It's so frustrating to talk to Kubernetes enthusiasts.
enthusiast: if you'll migrate service X from a small sets of VMs to K8s cluster (which will take N man-moths because of reasons) it will auto-scale
Old grumpy man: but the the load is low and predictable we don't need autoscale and if load will grow we will just create two more VMs
enthusiast: auto-scaling will save time in the distant and unlikely future and CTO agrees that K8s is the best way to run software so you have to migrate anyway
We are running a busy self service ERP for growing 400k employees, on a single database instance for 7 years now. Currently at 256gb ram, 8tb database, 32 P9 cores. I got 99 problems but the single database server ain't one of them!
Parent was probably talking about the application server, which is the typical example of K8s usage, whereas most databases don't fit well there and are usually separate from the K8s cluster.
Yeah. They must be. I’ve had several k8s enthusiast dissuade from putting a db in k8s for performance reasons. This may change overtime but I believe there is a performance bottleneck in disk I/o (or at least that’s how it was explained to me)
That really depends on your k8s environment. If you're running on bare metal and use host bind mounts for storage, you won't have a problem. But as soon as you introduce shared storage it gets messy...
We've just migrated our ERP from a regular deployment on VMs into Kubernetes. We have a very bursty self-service component, so being able to scale up easily for the week we need it is quite nice. We've taken advantage of this ability multiple times and seen great success each time.
Used to run Tivoli Storage Manager, now Commvault. Backup can be used for certain type of issues that are isolated to the database.
For DR, Entire stack is being continually synced with a set of hot server in separate datacentre. We run a fully DR exercise yearly and it works well. It's a fairly proprietary AIX / DB2 method though so not sure you'd find value unless you're using the exact same technology stack :-/
Quote them some Henry David Thoreau[1]: "Why should we live with such hurry and waste of life? We are determined to be starved before we are hungry. Men say that a stitch in time saves nine, and so they take a thousand stitches today to save nine tomorrow. As for work, we haven't any of any consequence."
The irony is the last time I fell for this "SRE/K8s/Platform Engineering" crap, they promised auto-scaling and used it as a motivation for the tech. Yet when the time came, and services inevitably needed to scale due to load, they had to manually log in to increase some random config to let it grow.
> Old grumpy man: but the the load is low and predictable we don't need autoscale and if load will grow we will just create two more VMs
That's perfectly ok, if your deployment costs are irrelevant and/or your company gladly pays up your infrastructure costs without a second thought.
This is not the case in some organizations, and toggling a setting to auto scale a deployment can automatically save you thousands of dollars per month.
Would you still be so casual about infrastructure costs if you had to bankroll the extra capacity you need to add to your baseline to support peaks?
The decision process involved in managing your single-box deployment is not the same that goes in managing global deployments with dozens of instances per region. Cloud providers charge a premium, and that premium is a lot.
It's like the thermostat in your office. If you're just running an AC in a single room then you can just set it to full blast to keep it day and night at a certain temperature. Once there's a decision to cut costs then you start to talk about the best time to turn off/turn on a AC unit.
VM deployment has a drawback though. If the machine runs out of memory, it will lock up (at least that's my experience with EC2). And then you need to set up some sort of health checks and auto scaling (!) for EC2 based on those health checks. This is much more cumbersome and fragile than just containerising the app and setting it up in a k8s cluster which will automatically take care of this scenario.
Also, deploying a new build on VMs is extremely manual compared to k8s, unless again you set up some sort of home brew rube Goldberg machine to auto deploy. It's just way better to use k8s in tandem with a simple GA workflow.
I think grumpy old man knows about the drawbacks with VMs, but grumpy old man also knows that for his particular service k8s has drawbacks. Grumpy old man is thinking that if his particular service goes down, some users just tell him it's not working and it's not a huge problem like people are losing $10 million per hour because of it. And in the event that happens sometime in the next 23 months (if it happens at all) he'll just do that little manual process for a couple more VMs, and that's all the time he'll spend on it. Contrast that with the time it would take to get it ready for k8s and keep someone around who understands k8s well enough to take care of it, even though their need for it might be so low.
