README

Warp is a extensively developed open-source particle-in-cell code designed to
simulate charged particle beams with high space-charge intensity. The name
"Warp" stems from the code's ability to simulate Warped (bent) Cartesian
meshes. This bent-mesh capability allows the code to efficiently simulate
space-charge effects in bent accelerator lattices (resolution can be placed
where needed) associated with rings and beam transfer lines with dipole
bends. Warp was first developed by Alex Friedman in the 1980s at LLNL
following a Livermore Lab model of steerable compiled code modules linked to
an interpreter. The Warp code has been continuously and extensively extended
by David Grote (LLNL/LBNL) from the late 1980s to the present with key
contributions from others in the Warp team to achieve it's present highly
developed state.

Warp has a broad variety of integrated physics models and extensive
diagnostics -- most of which work in multiple dimensions to allow examination
of modeling idealizations within a common framework. The code is setup around
the interactive python interpreter with dynamically loaded compiled code
modules. This allows flexible problem descriptions in python scripts which
can exploit the full versatility of python and compiled packages as well as
allowing interactive steering of runs. A mix of interpreter-linked compiled
code modules and flexible scripts can be employed for highly adaptable and
flexible problem descriptions without modifying or recompiling the source
code. Many diagnostics and tools are built in, mostly written in python,
which are documented in-line. A user can develop additional elaborate tools
and diagnostics, as needed, by extending and building on the scripts within
python for specific applications. Extensive provisions are included to define
realistic lattice elements in accelerator modeling and standard MAD-style
lattice element descriptions can be employed. Warp has a hierarchy of
multi-species models ranging from full 3D, transverse slice x-y (including
pz), and axisymmetric r-z (including ptheta), as well as simple envelope
models useful for problem setup. Warp can operate in a boosted-frame mode. A
broad variety of particle movers and field solvers are available. Particle
movers include leap-frog models as well as gyro-kinetic models. Electrostatic
and electromagnetic field solvers are included. Electrostatic field solvers
include FFT, multi-grid, and multigrid with mesh refinement (both static and
dynamic) options. The fieldsolvers work with bent (x-plane) meshes. A variety
of conducting structures can be loaded on the grid with subgrid resolution
and various boundary conditions can be employed. Particles can be scraped
consistently with conducting structures and secondary particles (for e-cloud
modeling) emitted. Electromagnetic field solvers are available in 3D,
transverse x-y, and r-z packages. There are options to control the dispersion
and smoothing. Extensive options exist to initialize the beam distributions
including space-charge effects and beams can also be consistently injected
off sources using Child-Langmuir-like emission. Warp is also a plasma code
and has been used for electron cloud and plasma modeling. Limited scattering
and inelastic collision models are available, as well as models for particle
interactions with surfaces. Warp can be compiled in serial and in parallel
using pyMPI or mpi4py.