I wonder if xooglers SREs would be interested in organizing regular surveys. It could be useful for both understanding the market landscape, but also to see patterns in career/personal satisfaction trajectories. Current googlers could find it very useful to know when it is time to move on.
Looking at the table of contents, it sounds quite interesting, but I find some bits missing. (Maybe due to 2012 date?)
DC networks generally follow a spine-leaf design, have ToR switches. They are DAGs and trees.
These are effectively ultrametric spaces, which have a lot of nice properties that can be exploited for optimization and ML (e.g. these are generally bipartite graphs, so belief revision is guaranteed to terminate).
In the same vein, basic abstract algebra and group theory (i.e. symmetries) could be quite useful as well.
I had a ~two years setback at work due to covid. It was not due a psychotic episode, but probably the effects felt similarly bad [to my coworkers]: my thoughts were all over the place, I could not carry a deep tech conversation as I would jump on a tangent and carry on different seemingly unrelated topics, I appeared confused and distracted all the time. I had compounding effects of brain fog and, then undiagnosed, ADHD: to the level that going out to get groceries was a struggle.
I have a great manager at work, but we were relatively new then: I had just transferred into the team before the virus broke loose. He was stellar and very supportive. My whole team was, so much that I shared my story as I recovered instead of hiding my diagnosis.
Nonetheless, It took me almost a year to recover from brain fog and rebuild the trust to work at my level. It was really hard to remain focused and get through working on simple things that I could have done 10y ago. It felt like quicksand for my brain. I almost left the company because of it.
Now, with all the layoffs, I wonder if I did the right thing. Am I at risk due to lower than expected performance [my performance had recovered since then, but who knows]? Is it better this way anyway, for a fatter severance package? Would the sparkle of new challenges in a new company gave me a faster personal trajectory?
TBH, I don't really care that much about career for the money. I just need the innovation of a different level of work to keep me motivated.
Borg–at least as it is published in those papers()–and these other schedulers are really design to maximize the availability of large distributed apps.
() The real Borg now does a lot more...
But there is a completely different world if you look elsewhere. In batch systems like supercomputers (HPC), you want to maximize throughput, not availability. It is common to have something closer to your design: a inventory of jobs and scheduler that allocates workers to them. E.g. https://hpc-wiki.info/hpc/Scheduling_Basics
Google Photos is the wrong product to ask this question, IMHO. One of its key highlight is sharing albums with others and preserving them for the future, so why waste a consumer device's energy on unnecessary tasks if the image is going to end in a Google datacenter anyway?
There are published and opensource (Apache license, according to the github LICENSE file) models for Tensorflow Lite which runs on Pixel Edge TPUs (and most embedded devices):
You can find some on https://recovery.com/
If you can afford it, I can personally recommend https://thetreehousethailand.com/