And grumpy old man also knows that compared to the 5-year overhead costs of some overcomplicated k8s system, it's quite reasonable to overprovision the bejesus at the colo bare metal>hypervisor layer as cheap insurance.
Every time I read something like this I think that there has to be a tipping point were a well-written (i.e. compiled, not interpreted) solution running on a beefy server has to more cost efficient than running the same thing spread over several containers.
Definitely not an expert, but I get the impression that this point is a lot higher that most people assume it is.
Spoiler alert: a well written application on a beefy server almost always beat the k8s rube Goldberg machines on cost. I still don't get why people refuse to think for themselves and just jump on the latest hype train.
K8s makes little sense unless you run on bare-metal. Once you jump to vms you are injecting another abstraction layer and take on a herculean level of ops without understanding what you're getting into.
VMs are nice when you can't fill a machine with a single task (plus whatever redundancy). Once you get to a single machine, you want to scale up that one machine; you can go a long way where scaling up is cheaper than scaling out. But at some point, scaling out gets cheaper.
I'm not sure where the check points are now, but typical points where cost jumps are desktop -> server socket, single socket -> two sockets, two sockets -> four sockets, four -> eight sockets. AFAIK, AMD EPYC isn't offered at more than two sockets, and going to four sockets used to be possible off the shelf but very expensive, and eight sockets was very expensive if off the shelf or very expensive because custom engineering. Sometimes ram costs go way up for the highest density too.
Men may be have a crisis of identity but women still expect their husbands to be the main breadwinners and suffer from the mismatch of expectation and reality too. Women avoid marrying men who earn less and have no potential to earn more.
As a result I'd expect a lot of single men with below average income and a lot of single women with above average income who struggle to find worthy men. Statistically it is impossible meet two condition at once: 1. women on average earn at leas as much as men 2. In a family a husband has income higher than a wife.
I've seen an insightful explanation (from an immigrant from Ukraine) why it doesn't matter is there is a problem with the Russian culture or not but it should be cancelled in any case: to win this war Ukrainian people need to mobilize and hatred to anything Russian will help to mobilize them. It is not enough for them to hate Putin, to fight they need to hate all Russians. For this reason it is important to convince all that all Russians are imperialists and Russian culture is all poisoned by imperialism. And it doesn't matter if it would backfire - in an authoritarian Russia it doesn't matter what ordinary people think. Ukraine on other hand is more or less democratic state - it is for the people to decide if they would accept defeat or will continue to fight.
As someone from a middle class soviet family born in early 1980s in a middle-size town the first time I've seen a western computer at home was only after USSR collapse (likely around 1995). It probably was a PC with i486 (or Cyrix clone of i486). During my school years (90s) only two or three my classmates had a computer at home and it were some Z80 clones produced either in the USSR/exUSSR or in the Eastern Block (in 90s PC were already available, but not affordable).
I know some soviet companies were able to procure western computers even in 1980s but for ordinary people it was out of reach until early 1990s or late 1980s at best (at least outside Moscow).
Different people have different motivation: among people who criticize Israel there are people driven by antisemitism. Of course not all critics are antisemitic. The same with China - some can be right (in critics of CCP) for wrong reasons (xenophobia/nationalism). Or one can criticize specific policies for good reasons and be labeled as a racist.
Not stealing directly but US military industrial complex is a huge money sink which delivers much much less than civilian counterparts for the same money. F-35 story is a shame IMHO.
Corruption is a big issues in the west but compared to what's happening in Russia and ex-USSR countries, it's not even in the same ballpark, hell, it's not even in the same planet.
In the west we have free press looking into it and exposing the theft, plus, the theft in the west has very complex schemes to make it look legit on the surface and to make sure nothing sticks to our Teflon politicians, but in Russia they don't need to worry about any red tape BS, they can just shove their hands in the public purse and take as much as they want and go directly to the Lamborghini dealership or the casino without any accountability, and if any reporter looks too deeply into their theft, they get death threats or even death through very suspicious circumstances.
