February 07, 2012
Update on "bigadv-16", the new bigadv rollout
As we've mentioned earlier, we have been preparing changes to the bigadv system –– both an increase in the number of cores required (and a shortening of deadlines to match) and the release of some new bigadv projects. The motivation for the core changes is as follows:
Bigadv is intentionally intended for the most powerful machines, which makes it naturally a moving target. Our goal with bigadv is to utilize the most powerful segment of (CPU-based) machines in the FAH project to work on projects that are particularly large (memory utilization, upload/download requirement) and require a large amount of computation. We are all fortunate in that processors get faster over time, so the highest-performing tier of donor machines also gets faster over time. We have a lot of exciting science being enabled by FAH donors, and it takes place at all levels of computational requirement and performance sensitivity. So it wouldn't help the project to have 50% of machines running bigadv. But it also wouldn't be a good match to have some of the older and/or bandwidth-limited machines running these most performance-sensitive projects.
As previously announced, our plan is to shorten the deadlines of the BA projects. As a result, assignments will have a 16 core minimum. We've been developing the new projects for the new "bigadv-16". This development has taken a bit longer than we expected, but we are now completing internal testing and reading beta projects for bigadv-16. We are bringing a new server online for bigadv-16. It will start by offering a new class of bigadv projects, but we will soon add in a number of projects on the same server that are more similar to bigadv projects donors have already seen. We want to make these work units available for testing, but at the same time we are still examining the points yield of these bigadv projects. So the points valuation remains a work in progress; we may alter points, bonuses, and/or deadlines in the process of testing.
Please expect a beta announcement soon for testing these new bigadv-16 work units. Then, after the new bigadv-16 projects stabilize, we will bring the bigadv-12 projects into line (points, deadlines) with the bigadv-16 projects and convert all projects to bigadv-16. We are not sure of the timescale for this yet, as we'd like to test the new projects in a thorough manner. We will endeavor to be as transparent as we can regarding upcoming changes in the bigadv program. Bigadv-8 projects will likely be phased out (and indeed are mostly not being assigned at this time).
As a side note, we recognize that the number of cores is a somewhat crude measure for system performance. Long-term, we have some ideas on how we'd like to improve this and use better metrics. But in the near term, we are using this admittedly imperfect metric.
Thank you for folding and for your support of the bigadv program and FAH more generally.