diff --git a/src/write_data.cpp b/src/write_data.cpp
new file mode 100644
index 0000000000..e3c98bc672
--- /dev/null
+++ b/src/write_data.cpp
@@ -0,0 +1,617 @@
+/* ----------------------------------------------------------------------
+   LAMMPS - Large-scale Atomic/Molecular Massively Parallel Simulator
+   http://lammps.sandia.gov, Sandia National Laboratories
+   Steve Plimpton, sjplimp@sandia.gov
+
+   Copyright (2003) Sandia Corporation.  Under the terms of Contract
+   DE-AC04-94AL85000 with Sandia Corporation, the U.S. Government retains
+   certain rights in this software.  This software is distributed under
+   the GNU General Public License.
+
+   See the README file in the top-level LAMMPS directory.
+------------------------------------------------------------------------- */
+
+#include "lmptype.h"
+#include "mpi.h"
+#include "string.h"
+#include "write_data.h"
+#include "atom.h"
+#include "atom_vec.h"
+#include "group.h"
+#include "force.h"
+#include "pair.h"
+#include "bond.h"
+#include "angle.h"
+#include "dihedral.h"
+#include "improper.h"
+#include "update.h"
+#include "modify.h"
+#include "domain.h"
+#include "universe.h"
+#include "comm.h"
+#include "output.h"
+#include "thermo.h"
+#include "memory.h"
+#include "error.h"
+
+using namespace LAMMPS_NS;
+
+enum{IGNORE,WARN,ERROR};                    // same as thermo.cpp
+
+/* ---------------------------------------------------------------------- */
+
+WriteData::WriteData(LAMMPS *lmp) : Pointers(lmp)
+{
+  MPI_Comm_rank(world,&me);
+  MPI_Comm_size(world,&nprocs);
+}
+
+/* ----------------------------------------------------------------------
+   called as write_data command in input script
+------------------------------------------------------------------------- */
+
+void WriteData::command(int narg, char **arg)
+{
+  if (domain->box_exist == 0)
+    error->all(FLERR,"Write_data command before simulation box is defined");
+  if (narg != 1) error->all(FLERR,"Illegal write_data command");
+
+  // if filename contains a "*", replace with current timestep
+
+  char *ptr;
+  int n = strlen(arg[0]) + 16;
+  char *file = new char[n];
+
+  if (ptr = strchr(arg[0],'*')) {
+    *ptr = '\0';
+    sprintf(file,"%s" BIGINT_FORMAT "%s",arg[0],update->ntimestep,ptr+1);
+  } else strcpy(file,arg[0]);
+
+  // init entire system since comm->exchange is done
+  // comm::init needs neighbor::init needs pair::init needs kspace::init, etc
+
+  if (comm->me == 0 && screen)
+    fprintf(screen,"System init for write_data ...\n");
+  lmp->init();
+
+  // move atoms to new processors before writing file
+  // do setup_pre_exchange to force update of per-atom info if needed
+  // enforce PBC in case atoms are outside box
+  // call borders() to rebuild atom map since exchange() destroys map
+
+  modify->setup_pre_exchange();
+  if (domain->triclinic) domain->x2lamda(atom->nlocal);
+  domain->pbc();
+  domain->reset_box();
+  comm->setup();
+  comm->exchange();
+  comm->borders();
+  if (domain->triclinic) domain->lamda2x(atom->nlocal+atom->nghost);
+
+  write(file);
+  delete [] file;
+}
+
+/* ----------------------------------------------------------------------
+   called from command()
+   later might let it be directly called within run/minimize loop
+------------------------------------------------------------------------- */
+
+void WriteData::write(char *file)
