nyu-cds/jw4339_assignment3

jw4339_assignment3

For Assignemtn 10:

Please find in mpi_assignment_1.py and mpi_assignment_2.py

For Assignemnt 9:

no assignemtn 9

For Assignment 8:

Please find in nbody_numba.py

For Assignment 7:

Please find in nbody_cython.pyx

For Assignment 6:

Please find in calculator.py

For Assignment 5:

Please find in nbody_iter.py

For Assignment 4:

This repository is comment by: Qiming Chen [email protected], Aditi S Nair [email protected], and Michael Christopher Higgins [email protected]

jw4339_assignment3's Issues

Don't need to create local_bodies_dict.keys() every time

Instead of calling local_bodies_dict.keys() in lines 56, 87, and 109 you can create a single global variable:
BODIES_keys = BODIES.keys()

Then where you call local_bodies_dict.keys(), you can instead create a local copy of BODIES_keys. This way, you don't have to create local_bodies_dict.keys() every time advance and report_energy are called.

missed global variable

In line 134,137 you are referencing Bodies instead of local version.

Reference to Global variables

In nbody_opt.py

    local_bodies_dict = BODIES
    body_keys = local_bodies_dict.keys()

In line 55-56/84-85/106-107, the reference to global variables can be in a better way by claiming in the parameter list.

Reduce repeated computations

            factor = dt * ((dx * dx + dy * dy + dz * dz) ** (-1.5))
            v1[0] -= dx * m2 * factor
            v1[1] -= dy * m2 * factor
            v1[2] -= dz * m2 * factor
            v2[0] += dx * m1 * factor
            v2[1] += dy * m1 * factor
            v2[2] += dz * m1 * factor

Here m2 * factor and m1 * factor are executed 3 times, which can be simplified for better performance.

Pre-compute values in BODIES

You can pre-compute the values in BODIES dict, so you don't have to keep looking up the global variables PI, SOLAR_MASS, and DAYS_PER_YEAR (and you can save time with the computation).

For example:

'sun': ([0.0, 0.0, 0.0], [0.0, 0.0, 0.0], SOLAR_MASS)

can be:

'sun': ([0.0, 0.0, 0.0], [0.0, 0.0, 0.0], 39.4784176044),

I think this doesn't make the code less clear, because there is no explanation of the values in the dictionary anyway.