Released in NAMD 2.11

The following features have been released in NAMD 2.11. Any bug fixes will appear in the nightly build version on the download site. Any documentation updates will appear in the Nightly Build User's Guide (online or 5.0M PDF) and release notes.

GPU-accelerated simulations up to twice as fast as NAMD 2.10

For example, on a dual E5-2680 v3 workstation with two GTX 980 GPUs the 92K-atom ApoA1 benchmark drops from 12.2 ms/step to 5.6 ms/step while the 1.06M-atom STMV benchmark drops from 79 ms/step to 57 ms/step. The largest performance increases are for implicit solvent simulations or runs that output energy on every timestep.

Enhanced Tcl scripting of collective variables and "cv" command

Collective variables module improvements including to histogram bias

TclForces query total net forces for atom groups

Replica-exchange multiplexing (fewer partitions than replicas)

New asynchronous inter-replica communication commands (described below) enable an arbitrary number of "logical" replicas to be multiplexed onto a smaller number of "physical" replical partitions. In short, the +replicas argument no longer needs to match the num_replicas variable, and replica exchange can be performed even on multicore builds. Multiplexing reduces the impact of performance differences between replicas and avoids synchronizing writes to output files. Multiplexed replica exchange is implemented in the new "multiplex.namd" script and uses the same configuration files as the still valid "replica.namd" script.

Tcl scripting multiple in-memory checkpoints

The new commands checkpointStore, checkpointLoad, checkpointSwap, and checkpointFree take a string key argument, by which checkpoints are distinguished.

Improved Tcl scripting multi-copy interface documentation

Tcl scripting asynchronous multi-copy remote in-memory checkpoints

The new checkpointStore/Load/Swap/Free commands take an optional second argument that is either the target replica on which to store the checkpoint or the keyword "global".

Tcl scripting asynchronous multi-copy remote trajectory output

The new command "replicaDcdFile index ?filename?" opens a trajectory file that is written to by any replica that uses the same index.

Tcl scripting asynchronous multi-copy remote script evaluation

The new command "replicaEval replica script" asynchronously executes its script in the top-level context of the target replica's Tcl interpreter and returns the result or error.

Tcl scripting asynchronous multi-copy central work queue support

See the enqueue_work and schedule_work procedures in the new "multiplex.namd" script.

Tcl scripting asynchronous multi-copy workflow-style programming

See the dependent_work and dependent_set procedures in the new "multiplex.namd" script.

Improved minimization for Drude force field and rigid bonds

Rigid bonds and Drude hard-wall constraints are now enforced during minimization.

Fix long-range LJ correction for systems larger than 11,000 atoms

An integer overflow in "16*numatoms*numatoms" resulted in the VDW tail correction (which is off by default) being too small and possibly of the wrong sign for all but the smallest systems, with the magnitude of the applied (incorrect) correction declining as system volume increased, as if no correction was applied.

Improved long-range LJ correction with VDW force switching

Improved alchemical calculations with VDW force switching

Pressure calculation with fixed atoms on GPU works as on CPU

Improved scaling for GPU-accelerated particle-mesh Ewald calculation

CPU-side operations overlap better and are parallelized across cores.

Improved scaling for GPU-accelerated simulations

Nonbonded force calculation results are streamed from the GPU for better overlap.

Improved scaling for multi-threaded "smp" builds

Prevent running smp builds with one thread per process

Support trajectory files larger than 2GB on Windows

NVIDIA CUDA GPU-acceleration binaries for Mac OS X

Improved Intel Xeon Phi coprocessor support

Update to Charm++ 6.7.0 with improved multi-copy support

Ignore ioformat statement in CHARMM topology and parameter files

Psfgen improvements including long resids and insertion codes

Psfgen package available in NAMD Tcl interpreter

Similar to VMD, "package require psfgen" will make psfgen commands available in a NAMD config file to enable building and running a structure in a single Tcl interpreter. NAMD and psfgen do not share data structures and must communicate via files.