+{
+  // special case where reneighboring is not done in integrator
+  //   on timestep data file is written (due to build_once being set)
+  // if box is changing, must be reset, else data file will have
+  //   wrong box size and atoms will be lost when data file is read
+  // other calls to pbc and domain and comm are not made,
+  //   b/c they only make sense if reneighboring is actually performed
+
+  //if (neighbor->build_once) domain->reset_box();
+
+  // natoms = sum of nlocal = value to write into data file
+  // if unequal and thermo lostflag is "error", don't write data file
+
+  bigint nblocal = atom->nlocal;
+  MPI_Allreduce(&nblocal,&natoms,1,MPI_LMP_BIGINT,MPI_SUM,world);
+  if (natoms != atom->natoms && output->thermo->lostflag == ERROR)
+    error->all(FLERR,"Atom count is inconsistent, cannot write data file");
+
+  // open data file
+
+  if (me == 0) {
+    fp = fopen(file,"w");
+    if (fp == NULL) {
+      char str[128];
+      sprintf(str,"Cannot open data file %s",file);
+      error->one(FLERR,str);
+    }
+  }
+
+  // proc 0 writes header, ntype-length arrays, force fields
+
+  if (me == 0) {
+    header();
+    type_arrays();
+    force_fields();
+  }
+
+  // per atom info
+
+  if (natoms) atoms();
+  if (natoms) velocities();
+  if (atom->nbonds) bonds();
+  if (atom->nangles) angles();
+  if (atom->ndihedrals) dihedrals();
+  if (atom->nimpropers) impropers();
+
+  // close data file
+
+  if (me == 0) fclose(fp);
+}
+
+/* ----------------------------------------------------------------------
+   proc 0 writes out data file header
+------------------------------------------------------------------------- */
+
+void WriteData::header()
+{
+  fprintf(fp,"LAMMPS data file via write_data, version %s, "
+          "timestep = " BIGINT_FORMAT "\n",
+          universe->version,update->ntimestep);
+
+  fprintf(fp,"\n");
+
+  fprintf(fp,BIGINT_FORMAT " atoms\n",atom->natoms);
+  fprintf(fp,"%d atom types\n",atom->ntypes);
+
+  if (atom->nbonds || atom->nbondtypes) {
+    fprintf(fp,BIGINT_FORMAT " bonds\n",atom->nbonds);
+    fprintf(fp,"%d bond types\n",atom->nbondtypes);
+  }
+  if (atom->nangles || atom->nangletypes) {
+    fprintf(fp,BIGINT_FORMAT " angles\n",atom->nangles);
+    fprintf(fp,"%d angle types\n",atom->nangletypes);
+  }
+  if (atom->ndihedrals || atom->ndihedraltypes) {
+    fprintf(fp,BIGINT_FORMAT " dihedrals\n",atom->ndihedrals);
+    fprintf(fp,"%d dihedral types\n",atom->ndihedraltypes);
+  }
+  if (atom->nimpropers || atom->nimpropertypes) {
+    fprintf(fp,BIGINT_FORMAT " impropers\n",atom->nimpropers);
+    fprintf(fp,"%d improper types\n",atom->nimpropertypes);
+  }
+
+  fprintf(fp,"\n");
+
+  fprintf(fp,"%g %g xlo xhi\n",domain->boxlo[0],domain->boxhi[0]);
+  fprintf(fp,"%g %g ylo yhi\n",domain->boxlo[1],domain->boxhi[1]);
+  fprintf(fp,"%g %g zlo zhi\n",domain->boxlo[2],domain->boxhi[2]);
+
+  if (domain->triclinic)
+    fprintf(fp,"%g %g %g xy xz yz\n",domain->xy,domain->xz,domain->yz);
+}
+
+/* ----------------------------------------------------------------------
+   proc 0 writes out any type-based arrays that are defined
+------------------------------------------------------------------------- */
+
+void WriteData::type_arrays()
+{
+  if (atom->mass) {
+    double *mass = atom->mass;
+    fprintf(fp,"\nMasses\n\n");