The problems on the Joint Strike Fighter program were caused more by flawed analysis of costs and military requirements than by outright corruption. Very little funding was stolen, and not many individuals have gotten personally wealthy from the program. Lockheed-Martin is now delivering an F-35 Block 4 product that, despite a few remaining minor flaws, substantially meets the customers' requirements. It's just that what the customers thought they wanted back in 1996 no longer aligns with what they actually need today.
1. Russian military performs worse than was expected by western experts but still has enough explosives to flatten out large swaths of land - as we see they don't care at all about collateral damage or outright target civilians as an easy prey.
2. NATO military was not really tested in full scale combat and can perform worse than expected too.
This is not really a poorly performing military. This is a terribly incompetent plan proudly made by only 5 people.
NATO has had plenty of logistical failures but it doesn't narrow it's planning to total top-down so those failures are dealt with and worked around by elements in the middle and on the field.
There was some corruption in Soviet Union (like in most countries around the world with the exception of economically developed or/and democratic countries) but it was not fatal.
Corruption in modern Russia is unique to Putin's regime - he used it (and still uses) as a tool to concentrate power - he puts in position of power only corrupt officials as a mean to get full loyalty - officials not loyal enough can be easily blackmailed, secret services have enough "kompromat" to put any government official into jail but are waiting for a signal from above. Corruption is the foundation for what in Russia called the "vertical of power", which essentially means that a person in a power structure have almost unlimited (limited only by Putin, but not by the law) power over people below.
> There was some corruption in Soviet Union ... but if[sic] was not fatal
It was actually fatal. I've studied the corruption that emerged under the Soviets; it was systemic and, in fact, necessary to mitigate the crippling adversities created by Soviet ideology. It is the primary reason the Soviet Union collapsed 30 years ago.
I was born in the Soviet Union, read book about Soviet economy and political system (albeit not scientific ones) and have a different take. Corruption is a broad term and one can probably say it was a contribution factor, but may take is that inefficiency of economy and misguided foreign policy (including Afghanistan invasion) were the fatal flaws.
When selecting who to delegate power to, all soviet leaders selected people who are loyal to them and suitable for the job. Loyalty was more important than competence so government was full of incompetent people, but competence was not a disqualifying factor so some of them was able to do the job.
In Putin's power structure people who not steal and cannot be easily blackmailed are not promoted high enough. Being not corrupt _is_ a disqualifying factor. Putin at least in the beginning of his career maintained visibility of a country with the rule of law, so instead jailing opponents on a whim he jailed them for breaking the law, but the law was enforced selectively only for ones who are not fully loyal.
As a result in modern Russia personal enrichment is a much bigger problem that it was in Soviet Union. Soviet nomenklatura [1] was rich compare to poor population but their quality of life was not much better than for upper-middle class in the US. The same nor the case for modern Russia.
Failure of soviet economy as I know was not result of corruption - much more resources were wasted than stolen. Unfortunately I cannot provide any sources here - all I read was in Russian and I cannot find translations to English.
I've seen this inefficiency myself in form of many thousands tons of trees which were cut so harvesting organization would meet KPI for amount of harvested wood but left to rot in the forest because an organization which should transport the wood had not enough capacity to do this (but likely transported all volume on paper, which you may say is corruption, but resources were wasted, not stolen).
Dude, the countries behind Iron Curtain had so much corruption it was insane. In a place where you have trouble buying meat or fruits the only way to get it is to bribe someone and for that someone to lie and/or steal to get it for you.
My country was behind an iron curtain and one of the motto's of that time was: "If one doesn't embelish from the state it means one's embelishing from their family".
Even with democracy this mentality won't fully go away for generations. Russia continued on that trajectory to a worse state.
enthusiast: if you'll migrate service X from a small sets of VMs to K8s cluster (which will take N man-moths because of reasons) it will auto-scale
Old grumpy man: but the the load is low and predictable we don't need autoscale and if load will grow we will just create two more VMs
enthusiast: auto-scaling will save time in the distant and unlikely future and CTO agrees that K8s is the best way to run software so you have to migrate anyway