Increase of Googlebot response time after CDN implementation

Problem: We have a site hosted in Germany that is suffixed with “.de”. Recently, we implemented a CDN with Cloudflare. Shortly after our Googlebot response times increased from 60-100ms to 400-500ms. We noticed that the origin from Googlebot requests come from the US. However, this is the usual case as we learned.

There are solution to overcome this, e.g. Caching. However, our goal is to understand the underlying change that led to the increase. We have reviewed a lot of our config in the last days and haven’t yet understood the problem. Cloudflare uses anycast ips, thus a website can not necessarily be geolocated.

Our current hypothesis: The Googlebot takes into account that a site is hosted on a different continent and subtracts a certain amount of response time.

Question aim: Receive hints what to look for to understand the underlying cause.

Mathematica 12.1 start-up time in Linux extremely slow

I installed Mathematica 12.1 on my Linux machine and the start-up time is extremely slow. I see the Spikey logo and messages like:

Initializing Kernel connections 

It takes about 45 seconds to open a standard small notebook.

Running SystemInformation[], I get:

{ {" Version", "12.1.0 for Linux x86 (64-bit) (March 14, 2020)"}, {" Release ID", "12.1.0.0 (6765490, 2020031402)"}, {" Patch Level", "0"}, {" Activation Key", "XXXX"}, {" Machine ID", "XXXX"}  }  { {" Machine Type", "PC"}, {" Operating System", "Unix"}, {" Processor Type", "x86-64"}  } 

I also can see under Packages & Files that 88 packages and 221 files were loaded.

Is someone else experiencing the same problem?

I also have Mathematica 11.2 installed and there, the start-up time is about 3 seconds or less. All the messages from the Spikey logo are shown really fast. Looking at how many Packages & Files have been loaded, I see 30 packages and 183 files.

Could this be the problem? I have 16GB RAM and an Intel i7-8650U CPU.

Does Time Stop apply a magical effect on only the caster, or does it affect everyone else too?

If the answers to this question are anything to go by, time stop seems to be a surprisingly nuanced spell, but one basic disagreement that’s appeared in the comments has been that one camp thinks time stop applies an effect only on the caster, whereas the other thinks it also applies a magical effect to everyone else, too.

Granted, time stop applies a magical effect on the caster – of that, I have no question. But does it also apply a magical effect on everyone else in the multiverse as well?

Why does time stop NOT apply a magical effect on other creatures

Here are the points I’ve understood for the side that thinks time stop applies only to the caster:

  1. It has a range of Self, meaning that it only applies a magical effect on the caster.

  2. There is no spell in D&D that applies an effect to every single creature in the multiverse, except maybe for wish.

Why DOES time stop apply a magical effect on other creatures

Here are the points from the side that thinks time stop applies a magical effect to everyone as well (I’m a member of this camp):

  1. Spells with a range of Self can still apply magical effects to other creatures, such as detect thoughts, dream, magic jar, and spirit guardians.

  2. The spell’s first sentence says: “You briefly stop the flow of time for everyone but yourself,” and that wording seems to directly state that everyone but the caster experiences a magical effect where the flow of time stops for them.

Consequences, implications

There are consequences for choosing one side over the other, which is the motivation for asking this question.

Consequences: Time stop does NOT apply a magical effect to other creatures

Going with the first camp, you can side step the confusion that comes with zones of antimagic and time stop. But, it doesn’t strictly follow the wording of the spell by ignoring the sentence “You briefly stop the flow of time for everyone but yourself.”

In addition, by claiming that spells with a range of Self do not impose a magical effect on other creatures, that same logic can be applied to, say, magic jar, where a humanoid inside a zone of antimagic can still be possessed because the spell has a range of Self, so it affects only the caster, and therefore the possession of other creatures is not a magical effect of the spell. The same logic can be applied to many other spells with a range of Self, some of which are listed above.

Consequences: Time stop DOES apply a magical effect to other creatures

Going with the second camp, yes, you can more closely follow the spell’s text as written. However, you also open the doors to the aforementioned confusion between zones of antimagic and time stop.

If time stop creates a magical effect on every other creature in the multiverse, across planes of existence (which seems to be the valid interpretation of “everyone but yourself” – and take note, this is a 9th level spell on par with wish), then being immune to it in any way (whether you’re inside a zone of antimagic, you are immune via the wish spell, or you’re one of the gods of the multiverse) will allow you to observe as everyone suddenly freezes in time whenever someone casts time stop.

It also fails to offer a resolution to the linked question, where initiative order between someone inside an antimagic field and a caster of time stop is not well defined.

The Question

With all the above in context, the question is as the title says: does Time Stop apply a magical effect on only the caster, or does it affect everyone else too?

If a caster readies Time Stop and casts it as a reaction during another creature’s turn, what happens to that creature’s turn?

