One use case where it would be beneficial is for tables that have batch-style patterns, where you (at least in my experience) need to provision a good enough baseline before autoscaling kicks in.
In some cases I've seen, autoscaling can manage going from ~500 to 10k WCU without heavy throttling (as the burst capacity can handle the five~ten minutes before autoscaling kicks in), but not with a smaller baseline. On tables with hot shards, the baseline usually has to be higher.
This kind of clause is usually known as a "DeWitt clause"[1], after a University of Wisconsin professor who benchmarked a couple databases, including Oracle (which performed poorly). Oracle/Larry Ellison didn't react well to that and decided to forbid benchmarks.
It seems more similar in spirit to triplestores, which are pretty neat all things considered. It also reminds me a lot of Datomic, but without the Clojure-ness and immutability.
44 million, according to their paper, and they used 5000 TPUs, which are capable of 4.6×10^17 operations per second.
(The operations the TPU can run are far simpler than what supercomputers can do, but just for the sake of comparison, the current top supercomputer in the world can do 1.25×10^17 floating point operations per second)
>So, using sports as a proxy is not the same as management decisions where there are no well-defined criteria or expected value.
Wouldn't it make his result even stronger? If the thesis fails to hold even in better conditions (like when there is a well-defined expected value), it should also fail to hold in worse conditions.
Aren't the bitcoin transaction fees actually higher than what you would get with banks? I don't know how it works in the US, but I can do free almost-instant transfers inside Germany with no problems at all, including for online transactions (like Amazon).
Not only that, but also languages you dislike enough to tell everyone who reads your CV you won't work with them, like ACow_Adonis said. I know many people who avoid R as much as they can, but would rather work with it than be unemployed.
Spark SQL does exactly that: declarative graph dataflow computations + a multicore (also distributed) scheduler and executor, although it doesn't really scale down that well to a single machine and the execution layer happens mostly on the JVM, even when using PySpark (with an ungodly mix of of serialized Python and JVM objects and code).
Ibis (also by Wes McKinney) does the first part, but it offloads scheduling and execution to the underlying database you are using.
Although China has such a large market (and potential to close the market if need be) that not being able to expand outside of China won't be a problem for a long time.
>That European chocolate in American supermarkets is quite often the lowest-quality stuff from here.
I guess this is somewhat universal: the only Brazilian cachaças available here in Germany are the worst ones in Brazil, but they have insane prices (e.g. 12€ for one that isn't worth the 2€ it costs in Brazil).
Mesos actually handles the lower level part (determining which machines are up and which resources they have (CPU, memory, disks, network ports, GPUs), running tasks on machines, isolating resources from each running task on each machine, either with its own containerizer or using Docker, etc), but the tasks themselves and their scheduling details (how many tasks to run, should it try to run each task on a different machine or not, etc) are determined by the framework you are using. There are some popular "meta-frameworks" (like Marathon and Aurora) that abstract away the Mesos interface and let you just say "run five instances of this Docker container, one per physical machine". They also might handle higher level details, like service registration/discovery, rolling updates and more.
>Because they are not down. Their shares are just sent to another worker from the pool seamlessly.
Eh, depends on what you consider by "seamlessly": basically if a node goes down, everything it was running will be rescheduled and ran again on other nodes, but this might take some minutes, and ensuring this failure is not perceived from the outside depends entirely on you. Disk management is entirely up to you also, so you have to roll your own SAN if you want to have something like that (at my company Mesos disk space is completely ephemeral, and truly persistent state is always stored on S3 or external databases).
How did Microsoft solve it with their C++ APIs? I've never ventured much into Windows programming, but I remember hearing that MFC uses C++, and newer stuff is mostly C++-based.
Here's a link to the original: https://www.zdf.de/video/dokus/megacitys-wenn-es-nacht-wird-...