ParallelCode - Revision history

import>Jss43: /* To do */

2008-05-02T17:18:08Z

To do

← Older revision		Revision as of 17:18, 2 May 2008
Line 8:		Line 8:

	= To do =		= To do =


	Distribute the UMatCache (and TMAT?).		Distribute the UMatCache (and TMAT?).

import>Jss43: /* To do */

2008-05-02T17:17:59Z

To do

← Older revision		Revision as of 17:17, 2 May 2008
Line 9:		Line 9:
	= To do =		= To do =

	Fix problem with shifting co-densities during FFT.
	--[[User:jss43\|james]] 19:10, 18 February 2008 (GMT)

	Distribute the UMatCache (and TMAT?).		Distribute the UMatCache (and TMAT?).

import>Jss43: /* Parallelising NECI */

2008-05-02T17:17:33Z

Parallelising NECI

← Older revision		Revision as of 17:17, 2 May 2008
Line 17:		Line 17:

	= Parallelising NECI =		= Parallelising NECI =

	Once this is done, we might be close to having NECI completely parallel. Is there anything else we'd need to do?
	--[[User:jss43\|james]] 19:10, 18 February 2008 (GMT)

	See discussion here: [[NECI Parallelization]].		See discussion here: [[NECI Parallelization]].

import>Jss43: /* Status */

2008-05-02T17:17:20Z

Status

← Older revision		Revision as of 17:17, 2 May 2008
Line 4:		Line 4:
	= Status =		= Status =

			The integral engines in CPMD are now fully parallel, tested and merged back to trunk.
	I have (mainly) made the required changes to the integral routines. The only thing to deal with is the shifted integrals, but I think I know how to do this without any further communication between processors. We can calculate the exchange energy, hartree energy and KS determinant in parallel. Tardis is pleasingly fast.
	--[[User:jss43\|james]] 19:10, 18 ~~February~~ 2008 (~~GMT~~)		--[[User:jss43\|james]] 18:17, 2 May 2008 (BST)

	= To do =		= To do =

import>Jss43: /* Parallelising NECI */

2008-04-10T16:03:02Z

Parallelising NECI

← Older revision		Revision as of 16:03, 10 April 2008
Line 20:		Line 20:
	Once this is done, we might be close to having NECI completely parallel. Is there anything else we'd need to do?		Once this is done, we might be close to having NECI completely parallel. Is there anything else we'd need to do?
	--[[User:jss43\|james]] 19:10, 18 February 2008 (GMT)		--[[User:jss43\|james]] 19:10, 18 February 2008 (GMT)

			See discussion here: [[NECI Parallelization]].

import>Jss43 at 19:19, 18 February 2008

2008-02-18T19:19:32Z

← Older revision		Revision as of 19:19, 18 February 2008
Line 1:		Line 1:
	I am currently in the process of parallelising the CPMD-NECI code. Our approach is:		I am currently in the process of parallelising the CPMD-NECI code. Our approach is to note that integrals are the bottleneck, and so we just parallelise the integrals. The bulk of the work is already done for us in CPMD: the wavefunctions and fourier mesh are already distributed over processors, and the FFTs and sums over G-vectors are automatically done in parallel.

	* Integrals are the bottleneck.
	* Parallelise the integrals. This is already (mainly) done in CPMD: the wavefunctions and fourier mesh are already distributed over processors.
	--[[User:jss43\|james]] 19:10, 18 February 2008 (GMT)		--[[User:jss43\|james]] 19:10, 18 February 2008 (GMT)

import>Jss43 at 19:10, 18 February 2008

2008-02-18T19:10:05Z

New page

I am currently in the process of parallelising the CPMD-NECI code. Our approach is:

* Integrals are the bottleneck.
* Parallelise the integrals. This is already (mainly) done in CPMD: the wavefunctions and fourier mesh are already distributed over processors.
--[[User:jss43|james]] 19:10, 18 February 2008 (GMT)

= Status =

I have (mainly) made the required changes to the integral routines. The only thing to deal with is the shifted integrals, but I think I know how to do this without any further communication between processors. We can calculate the exchange energy, hartree energy and KS determinant in parallel. Tardis is pleasingly fast.
--[[User:jss43|james]] 19:10, 18 February 2008 (GMT)

= To do =

Fix problem with shifting co-densities during FFT.
--[[User:jss43|james]] 19:10, 18 February 2008 (GMT)

Distribute the UMatCache (and TMAT?).
The best way to accomplish this might be to split up the calculation into chunks, and have each processor store (only) the information required for its chunk. How should we go about this? Or is it better to split it up on a node-wise basis?
--[[User:jss43|james]] 19:10, 18 February 2008 (GMT)

= Parallelising NECI =

Once this is done, we might be close to having NECI completely parallel. Is there anything else we'd need to do?
--[[User:jss43|james]] 19:10, 18 February 2008 (GMT)