To Dara O Shayda

Hi Dara, when your CUDA/OpenCL server is operational, feel free to send me some login details along with a quick pointer to which compiler it uses for OpenCL. I don't promise to get around to using it very quickly but I might get to try some Azimuth stuff on it (and will be suitably responsible in any usage of your systems.)

My email address is david dot tweed signOfAt gmail dot com.

Regards, Dave

Source Text:hoidle

Comments

  • 1.

    Hello David it is ready now as was promised.

    However it is best if you could give a short synopsis of what you like to code and accomplish, because there are 2 server programmers and myself need to be involved to help out and there are other admin considerations.

    I do this myself with them as well, this way we keep the team enthused.

    Let me know if that works with you

    Dara

    Source Text:Hello David it is ready now as was promised. However it is best if you could give a short synopsis of what you like to code and accomplish, because there are 2 server programmers and myself need to be involved to help out and there are other admin considerations. I do this myself with them as well, this way we keep the team enthused. Let me know if that works with you Dara
  • 2.

    That sounds eminently sensible: Do you want to do this over private email or via the forum here? (I've no idea what your email address is.)

    Source Text:That sounds eminently sensible: Do you want to do this over private email or via the forum here? (I've no idea what your email address is.)
  • 3.

    for the admin stuff private since there are passwords and other security issues, but please submit your synopsis to the group here and see how other researchers react.

    Also in doing so please think of compute intensive applications, if it runs on your laptop then it would be a futile exercise of using GPU, also mind you it is very hard work to get this to work properly, not that I am your boss but requires focus.

    Meanwhile, I introduce you to my server guru and he will guide you through the server, being Fri if he does not get you today, Monday :)

    dara

    Source Text:for the admin stuff private since there are passwords and other security issues, but please submit your synopsis to the group here and see how other researchers react. Also in doing so please think of compute intensive applications, if it runs on your laptop then it would be a futile exercise of using GPU, also mind you it is very hard work to get this to work properly, not that I am your boss but requires focus. Meanwhile, I introduce you to my server guru and he will guide you through the server, being Fri if he does not get you today, Monday :) dara
  • 4.

    I've started another thread on this.

    Source Text:I've started another thread on this.
  • 5.

    Hello David

    Please make sure you are planning what John was talking about with regards to your work and write to me so one of my engineers will set you up on GPU server.

    Please kindly make a calendar, like I did here as John does regularly.

    Dara

    Source Text:Hello David Please make sure you are planning what John was talking about with regards to your work and write to me so one of my engineers will set you up on GPU server. Please kindly make a calendar, like I did here as John does regularly. Dara
  • 6.

    Hi Dara,

    Thank you very much for your kind offer. However, I now believe that the most productive course for me to take is to run my models on a commercial cloud computing provider. I will however make the code that I write available on a personal github account and, when it's in a suitable state, try to move it to the proper Azimuth software repository.

    Best wishes, Dave

    Source Text:Hi Dara, Thank you very much for your kind offer. However, I now believe that the most productive course for me to take is to run my models on a commercial cloud computing provider. I will however make the code that I write available on a personal github account and, when it's in a suitable state, try to move it to the proper Azimuth software repository. Best wishes, Dave
  • 7.

    Hello Sir

    As you please, made the offer and gesture to cooperate.

    Dara

    Source Text:Hello Sir As you please, made the offer and gesture to cooperate. Dara
  • 8.

    I think in my last reply my wording was a bit misleading. What I meant by putting stuff on my github was just that a lot of my programming tends to be a bit scrappy in the initial stages, probably not of suitable quality to be associated with Azimuth initially. The git hub account will be here -- there's nothing there yet.

    I'm sure Dara will do some amazing coding things for Azimuth and it's a very generous offer to use his resources. I'm one of these people who doesn't really fit into an organised management hierarchy (which has a lot to recommend it), I just do stuff like implementation as I figure it out. It's just a difference in working styles, and I'm sure we'll both make good contributions to Azimuth's goals.

    Source Text:I think in my last reply my wording was a bit misleading. What I meant by putting stuff on my github was just that a lot of my programming tends to be a bit scrappy in the initial stages, probably not of suitable quality to be associated with Azimuth initially. The git hub account will be [here](https://github.com/davidtweed/multicoreBilinearRegression) -- there's nothing there yet. I'm sure Dara will do some amazing coding things for Azimuth and it's a very generous offer to use his resources. I'm one of these people who doesn't really fit into an organised management hierarchy (which has a lot to recommend it), I just do stuff like implementation as I figure it out. It's just a difference in working styles, and I'm sure we'll both make good contributions to Azimuth's goals.
  • 9.

    David I totally understand no pressure from me and no structure.

    The algorithms are most difficult and quirky and breakdown and they require several areas of expertise to make use of them i.e. require teamwork on my end.

    Dara

    Source Text:David I totally understand no pressure from me and no structure. The algorithms are most difficult and quirky and breakdown and they require several areas of expertise to make use of them i.e. require teamwork on my end. Dara
Sign In or Register to comment.