Kernel 4.18 is _ancient_, weird to test on that. Also I wonder if the whole performance increase can't just be achieved by lowering the readahead size a bit. I see in their notes they have a decent SSD but set the blockdev readahead to 64. Depending on your workload it's much more performant to lower that to 16. As mentioned in TFA, fast storage is the most important factor here, and in my experience combining that with a smaller readahead pretty much fixes any read amplification issues you get with the larger readahead.
There are two performance problems presented here. One is the fact that the OS waits till a static memory bank water-level before triggering a cleanup.
The second is the one you pointed out; here again one thing to consider is cassandra's compression chunk size `chunk_length_in_kb`. readahead value less than chunk size makes cassandra slow in general. take a look at this https://thelastpickle.com/blog/2018/08/08/compression_perfor... for more info on it.
I get the feeing you're looking at this through the specific lens of a programmer. Terraform isn't made for programmers - you'll miss all the flexibility a real language gives you. It's made for ops people who deal with wrangling a whole bunch of different types of systems with different API's and languages and just need some way to standardize the management of disparate systems, whatever counts as "infrastructure".
The state file thing gets a relatively large part of the hate but it's that and the limitations of the DSL that make the DAG possible and useful. Pulumi and all the other wrappers don't solve this, though they can plausibly solve the "closer to programming" problem and I'm sure that has a valid audience.
I guess what I'm saying is, I think it'll stick around and we will in fact settle on it for a large part of operational work. I'll add that I also think k8s should die a quiet death and _that_ will be seen in retrospect as a necessary step to something better.
Personally I just push to an offsite restic backup on a cheap Hetzner storage box. It's not super fast and only hosted in EU AFAICT but it fits the bill for being cheap and reliable enough. It's just there for me to retrieve sentimental-value things like photos in the off chance all my other backup methods have failed.
Initial setup is a bit weird with their SSH key setup IIRC but manageable.
As others have noted, PowerToys (https://en.wikipedia.org/wiki/Microsoft_PowerToys) has a long history. I have some fond memories of the TweakUI utility by Raymond Chen as it made the OS much more usable.
The original is "Krigsskipet gjør først en unnamanøver i siste liten". Now, I don't speak Norwegian but I'm guessing the word "unnamanøver" is pretty versatile considering the translations you get when making minor changes.
Krigsskipet = war ship (german-ish)
gjør = verb to do
først = first
unnamanøver = evasive action
(in other context apparently "nightmare" or "kidney robbery"?)
i siste liten = at the last minute
It's a short sentence so it seems it guessed the wrong context. Just looking at the word though makes one think of sort some of "maneuver".
I'm guessing the word "unnamanøver" is pretty versatile considering the translations you get when making minor changes.
That is pretty bizzare. I cannot think of any context or situation where "unnamanøver" could possibly mean anything other than evasive maneuver, nor do the Norwegian words for either kidney or nightmare look or sound anything even vaguely close to "unnamanøver"
"Krigsskipet gjør en unnamanøver" translates to "The war ship is doing nightmare". "først en unnamanøver" to "first a young man", "unnamanøver i siste liten" is "last minute kidnap".
While it's a relatively young product, I've been running InfluxDB at scale for over a year in production and it's been a joy to use. Hundreds of hosts reporting stats every 10 seconds and the query language is straightforward and powerful once you grok the data structures a bit. It fits the bill, I think.
+1 for InfluxDB and the TICK stack. We evaluated it for our monitoring needs, decided not to go with it but were damned impressed with how easy it was to setup and use, the features (including the query language) and the overall quality of the stack.
Less impressed with the Influx org and their SaaS, but I definitely want to find a usecase for getting stuck into Influx for time series collection
Can you describe your workload and your experience with InfluxDB? We evaluated InfluxDB in 2015 and found that it fell apart due to the large amount of data we produce. I'd be interested in hearing your experiences to see if we need to reevaluate.
Workload is mostly writing incoming metrics from many dynamic instances (AWS). In 2015 Influx was at pre-1.0 and definitely had a tendency to "fall apart" under stress - for one, there were no memory limitations implemented, so an accidental large query would eat up all your memory and kill the database. To be fair, no one recommended putting it in production at that point. It's just that I needed something to run a Graphite-like metrics platform at scale and it had to be future-proof. Bit of a gamble but it turned out well! :)
I'd like to second this, while also noting that a side-effect of this noodling-with-languages is that my fully English OS and browser is getting a page with some content (full paragraphs!) in Dutch, while other content is in English. On the same page. An interesting bug, but not one that inspires confidence.
Kind of amazing that the state of the art in this area hasn't changed in nearly 10 years. The marketing angle is funny as well - everything has to be a cloud now. I think I used to call what we built at Optiver a "private cloud" but "Metal Cloud" is nicely buzzwordy as well.
PS - Hi Marty, unknown nick here but I'm sure you can figure out who I am. :D
> You don't hear the former at all; but you hear the latter as "plus plus," don't you?
I really do not. A short silence, both of them. I suppose it depends on how you normally code. I have never had a need for speaking out code and am not well versed in it.
I experience reading similarly to you. I will often become familiar with written words before I know how to pronounce them. When I want to say the word for the first time, I will have to pause and think about how it would sound out loud.
This is especially true of code and symbols. For example, consider this poem
> is it already too dark
> to play tennis with a racket
> i asked?
> while I code with [
and contrast with this one
> The house filled with laughter
> from mother and daughter
> Both were fiends
> but neither friends
The "it works after traceroute" line was the moment I was sure it was ARP caching. Then I was just wondering if it had anything to do with AWS specifically, which could have pointed towards a serious VPC bug. That doesn't seem to be the case though. You're not weird. Just burnt by real experience.
reply