The idea that a publisher can sell a live service game and shut it down in 1 month with no legal repercussions is ridiculous to me.
This is nice and all but the hard part is replication and consistency in a distributed database. In memory has its uses, also disk backed tables can have their uses. Pretty much normal databases already do this, just writing domain logic in stored procs is kind of annoying.
I'd imagine embedding sqlite in your binary using memory tables is equivalent at the moment. Well you'd have to write code to publish table updates to clients. I suppose it has that going for it.
I've seen some hand wavy docs about clustering but nothing concrete.
Alice measures at angle X, gets value V1
Calls Bob on the phone, okay I measured angle X.
Bob measures at angle X, also gets value V1
Bob measures at angle Y, gets value V2.
Bob calls Alice back says, okay I measured angle Y.
Alice measures angle Y, also gets V2.
The correlation here is nobody can do other measurements while the other party is in the process of measuring. Each party can't know the other party is done until traditional communication has happened.If each party acted independently they would randomly change the state on the other side and each party would get what appears to be random values.
In theory models can/could do the same thing. Think of the current text output of a model being those thoughts inside your head, you have an internal scratch space you work on ideas, then after you perform 'chain of thought' on them, you output to the world. What you're outputting to the world isn't your directly thoughts, but more of a calculation of what you wanted to achieve.
Really for LLMs you just need to have the model put it's output to an internal buffer, read that buffer and make sure it makes sense, then output that to the end user.
It's probably not fast or compute efficient to do that at this time, but at some point it likely will be.
Makes sense to what. The LLM doesn't have a goal, other then to spew text that looks like it should be there.
Does redis do async operations? I'm not sure how that would work because it's known to be in memory. But I do know it persists to disk. So basically my question is:
Does almost absolutely every operation on a redis database happen serially? Maybe not every single operation, but in general.
House approves debt ceiling extension through early December: https://edition.cnn.com/2021/10/12/politics/house-vote-debt-...
White House rules out concessions over debt ceiling while GOP refuses to help avert crisis : https://www.washingtonpost.com/us-policy/2021/09/20/white-ho...
House passes debt ceiling increase, sending it to Biden to avoid default hours before deadline : https://www.cnbc.com/2021/12/14/debt-ceiling-democrats-to-vo...
select id from users where id = ? for update;
if row_count() < 1 then raise 'no user' end if;
insert into sub_resource (owner, thing) values (?, ?);
commit;
??