Imagine Tim and Charlie are in combat. Tim takes the Ready action to cast time stop, with some trigger specified that should occur on Charlie’s turn. Charlie takes his turn, triggers Tim’s readied action in the middle of that turn, and time stop is cast.

What happens to Charlie’s turn? Does it immediately end? Does Tim get to take his turns, and when he’s done, Charlie can continue where he left off?

Time efficient way to implement Multi-Armed-Bandits?

I’m doing a research on Multi-Armed Bandit (MAB) problem with approx. 1 million arms. In contrast, the number of iterations is of course much larger, about 10-20 million.

Most MAB-algorithms require an argmax operator (argmax of the action space) that has to be executed in each iteration in order to select the current arm (which maximizes a given selection criterion). Regardless of the chosen programming language for implementation, this procedure/ this argmax operator over the entire action space (1 million arms) is very time-consuming.

Does anyone have some ideas on how to implement MAB algorithms in a time-efficient way?

Can a creature take turns as normal if they are inside an Antimagic Field while another creature casts Time Stop?

Time stop is a spell that stops time for other creatures, allowing one creature to take multiple turns in a row. It says:

You briefly stop the flow of time for everyone but yourself. No time passes for other creatures, while you take 1d4 + 1 turns in a row, during which you can use actions and move as normal.

Certainly, this is a magical effect. The spell causes the flow of time to stop for other creatures, and while you are taking multiple turns, no time flows for them. I imagine antimagic field can defeat it. The relevant text says:

Spells. Any active spell or other magical effect on a creature or an object in the sphere is suppressed while the creature or object is in it.

So, imagine combat between Annie, Tim, and Charlie. Ordinarily, initiative might look something like this:

Annie → Charlie → Tim

Suppose Annie casts time stop and rolled a 1 on their d4. Thus, they take 2 turns in a row. Initiative would look like this:

Annie → Annie → Charlie → Tim

Now imagine that Tim cast antimagic field, and following their turn, Annie casts time stop. Suppose they rolled a 1 on their d4 so that they can take 2 turns in a row. What would the initiative order look like?

Here are some possible resolutions I can think of, but none satisfy me totally:

  1. Time stop defeats antimagic field. The initiative order is: Annie → Annie → Charlie → Tim. The reason this is unsatisfactory is that time stop shouldn’t seem to prevail because it’s a spell, and antimagic field defeats spells.

  2. Time stop cannot be cast while there is an active antimagic field, because there exists some creatures you can’t stop time for. The reason this is unsatisfactory is there is no rule that prevents these two spells from being active at the same time. Also, since things like beholders exist, it’s not unreasonable to say there is almost always an active area of antimagic somewhere in the world, and that means time stop can almost never be used.

  3. The caster of time stop and antimagic field take their turns as normal while everyone else is frozen in time. Thematically and narratively, this seems the most logical. So we go through the turn order, treating every turn Alice would have taken as one full round. For this scenario, since Alice takes two turns in a row, then we can imagine two rounds going by. Ordinarily, everyone but Alice takes a turn, but now we unfreeze anyone inside an area of antimagic. So initiative would be: Annie (time stop starts) → Charlie (frozen in time) → Tim (unfrozen) → Annie (time stop ends) → Charlie → Tim. The reason this is unsatisfactory is because we’re advancing the “round count” now, which feasibly triggers things like lair actions that activate on a certain initiative count. It does have the side effect of allowing Tim to act normally though, affecting other creatures if he wants, because he isn’t bound by time stop and Annie isn’t the one doing the violations of the rules of the spell.

  4. There is no answer to this question, and this is solidly in the zone of DM adjudication. This is unsatisfactory because, well, all questions answered that way tend to be unsatisfactory.

So which is it? Or is it an option I haven’t listed here? Can Tim take turns as normal while inside an antimagic field if Annie casts time stop?

Time complexities of state-of-the-art SAT solvers with respect to length of the formula

I am learning about DPLL and CDCL SAT solvers, and I know that they have time complexity exponential to the number of variables.

If I am not mistaken, one of the reasons why most believe P does not equal to NP is that there exists no polynomial-time algorithm for SAT despite tremendous effort from mathematicians and computer scientists. However, the P vs NP problem only cares about the time complexity with respect to the length of the formula, not the number of variables.

If these state-of-the-art SAT solvers were run on 3CNF formulas, then a time complexity exponential to the number of variables implies a time complexity exponential to the length of the 3CNF.

However, if they were run on arbitrary CNFs, whose length can be exponential to the number of variables, then a time complexity exponential to the number of variables would not necessarily imply a time complexity exponential to the length of the CNF.

Thus, a related question is, are these SAT solvers run on 3CNF or CNF while measuring time complexity?