As mentioned in the class, Search Engine is one of the typical examples in the modern web site category. According to Google's research, 3 seconds seems to be line where users would be more likely to choose a faster engine. So how to generate the best possible result in certain amount of time is important for search engines.
holard
Video streaming may qualify, to a degree. The primary goal is to provide video in real time, and better performance results in better video quality.
ask
Is video-streaming / video-conferencing a time bound scaling problem? To me it also seems like a bandwidth bound problem. It would be great to get some clarification on this.
holard
@ask actually you're probably right about this being a bandwidth bound problem.
rc0303
Another example would be multiplayer games with a server computing the state of the game world for each client.
shpeefps
Self-driving cars definitely a time constrained example. Doesn't matter if you're able to get the most optimal speed or direction, but it would be extremely important that you know where you're going and don't hurt anyone/violate any rules.
jk2d
This reminds me of the movie The Imitation Game. Alan Turing (played by Benedict Cumberbatch) led a team to break the Enigma machine during World War II. The Enigma machine, as you might already know, is the machine that Germans used to encrypt communications. It changes its settings on a daily basis. Thus, any scaling must be within the time constraint. If the algorithm is not able to break the setting before the Enigma changes it, all the progress will be in vain and you have to repeat the entire process all over again.
eosofsky
Another example would be virtual reality. For VR, it is important to render scenes at a high enough rate in order to prevent motion sickness in users. Thus when we want to render more, we need to do it in the same amount of time.
CSGuy
Finding an Uber/Lyft driver is a time constrained problem, if it takes too long a person would likely try another service.
nishadg
What kind of scaling is prioritized in our personal computers? Or is scaling a concept only applied to highly parallel, high performance machines?
Question: What are some other examples?
As mentioned in the class, Search Engine is one of the typical examples in the modern web site category. According to Google's research, 3 seconds seems to be line where users would be more likely to choose a faster engine. So how to generate the best possible result in certain amount of time is important for search engines.
Video streaming may qualify, to a degree. The primary goal is to provide video in real time, and better performance results in better video quality.
Is video-streaming / video-conferencing a time bound scaling problem? To me it also seems like a bandwidth bound problem. It would be great to get some clarification on this.
@ask actually you're probably right about this being a bandwidth bound problem.
Another example would be multiplayer games with a server computing the state of the game world for each client.
Self-driving cars definitely a time constrained example. Doesn't matter if you're able to get the most optimal speed or direction, but it would be extremely important that you know where you're going and don't hurt anyone/violate any rules.
This reminds me of the movie The Imitation Game. Alan Turing (played by Benedict Cumberbatch) led a team to break the Enigma machine during World War II. The Enigma machine, as you might already know, is the machine that Germans used to encrypt communications. It changes its settings on a daily basis. Thus, any scaling must be within the time constraint. If the algorithm is not able to break the setting before the Enigma changes it, all the progress will be in vain and you have to repeat the entire process all over again.
Another example would be virtual reality. For VR, it is important to render scenes at a high enough rate in order to prevent motion sickness in users. Thus when we want to render more, we need to do it in the same amount of time.
Finding an Uber/Lyft driver is a time constrained problem, if it takes too long a person would likely try another service.
What kind of scaling is prioritized in our personal computers? Or is scaling a concept only applied to highly parallel, high performance machines?