+    for (int i = 1; i <= atom->ntypes; i++) fprintf(fp,"%d %g\n",i,mass[i]);
+  }
+}
+
+/* ----------------------------------------------------------------------
+   proc 0 writes out force field info
+------------------------------------------------------------------------- */
+
+void WriteData::force_fields()
+{
+  if (force->pair && force->pair->writedata) {
+    fprintf(fp,"\nPair Coeffs\n\n");
+    force->pair->write_data(fp);
+  }
+  if (atom->avec->bonds_allow && force->bond) {
+    fprintf(fp,"\nBond Coeffs\n\n");
+    force->bond->write_data(fp);
+  }
+  if (atom->avec->angles_allow && force->angle) {
+    fprintf(fp,"\nAngle Coeffs\n\n");
+    force->angle->write_data(fp);
+  }
+  if (atom->avec->dihedrals_allow && force->dihedral) {
+    fprintf(fp,"\nDihedral Coeffs\n\n");
+    force->dihedral->write_data(fp);
+  }
+  if (atom->avec->impropers_allow && force->improper) {
+    fprintf(fp,"\nImproper Coeffs\n\n");
+    force->improper->write_data(fp);
+  }
+}
+
+/* ----------------------------------------------------------------------
+   write out Atoms section of data file
+------------------------------------------------------------------------- */
+
+void WriteData::atoms()
+{
+  // communication buffer for all my Atom info
+  // max_size = largest buffer needed by any proc
+
+  int ncol = atom->avec->size_data_atom + 3;
+
+  int sendrow = atom->nlocal;
+  int maxrow;
+  MPI_Allreduce(&sendrow,&maxrow,1,MPI_INT,MPI_MAX,world);
+
+  double **buf;
+  if (me == 0) memory->create(buf,maxrow,ncol,"write_data:buf");
+  else memory->create(buf,sendrow,ncol,"write_data:buf");
+
+  // pack my atom data into buf
+
+  atom->avec->pack_data(buf);
+
+  // write one chunk of atoms per proc to file
+  // proc 0 pings each proc, receives its chunk, writes to file
+  // all other procs wait for ping, send their chunk to proc 0
+
+  int tmp,recvrow;
+  MPI_Status status;
+  MPI_Request request;
+
+  if (me == 0) {
+    fprintf(fp,"\nAtoms\n\n");
+    for (int iproc = 0; iproc < nprocs; iproc++) {
+      if (iproc) {
+        MPI_Irecv(&buf[0][0],maxrow*ncol,MPI_DOUBLE,iproc,0,world,&request);
+        MPI_Send(&tmp,0,MPI_INT,iproc,0,world);
+        MPI_Wait(&request,&status);
+        MPI_Get_count(&status,MPI_DOUBLE,&recvrow);
+      } else recvrow = sendrow;
+      
+      atom->avec->write_data(fp,recvrow,buf);
+    }
+    
+  } else {
+    MPI_Recv(&tmp,0,MPI_INT,0,0,world,&status);
+    MPI_Rsend(&buf[0][0],sendrow*ncol,MPI_DOUBLE,0,0,world);
+  }
+
+  memory->destroy(buf);
+}
+
+/* ----------------------------------------------------------------------
+   write out Velocities section of data file
+------------------------------------------------------------------------- */
+
+void WriteData::velocities()
+{
+  // communication buffer for all my Atom info
+  // max_size = largest buffer needed by any proc
+
+  int ncol = atom->avec->size_velocity + 1;
+
+  int sendrow = atom->nlocal;
+  int maxrow;
+  MPI_Allreduce(&sendrow,&maxrow,1,MPI_INT,MPI_MAX,world);
+
+  double **buf;
+  if (me == 0) memory->create(buf,maxrow,ncol,"write_data:buf");
+  else memory->create(buf,sendrow,ncol,"write_data:buf");
+
+  // pack my velocity data into buf
+
+  atom->avec->pack_vel(buf);
+
+  // write one chunk of velocities per proc to file
+  // proc 0 pings each proc, receives its chunk, writes to file
+  // all other procs wait for ping, send their chunk to proc 0
