Everyone has to start somewhere, but if you don't know what ZIRP stands for you just don't have the fundamentals/haven't taken enough interest in economics to participate in a conversation related to macroeconomics and posting grumpy comments out of ignorance won't help you. You're going to have to read up on it.
Although FWIW you probably aren't a bot, an AI would likely know all about ZIRP.
I recommend against books; they're expensive and typically the page count is too high for the number of ideas unless you are really interested in the academics of it. I get a lot more out of Wikipedia [0], blogs and YouTube (although the quality of information on YouTube is typically low).
That being said, if you want some recommendations I'd suggest Tyler Cowen [1] as a good clearinghouse for topical ideas and John Cochrane [2] as an interesting read. I picked an article from the last 7 days where you can see the ZIRP in the graphs, it is all the points where US Fed Funds rate was at 0.
It really wasn't, but it did assume the audience was familiar with a specific bit of discourse about, well, ZIRPs. In particular that when debt is super cheap, companies can afford to do a lot of things that would be stupid when capital comes at a higher price. Which generally seems to have been true about a lot of SV companies.
The cause or mutability the attribute has nothing to do with the reasonableness of ascribing a certain prior probability to the attribute.
Also, many hundreds of millions of people in the world are born into families that raise them as vegetarians, is that so dissimilar from being born with a certain skin tone?
I don't see what's wrong about the note. I don't see anything suggesting the note is inaccurate. I don't see anything that says the court forced Reuters to add the note. It appears Reuters added the note of their own free will.
Confirmed. Also, ClickBench is working from a batch-loaded data set, which is kind of antithetical to a real-world, real-time analytical database workload.
[Disclosure: I work at StarTree, and we're powered by Apache Pinot.]
We are currently considering / evaluating different methodologies to benchmark more realistic situations for real-time analytics. Potential consideration for your own benchmarking / POCs, or for a future industry benchmark spec:
1. Some sort of "freshness" (data latency) measurement: time for streaming ingestion / indexing / data ready for query. Is it consistent, or are there pauses in ingestion?
2. Some sort of "ingestion scaling" measurement: how many objects per second can you get to before you choke IO? What happens to ingested objects at different payload sizes? (This interacts with "freshness" above; i.e., you might [or might not] be able to throttle ingestion to improve freshness.)
3. Query concurrency/throughput: does your query capacity scale linearly or non-linearly? What happens at 10 QPS? 100 QPS? 1000 QPS? 10000 QPS? 100000 QPS? (Or when does it top out?)
4. Data volume: Are you querying against 1TB? 10TB? 100TB? 1 PB? (More?) This interacts with query concurrency. Because driving 100 QPS against 1 PB is a totally different case than driving 100000 QPS against 1 TB.
5. Data storage type: Are you running against local NVMe, EBS, or S3 buckets? (Maybe even HDD?) Is it all uniform storage, or is it in a tiered storage topology? If tiered, what's the % mix of the different storage types? This is just an increasing reality all vendors need to deal with. Customers want to optimize their spend per use case.
6. Query complexity: Before talking simple "latencies," you have to understand what sort of queries you're running. These aren't simple atomic row CRUD operations like a Redis or a ScyllaDB. How are you doing aggregates? Are you running queries against denormalized data in a single table, or are you doing single JOINs or multiple table complex JOINs?
7. Indexing: As pointed out by shadow28, indexes are vital for best performance. Which type of index was used? (Apache Pinot supports about a dozen different types).
And my personal favorite to throw into the mix:
8. Cluster resilience: Great. All of the above worked on a fully-provisioned stable cluster. Now knock out a node. Do it. See what happens. How long before the cluster rebalances and quiesces? What happens to your QPS and latencies during the rebalance, and then after quiescence? Measure that. Now knock out a 2nd node. Maybe a third. How many nodes can you bring down before performance goes non-linear, or the cluster is rendered as utterly unreliable.
This latter I call the "Torpedo test;" and I've been preaching about it for years[1]. How many "torpedos" can you cluster take before it sinks under the waves. It's not specific to real-time OLAP. You can use this kind of methodology to test resilience of any distributed system. And you should probably do this before you hit production.
"Backend" is absolutely not a simpler problem to solve, it's an umbrella term encompassing a wide variety of domains. If you're talking specifically about Web backends, then sure, maybe there's a case to be made there.
According to Wikipedia, Yugabyte (the company) has taken 290 million dollars of VC money. It's probably a safe assumption that they will follow the same path soon enough.
