1,208 karma · joined October 7, 2009
My team did an integration with Kodak-Alaris a few years back and we toured their main office in Rochester.
I've never heard of seals doing that kind of training.
[1] https://faseb.onlinelibrary.wiley.com/doi/abs/10.1096/fj.201...
Yeah, but the title is made up. There is noting in the official report about this being "nationwide".
The report only states that there is a metar outage consisting of 167 stations.
It doesn't provide any context on which stations are missing or how they are distributed.
Adding "nationwide" to the title is pure FUD IMO.
The report is that 167 are missing.
That would be roughly 5% of US stations or 0.2% of worldwide stations.
I was able to look up meters for airports that I'm familiar with. https://www.aviationweather.gov/metar
So, it is real but far from a total collapse of the system.
I would say something like: "I think a balanced tree, such as an rb-tree, would be useful here for <reasons that make sense given the problem and the properties of rb trees>. I've written rb trees before and think I could write a basic one in 10-15 minutes or I could use <class from the std library, which uses a balanced tree>. Which would you prefer?"
Assuming what you said made sense I would take an interaction like that as a positive signal.
There is a system for using git* in conjunction with the monorepo that some developers use, but that system has nothing to do with the prodkernel.
* there is also a similar system for using mecurial with the monorepo that some folks love.
'Please don't comment on whether someone read an article. "Did you even read the article? It mentions that" can be shortened to "The article mentions that."'
It doesn't add anything to the discussion. And in this case it was very clear that the OP read the article and just missed that part.
Your comment would have been just as informative and less snarky if it was simply:
There's a small part about whiskers that already convey enough meaning:
> UNACCEPTABLE WHISKER
> A whisker is a slender needle-shaped metallic growth between a conductor and a land. Typically the result of mechanical stresses in high tin compounds, it is a reliability concern."
We have teams at Google whose sole purpose is to ensure that data that should be deleted is deleted in a timely manner (as defined by TOS and the law).
Basically, every piece of user data that is stored at Google has to have a documented and audited retention plan that covers how it is stored, how long it is retained, how it can be used, etc. The retention plans are reviewed by legal counsel and are audited by independent teams.
From a technical standpoint, Photos doesn't deal directly with individual machines or hard drives. There are several layers of abstraction to map what we think of as a "file" and the storage system are doing something that is _effectively_ at least a defrag on a regular basis.
[I work on Google Photos and used to work on storage infrastructure]
I work on Google Photos and prior to that I worked on datacenter level infrastructure.
"And, as always, we uphold our commitment to not use information in Google Photos for advertising purposes"
https://blog.google/technology/safety-security/keeping-priva...
<I've worked on Google Photos since before it was launched>
<disclaimer - I work on Google Photos>
It will show things like your largest photos and low quality (blurry, dark, etc) photos that you can review and decide what to delete.
At 40,000 feet, time of useful consciousness is 15-20 seconds.
See https://expertaviator.com/2012/04/19/oxygen-requirements-tim...
Basically the partial pressure of oxygen is low enough at altitude that the oxygen is driven out of your blood instead of in.
There are some breathing techniques that can offset that somewhat, but only if you are prepared and trained, which is unlikely in a sudden decompression event.
https://www.who.int/ith/mode_of_travel/tcd_aircraft/en/
"Research has shown that there is very little risk of any communicable disease being transmitted on board an aircraft.
The quality of aircraft cabin air is carefully controlled. Ventilation rates provide a total change of air 20–30 times per hour. Most modern aircraft have recirculation systems, which recycle up to 50% of cabin air. The recirculated air is usually passed through HEPA (high-efficiency particulate air) filters, of the type used in hospital operating theatres and intensive care units, which trap dust particles, bacteria, fungi and viruses."
'Please don't comment on whether someone read an article. "Did you even read the article? It mentions that" can be shortened to "The article mentions that."'
'Be kind. Don't be snarky. Comments should get more thoughtful and substantive, not less, as a topic gets more divisive.'
Nope, tomatoes are technically fruits, though they are commonly considered vegetables.
Just like pterosaurs are technically not dinosaurs, but are commonly considered to be in our culture.
I used to live less then a mile from one. Pretty solid pizza.
FWIW, it always had the feel of a local place. I was surprised that they had 20 locations and used consultants.
From what I can see of the screen shots it looks like at least the core of that software is still in use.
The most surprising thing to me was the correlation between age and RBER.
I would not have guessed that normal aging would play a significant role in reliability. It would be fun to understand what is happening there.
From the paper:
We find that UBER (uncorrectable bit error rate), the
standard metric to measure uncorrectable errors, is not
very meaningful. We see no correlation between UEs
and number of reads, so normalizing uncorrectable errors
by the number of bits read will artificially inflate the
reported error rate for drives with low read count."
They don't differentiate UBER from a manufacturer spec and UBER as measured in production.
For any given type of flash chip the manufacturer will provide a spec like "you must be able to correct N bits per M bytes". The firmware for a flash drive must use forward error correction codes (e.g. BCH or LDPC) of sufficient strength to correct the specified number of bit errors.
Dealing with a certain amount of bit-errors is just part of dealing with flash.
For example, a chip could have a spec that you must be able to correct up to 8 bit errors per 512 bytes (made up numbers). If the chip had 4KiB pages, each page would be split into 8 "chunks" that were each protected by error correcting codes that were capable of correcting up to 8 single-bit errors in that chunk. As long as no "chunk" had more than 8 bit errors, the read would succeed.
So in this case you could theoretically have a page read with 64 bit errors that succeeded.
This is alluded to in the paper: "The first generation of drives report accurate counts for the number of bits read, but for each page, consisting of 16 data chunks, only report the number of corrupted bits in the data chunk that had the largest number of corrupted bits."