+
+  int tmp,recvrow;
+  MPI_Status status;
+  MPI_Request request;
+
+  if (me == 0) {
+    fprintf(fp,"\nVelocities\n\n");
+    for (int iproc = 0; iproc < nprocs; iproc++) {
+      if (iproc) {
+        MPI_Irecv(&buf[0][0],maxrow*ncol,MPI_DOUBLE,iproc,0,world,&request);
+        MPI_Send(&tmp,0,MPI_INT,iproc,0,world);
+        MPI_Wait(&request,&status);
+        MPI_Get_count(&status,MPI_DOUBLE,&recvrow);
+      } else recvrow = sendrow;
+      
+      atom->avec->write_vel(fp,recvrow,buf);
+    }
+    
+  } else {
+    MPI_Recv(&tmp,0,MPI_INT,0,0,world,&status);
+    MPI_Rsend(&buf[0][0],sendrow*ncol,MPI_DOUBLE,0,0,world);
+  }
+
+  memory->destroy(buf);
+}
+
+/* ----------------------------------------------------------------------
+   write out Bonds section of data file
+------------------------------------------------------------------------- */
+
+void WriteData::bonds()
+{
+  // communication buffer for all my Bond info
+  // max_size = largest buffer needed by any proc
+
+  int ncol = 3;
+
+  int *tag = atom->tag;
+  int *num_bond = atom->num_bond;
+  int **bond_atom = atom->bond_atom;
+  int nlocal = atom->nlocal;
+  int newton_bond = force->newton_bond;
+
+  int i,j;
+  int sendrow = 0;
+  if (newton_bond) {
+    for (i = 0; i < nlocal; i++)
+      sendrow += num_bond[i];
+  } else {
+    for (i = 0; i < nlocal; i++)
+      for (j = 0; j < num_bond[i]; j++)
+        if (tag[i] < bond_atom[i][j]) sendrow++;
+  }
+
+  int maxrow;
+  MPI_Allreduce(&sendrow,&maxrow,1,MPI_INT,MPI_MAX,world);
+
+  int **buf;
+  if (me == 0) memory->create(buf,maxrow,ncol,"write_data:buf");
+  else memory->create(buf,sendrow,ncol,"write_data:buf");
+
+  // pack my bond data into buf
+
+  atom->avec->pack_bond(buf);
+
+  // write one chunk of atoms per proc to file
+  // proc 0 pings each proc, receives its chunk, writes to file
+  // all other procs wait for ping, send their chunk to proc 0
+
+  int tmp,recvrow;
+  MPI_Status status;
+  MPI_Request request;
+
+  int index = 1;
+  if (me == 0) {
+    fprintf(fp,"\nBonds\n\n");
+    for (int iproc = 0; iproc < nprocs; iproc++) {
+      if (iproc) {
+        MPI_Irecv(&buf[0][0],maxrow*ncol,MPI_DOUBLE,iproc,0,world,&request);
+        MPI_Send(&tmp,0,MPI_INT,iproc,0,world);
+        MPI_Wait(&request,&status);
+        MPI_Get_count(&status,MPI_DOUBLE,&recvrow);
+      } else recvrow = sendrow;
+      
+      atom->avec->write_bond(fp,recvrow,buf,index);
+      index += recvrow;
+    }
+    
+  } else {
+    MPI_Recv(&tmp,0,MPI_INT,0,0,world,&status);
+    MPI_Rsend(&buf[0][0],sendrow*ncol,MPI_DOUBLE,0,0,world);
+  }
+
+  memory->destroy(buf);
+}
+
+/* ----------------------------------------------------------------------
+   write out Angles section of data file
+------------------------------------------------------------------------- */
+
+void WriteData::angles()
+{
+  // communication buffer for all my Angle info
+  // max_size = largest buffer needed by any proc
+
+  int ncol = 4;
+
+  int *tag = atom->tag;
+  int *num_angle = atom->num_angle;
+  int **angle_atom2 = atom->angle_atom2;
+  int nlocal = atom->nlocal;
+  int newton_bond = force->newton_bond;
+
+  int i,j;
+  int sendrow = 0;
+  if (newton_bond) {
+    for (i = 0; i < nlocal; i++)
+      sendrow += num_angle[i];
+  } else {
+    for (i = 0; i < nlocal; i++)
+      for (j = 0; j < num_angle[i]; j++)
