the Dirac operator raised to a fractional power is calculated by introducing bosonic fields (pseudofermions), and sandwiching this Dirac operator raised to minus said fractional power between them. This fractional power of the Dirac operator is approximated to machine accuracy by a rational approximation. After defining this this theory on a discrete space-time lattice, the inversions required by the partial-fraction expansion of the rational approximation are performed using Krylov space methods, in particular a multi-shift extension of the conjugate gradient algorithm. A global Metropolis Monte-Carlo accept/reject step applied at the end of each trajectory removes discretization errors introduced by the numerical integration of these stochastic equations of motion. We parallelize the code by assigning a fixed number of adjacent lattice sites to each MPI task. Network bandwidth ultimately limits how small a chunk of the lattice can be assigned to each task. Our SU(2) codes will also be based on our colour-triplet SU(3) codes. In fact the fermionic part of the code involves replacing the complex 3-vector fermion fields on the lattice sites with real 3-vector fields, and the complex 3 X 3 matrix gauge fields acting on the fermions with real 3 X 3 matrix gauge fields. The simulation algorithms will be identical to those described above. A short benchmark run of the 12^3*8 sextet code on 192 cores of Blues, indicates a performance of 517 Gflops or 2.7 Gflops/core. In each case we have checked on Fusion the per core performance remains nearly constant as the number of cores is varied, so that maximum throughput is achieved at the largest number of cores allowed by the current implementation (in this case 192). We expect similar scaling behaviour on Blues. We expect the SU(2) codes to perform similarly. The 8^3*4 simulations will be run on 16 or 32 cores (1 or 2 nodes), and the 16^3*8 simulations will be performed on 128 cores (8 nodes). It should be noted that such small (in numbers of cores) jobs are not well suited to the queue structures on Hopper or Edison at NERSC or Kraken at NICS, where large jobs are preferred. In addition, these jobs require a number of cores which is a power of 2, so that on machines that have 24 or 12 cores per node, cores would be left idle. Current: undetermined amount Justification: The measured performance of the 24^3*8 sextet code running on 192 cores of Blues is 2.7 Gflops/core. On a long production run the performance of Blues and Fusion running on the same number of cores (192) was similar. (The Blues run took 46hours 44minutes and the Fusion run took 44hours 21minutes.) Scaling tests on Fusion have shown close to linear speedups with the number of cores used, and we suspect that the same is true for Blues. Our first quarter estimates are based on the assumption that this quarter will be devoted to finishing our sextet runs with 25 1000-trajectory runs on a 24^3*8 lattice using 192 cores. We estimate that each run will take an average of 50 wallclock-hours. In the second and subsequent quarters we will switch to the new SU(2) codes. Based on past experience, we estimate that producing, debugging, testing and tuning of these codes will take approximately 1 man-week. We make our time estimates based on the assumption that we will run either 4 32-core simulations on an 8^3*4 lattice, or 1 128-core simulation on a 16^3*8 lattice, 80% of the time. Because the fermion part of the SU(2) code is a real version of the SU(3) fundamental (triplet) fermion code, and this part of the code uses most of the CPU time in production, the SU(2) code is expected to scale similarly to the SU(3) codes. It is to be hoped that information concerning the graphics coprocessor will be forthcoming, allowing us to make the best use of Blues, thus increasing our productivity. For more information concerning code performance and scaling, read previous years' requests. Requested: 50000 A specific reason has been given: Needed to finish the current set of runs. This takes into account that the project currently has a negative reserve, This needs to be approved and the final allocation amount decided upon. Thank You, The LCRC Accounts System