Hi all Has there ever been any attempt to write an extension of the sparse-diagonalization code that allows the matrix to be stored across multiple machines? In other words so that one can use the combined processing power and storage of multiple nodes in a cluster? If I see it correctly, alps can use MPI but doesn't share resources across nodes. Is that correct? Thanks Benedikt
Hi Benedikt, Such a code can be written but is not implemented in ALPS Matthias
On 6 Dec 2016, at 17:41, Brandt, Benedikt B <benbra@gatech.edu> wrote:
Hi all
Has there ever been any attempt to write an extension of the sparse-diagonalization code that allows the matrix to be stored across multiple machines? In other words so that one can use the combined processing power and storage of multiple nodes in a cluster?
If I see it correctly, alps can use MPI but doesn't share resources across nodes. Is that correct?
Thanks
Benedikt
---- Comp-phys-alps-users Mailing List for the ALPS Project http://alps.comp-phys.org/ <http://alps.comp-phys.org/>
List info: https://lists.phys.ethz.ch//listinfo/comp-phys-alps-users <https://lists.phys.ethz.ch//listinfo/comp-phys-alps-users> Archive: https://lists.phys.ethz.ch//pipermail/comp-phys-alps-users <https://lists.phys.ethz.ch//pipermail/comp-phys-alps-users>
Unsubscribe by writing a mail to comp-phys-alps-users-leave@lists.phys.ethz.ch <mailto:comp-phys-alps-users-leave@lists.phys.ethz.ch>.
Hi Benedikt, We wrote and released an exact diagonalisation code for sparse diagonalisation on distributed systems sometime ago. This is capable of treating spin half, spin one and spinless fermionic systems, but can easily be extended to treat others. It is no longer under active development but you might find it useful. latest code: https://github.com/nmoran/DoQO paper describing it: http://www.sciencedirect.com/science/article/pii/S0010465511000099 Kind Regards, Niall. On 6 December 2016 at 23:44, Matthias Troyer <troyer@phys.ethz.ch> wrote:
Hi Benedikt,
Such a code can be written but is not implemented in ALPS
Matthias
On 6 Dec 2016, at 17:41, Brandt, Benedikt B <benbra@gatech.edu> wrote:
Hi all
Has there ever been any attempt to write an extension of the sparse-diagonalization code that allows the matrix to be stored across multiple machines? In other words so that one can use the combined processing power and storage of multiple nodes in a cluster?
If I see it correctly, alps can use MPI but doesn't share resources across nodes. Is that correct?
Thanks
Benedikt
---- Comp-phys-alps-users Mailing List for the ALPS Project http://alps.comp-phys.org/
List info: https://lists.phys.ethz.ch//listinfo/comp-phys-alps-users Archive: https://lists.phys.ethz.ch//pipermail/comp-phys-alps-users
Unsubscribe by writing a mail to comp-phys-alps-users-leave@ lists.phys.ethz.ch.
---- Comp-phys-alps-users Mailing List for the ALPS Project http://alps.comp-phys.org/
List info: https://lists.phys.ethz.ch//listinfo/comp-phys-alps-users Archive: https://lists.phys.ethz.ch//pipermail/comp-phys-alps-users
Unsubscribe by writing a mail to comp-phys-alps-users-leave@ lists.phys.ethz.ch.
Hi Niall Thanks so much for pointing this out! This looks very promising, I will give it a try. If we end up using it we will definitely give a reference to the aforementioned paper. Thanks a lot Benedikt ________________________________ From: Comp-phys-alps-users <comp-phys-alps-users-bounces@lists.phys.ethz.ch> on behalf of Niall Moran <nmoran@thphys.nuim.ie> Sent: Wednesday, December 7, 2016 5:23:56 AM To: comp-phys-alps-users@lists.phys.ethz.ch Subject: Re: [ALPS-users] Distributed diagonalization Hi Benedikt, We wrote and released an exact diagonalisation code for sparse diagonalisation on distributed systems sometime ago. This is capable of treating spin half, spin one and spinless fermionic systems, but can easily be extended to treat others. It is no longer under active development but you might find it useful. latest code: https://github.com/nmoran/DoQO paper describing it: http://www.sciencedirect.com/science/article/pii/S0010465511000099 Kind Regards, Niall. On 6 December 2016 at 23:44, Matthias Troyer <troyer@phys.ethz.ch<mailto:troyer@phys.ethz.ch>> wrote: Hi Benedikt, Such a code can be written but is not implemented in ALPS Matthias On 6 Dec 2016, at 17:41, Brandt, Benedikt B <benbra@gatech.edu<mailto:benbra@gatech.edu>> wrote: Hi all Has there ever been any attempt to write an extension of the sparse-diagonalization code that allows the matrix to be stored across multiple machines? In other words so that one can use the combined processing power and storage of multiple nodes in a cluster? If I see it correctly, alps can use MPI but doesn't share resources across nodes. Is that correct? Thanks Benedikt ---- Comp-phys-alps-users Mailing List for the ALPS Project http://alps.comp-phys.org/ List info: https://lists.phys.ethz.ch//listinfo/comp-phys-alps-users Archive: https://lists.phys.ethz.ch//pipermail/comp-phys-alps-users Unsubscribe by writing a mail to comp-phys-alps-users-leave@lists.phys.ethz.ch<mailto:comp-phys-alps-users-leave@lists.phys.ethz.ch>. ---- Comp-phys-alps-users Mailing List for the ALPS Project http://alps.comp-phys.org/ List info: https://lists.phys.ethz.ch//listinfo/comp-phys-alps-users Archive: https://lists.phys.ethz.ch//pipermail/comp-phys-alps-users Unsubscribe by writing a mail to comp-phys-alps-users-leave@lists.phys.ethz.ch<mailto:comp-phys-alps-users-leave@lists.phys.ethz.ch>.
participants (3)
-
Brandt, Benedikt B
-
Matthias Troyer
-
Niall Moran