+        if (tag[i] == angle_atom2[i][j]) sendrow++;
+  }
+
+  int maxrow;
+  MPI_Allreduce(&sendrow,&maxrow,1,MPI_INT,MPI_MAX,world);
+
+  int **buf;
+  if (me == 0) memory->create(buf,maxrow,ncol,"write_data:buf");
+  else memory->create(buf,sendrow,ncol,"write_data:buf");
+
+  // pack my angle data into buf
+
+  atom->avec->pack_angle(buf);
+
+  // write one chunk of atoms per proc to file
+  // proc 0 pings each proc, receives its chunk, writes to file
+  // all other procs wait for ping, send their chunk to proc 0
+
+  int tmp,recvrow;
+  MPI_Status status;
+  MPI_Request request;
+
+  int index = 1;
+  if (me == 0) {
+    fprintf(fp,"\nAngles\n\n");
+    for (int iproc = 0; iproc < nprocs; iproc++) {
+      if (iproc) {
+        MPI_Irecv(&buf[0][0],maxrow*ncol,MPI_DOUBLE,iproc,0,world,&request);
+        MPI_Send(&tmp,0,MPI_INT,iproc,0,world);
+        MPI_Wait(&request,&status);
+        MPI_Get_count(&status,MPI_DOUBLE,&recvrow);
+      } else recvrow = sendrow;
+      
+      atom->avec->write_angle(fp,recvrow,buf,index);
+      index += recvrow;
+    }
+    
+  } else {
+    MPI_Recv(&tmp,0,MPI_INT,0,0,world,&status);
+    MPI_Rsend(&buf[0][0],sendrow*ncol,MPI_DOUBLE,0,0,world);
+  }
+
+  memory->destroy(buf);
+}
+
+/* ----------------------------------------------------------------------
+   write out Dihedrals section of data file
+------------------------------------------------------------------------- */
+
+void WriteData::dihedrals()
+{
+  // communication buffer for all my Dihedral info
+  // max_size = largest buffer needed by any proc
+
+  int ncol = 5;
+
+  int *tag = atom->tag;
+  int *num_dihedral = atom->num_dihedral;
+  int **dihedral_atom2 = atom->dihedral_atom2;
+  int nlocal = atom->nlocal;
+  int newton_bond = force->newton_bond;
+
+  int i,j;
+  int sendrow = 0;
+  if (newton_bond) {
+    for (i = 0; i < nlocal; i++)
+      sendrow += num_dihedral[i];
+  } else {
+    for (i = 0; i < nlocal; i++)
+      for (j = 0; j < num_dihedral[i]; j++)
+        if (tag[i] == dihedral_atom2[i][j]) sendrow++;
+  }
+
+  int maxrow;
+  MPI_Allreduce(&sendrow,&maxrow,1,MPI_INT,MPI_MAX,world);
+
+  int **buf;
+  if (me == 0) memory->create(buf,maxrow,ncol,"write_data:buf");
+  else memory->create(buf,sendrow,ncol,"write_data:buf");
+
+  // pack my dihedral data into buf
+
+  atom->avec->pack_dihedral(buf);
+
+  // write one chunk of atoms per proc to file
+  // proc 0 pings each proc, receives its chunk, writes to file
+  // all other procs wait for ping, send their chunk to proc 0
+
+  int tmp,recvrow;
+  MPI_Status status;
+  MPI_Request request;
+
+  int index = 1;
+  if (me == 0) {
+    fprintf(fp,"\nDihedrals\n\n");
+    for (int iproc = 0; iproc < nprocs; iproc++) {
+      if (iproc) {
+        MPI_Irecv(&buf[0][0],maxrow*ncol,MPI_DOUBLE,iproc,0,world,&request);
+        MPI_Send(&tmp,0,MPI_INT,iproc,0,world);
+        MPI_Wait(&request,&status);
+        MPI_Get_count(&status,MPI_DOUBLE,&recvrow);
+      } else recvrow = sendrow;
+      
+      atom->avec->write_dihedral(fp,recvrow,buf,index);
+      index += recvrow;
+    }
+    
+  } else {
+    MPI_Recv(&tmp,0,MPI_INT,0,0,world,&status);
+    MPI_Rsend(&buf[0][0],sendrow*ncol,MPI_DOUBLE,0,0,world);
+  }
+
+  memory->destroy(buf);
+}
+
+/* ----------------------------------------------------------------------
