Hello from the Ramanujan Group!

Over the past couple of months, we have been collaborating with the Scientific Software Engineering Center at Georgia Tech (https://ssecenter.cc.gatech.edu/), following a grant awarded to us by the Schmidt Futures’ Virtual Institute for Scientific Software (https://www.schmidtfutures.org/our-work-old/virtual-institute-for-scientific-software/). This collaboration aims to enhance our use of the BOINC infrastructure.

We have largely automated the process that translated our algorithm into the BOINC application, including compilation, cryptographic signing, and QA testing. Additionally, we have upgraded how we generate jobs and collect results.

Notably, we will now have significantly improved visibility into the status of workers, common bugs, and issues. This enhancement will enable us to iterate more rapidly and respond more effectively to common problems.

As part of this upgrade, we have changed the cryptographic key and revamped the devops processes surrounding the signing procedure. You will be prompted by the BOINC client to remove and re-add the project.
Please follow these instructions.
We apologize for any inconvenience this may cause.

The first application and associated jobs created through the upgraded process are already operational, running a new algorithm. For more details on this algorithm, including the approach and results from a small-scale experiment involving 1.5 million formulas, please refer to our latest paper below.

Hello from the Ramanujan Group!

Over the past couple of months, we have been collaborating with the Scientific Software Engineering Center at Georgia Tech (https://ssecenter.cc.gatech.edu/), following a grant awarded to us by the Schmidt Futures’ Virtual Institute for Scientific Software (https://www.schmidtfutures.org/our-work-old/virtual-institute-for-scientific-software/). This collaboration aims to enhance our use of the BOINC infrastructure.

Do you plan a gpu app?