Check out the following exciting (at least for me) new paper
OptQC: An optimised parallel quantum compiler
by T. Loke, J. B. Wang, Y.H. Chen,
School of Physics, The University of Western Australia, 6009 Perth, Australia
The Perth paper describes new software, a quantum compiler called OptQC that is based on the CSD (Cosine Sine Decomposition). OptQC is designed to take advantage of high-performance computers with a multiprocessor architecture using MPI.
The first ever quantum compiler software based on the CSD was written by me, and is called Qubiter. Here is a previous blog post of mine giving a general introduction to quantum compilers based on CSD.
Qubiter brings back somewhat sad memories for me. At the same time that I released the Qubiter software publicly circa 1999, I also submitted a patent and published an arXiv paper. My logic was that I would be able to parlay the patent into a government grant to continue research in this promising new idea. I dreamt of bringing together 2 great communities: the quantum computing community and the numerical linear algebra community (the guys who invented CSD and are responsible for the awesome treasures called LAPACK and MatLab, both of which have a CSD subroutine). I befriended an outstanding member of the numerical linear algebra community, Steve Leon, Prof. at Univ. of Mass, Dartmouth. Together we applied for an ARO grant (Mark Everitt was the ARO official in charge of this grant). What happened next is described in my Qubiter webpage
Dec 10, 2002: our grant proposal goes down in flames. Grant Proposal submitted to ARO (Army Research Office) to continue Qubiter research is rejected. Here is the Project Description section of the losing proposal. Here is the website of Prof. Steven J. Leon, with whom I had the honor of writing this proposal.
What I don’t say in my Qubiter webpage is that the proposal was rejected without any explanation or any referee reports. I asked for my evaluations and they said, basically, are you crazy, go away. So then I filed a FOIA request (freedom of information act request) asking for my evaluations. What happened next was kind of funny and sad. They sent me about half a dozen referee reports, all undated (I believe they were written in direct response to the FOIA request, long after the grant winners had been announced.) I remember one referee report in particular: there were about 10 questions. The referee had no comments. He just put the lowest possible grade next to each question. Nice guy. ARO is such an honest, fair and patriotic organization.
The interesting thing is that my Qubiter patent is still active. It’s only a US patent so it doesn’t apply outside the US. But if ARO (or the NSF or the NSA or any other American federal agency) awards a government grant for research applying the CSD to quantum computing, and they don’t ask for my permission first, they will be violating US federal law. I doubt they care though.