+   write out Impropers section of data file
+------------------------------------------------------------------------- */
+
+void WriteData::impropers()
+{
+  // communication buffer for all my Improper info
+  // max_size = largest buffer needed by any proc
+
+  int ncol = 5;
+
+  int *tag = atom->tag;
+  int *num_improper = atom->num_improper;
+  int **improper_atom2 = atom->improper_atom2;
+  int nlocal = atom->nlocal;
+  int newton_bond = force->newton_bond;
+
+  int i,j;
+  int sendrow = 0;
+  if (newton_bond) {
+    for (i = 0; i < nlocal; i++)
+      sendrow += num_improper[i];
+  } else {
+    for (i = 0; i < nlocal; i++)
+      for (j = 0; j < num_improper[i]; j++)
+        if (tag[i] == improper_atom2[i][j]) sendrow++;
+  }
+
+  int maxrow;
+  MPI_Allreduce(&sendrow,&maxrow,1,MPI_INT,MPI_MAX,world);
+
+  int **buf;
+  if (me == 0) memory->create(buf,maxrow,ncol,"write_data:buf");
+  else memory->create(buf,sendrow,ncol,"write_data:buf");
+
+  // pack my improper data into buf
+
+  atom->avec->pack_improper(buf);
+
+  // write one chunk of atoms per proc to file
+  // proc 0 pings each proc, receives its chunk, writes to file
+  // all other procs wait for ping, send their chunk to proc 0
+
+  int tmp,recvrow;
+  MPI_Status status;
+  MPI_Request request;
+
+  int index = 1;
+  if (me == 0) {
+    fprintf(fp,"\nImpropers\n\n");
+    for (int iproc = 0; iproc < nprocs; iproc++) {
+      if (iproc) {
+        MPI_Irecv(&buf[0][0],maxrow*ncol,MPI_DOUBLE,iproc,0,world,&request);
+        MPI_Send(&tmp,0,MPI_INT,iproc,0,world);
+        MPI_Wait(&request,&status);
+        MPI_Get_count(&status,MPI_DOUBLE,&recvrow);
+      } else recvrow = sendrow;
+      
+      atom->avec->write_improper(fp,recvrow,buf,index);
+      index += recvrow;
+    }
+    
+  } else {
+    MPI_Recv(&tmp,0,MPI_INT,0,0,world,&status);
+    MPI_Rsend(&buf[0][0],sendrow*ncol,MPI_DOUBLE,0,0,world);
+  }
+
+  memory->destroy(buf);
+}
diff --git a/src/write_data.h b/src/write_data.h
new file mode 100644
index 0000000000..971de1cc84
--- /dev/null
+++ b/src/write_data.h
@@ -0,0 +1,53 @@
+/* ----------------------------------------------------------------------
+   LAMMPS - Large-scale Atomic/Molecular Massively Parallel Simulator
+   http://lammps.sandia.gov, Sandia National Laboratories
+   Steve Plimpton, sjplimp@sandia.gov
+
+   Copyright (2003) Sandia Corporation.  Under the terms of Contract
+   DE-AC04-94AL85000 with Sandia Corporation, the U.S. Government retains
+   certain rights in this software.  This software is distributed under
+   the GNU General Public License.
+
+   See the README file in the top-level LAMMPS directory.
+------------------------------------------------------------------------- */
+
+#ifdef COMMAND_CLASS
+
+CommandStyle(write_data,WriteData)
+
+#else
+
+#ifndef LMP_WRITE_DATA_H
+#define LMP_WRITE_DATA_H
+
+#include "stdio.h"
+#include "pointers.h"
+
+namespace LAMMPS_NS {
+
+class WriteData : protected Pointers {
+ public:
+  WriteData(class LAMMPS *);
+  void command(int, char **);
+  void write(char *);
+
+ private:
+  int me,nprocs;
+  FILE *fp;
+  bigint natoms;         // natoms (sum of nlocal) to write into file
+
+  void header();
+  void type_arrays();
+  void force_fields();
+  void atoms();
+  void velocities();
+  void bonds();
+  void angles();
+  void dihedrals();
+  void impropers();
+};
+
+}
+
+#endif
+#endif