Newsletter Subject

Coding Horror

From

codinghorror.com

Email Address

noreply+feedproxy@google.com

Sent On

Sun, Feb 17, 2019 01:54 PM

Email Preheader Text

--------------------------------------------------------------- Posted: 16 Feb 2019 06:15 PM PST Whe

[Coding Horror]( --------------------------------------------------------------- [The Cloud Is Just Someone Else's Computer]( Posted: 16 Feb 2019 06:15 PM PST When we started [Discourse]() in 2013, our server requirements were high: - 1GB RAM - modern, fast dual core CPU - speedy solid state drive with 20G+ I'm not talking about a cheapo shared cpanel server, either, I mean a dedicated virtual private server with those specifications. We were OK with that, because we were [building in Ruby]( for the next decade of the Internet. I predicted early on that the cost of renting a VPS with those specs would drop to $5 per month, and courtesy of Digital Ocean that indeed [happened in January 2018](. The cloud got cheaper, and faster. Not really a surprise, since [the price of hardware trends to zero over time](. But it's still the cloud, and that means it isn't exactly cheap, because it is, after all, someone else's computer that you pay for the privilege of renting. [there-is-no-cloud] But wait … what if you could put your own computer "in the cloud"? Wouldn't that be the best of both worlds? Reliable connectivity, plus a nice low monthly price for extremely fast hardware? If this sounds crazy, it shouldn't – Mac users have been doing this [for years now](. [mac-colocation-2019-1] I suppose it's understandable that Mac users would be on the cutting edge here since Apple [barely makes server hardware](, whereas the PC world has always been [the literal de-facto standard for server hardware](. [mac-stadium-colocation] Given the prevalence and maturity of cloud providers, it's even a little controversial these days to [colocate actual servers](, but we've also experimented with [colocating mini-pcs]( in various hosting roles. I'm still curious why there isn't more of a cottage industry for colocating mini PCs. Because … I think there should be. I originally wrote about [the scooter computers]( we added to our Discourse infrastructure in 2016, plus my own colocation experiment that ran concurrently. Over the last three years of both experiments, I've concluded that these little boxes are plenty reliable, with one role specific caveat that I'll explain in the comments. I remain an unabashed fan of mini-PC colocation. I like it so much I put together a new 2019 iteration: 2017 — $670 2019 — $820 [i7-7500u]( 2.7-3.5 Ghz, 2c / 4t [i7-8750h]( 2.2-4.1 Ghz, 6c / 12t 16GB DDR3 RAM 32GB DDR4 RAM 500GB SATA SSD 500GB NVMe SSD This year's iteration of the scooter computer offers 3× the cores, 2× the memory, and 3× faster drive. It is, as the kids say … an absolute unit. 😱 [2019-scooter-computer-top-interior-1] [2019-scooter-computer-bottom-interior] [2019-scooter-computer-front-and-back] It also has a rather elegant dual-sided internal layout. There is a slot for an old-school 2.5" drive, plus built in wi-fi, but you won't see it in these pictures because I physically removed both. I vetted each box via my [recommended burn in and stability testing]( and they all passed with flying colors, though I did have to RMA one set of bodgy RAM sticks in the process. The benchmarks tell the story, as compared to the average Digital Ocean droplet: Per-core performance sysbench cpu --cpu-max-prime=20000 run DO Droplet 2,988 2017 Mini-PC 4,800 2019 Mini-PC 5,671 Multi-core performance sysbench cpu --cpu-max-prime=40000 --num-threads=8 run DO Droplet 2,200 2017 Mini-PC 5,588 2019 Mini-PC 14,604 Disk performance dd bs=1M count=512 if=/dev/zero of=test conv=fdatasync hdparm -Tt /dev/sda DO Droplet 701 / 8818 / 471 MB/sec 2017 Mini-PC 444 / 12564 / 505 MB/sec 2019 Mini-PC 1200 / 17919 / 3115 MB/sec Discourse rebuild time ./launcher rebuild app DO Droplet 6:59 2017 Mini-PC 3:41 2019 Mini-PC 3:24 Power consumption could be a concern, as the 2017 version had a much lower 15 watt TDP, compared to the 45 watts of this version. That 3× increase in core count ain't free! So I tested that, too, with a combination of i7z, stress, and my handy dandy [watt meter](. [2019-mini-pc-i7z-testing] (idle login) 800 Mhz 10w stress --cpu 1 4.1 GHz 30w stress --cpu 2 4.1 GHz 42w stress --cpu 3 4.0 GHz 53w stress --cpu 4 3.9 GHz 65w stress --cpu 5 3.7 GHz 65w stress --cpu 6 3.5 GHz 65w stress --cpu 12 3.3 Ghz 65w I'd expect around 10 - 20 watts doing typical low-load stuff that isn't super CPU intensive. Note that running current-ish versions of mprime jacks power consumption up to 75w 🔥 and the overall clock scales down to 3.1 Ghz … let me tell you, I've learned to be [very, very afraid of AVX2 extensions](. (If you're worried about noise, don't be. This active cooling solution is clearly overkill for a 65w load, because it barely spun up at all even under full core load. It was extremely quiet.) So we're happy that this machine is a slammin' deal for $820, it's super fast, and plenty reliable. But how about colocation costs? My colocation provider is [EndOffice]( out of Boston, and they offer very competitive rates, at $29/month for colocating a Mini-PC. [endoffice-mini-pc-colocation]( I personally colocate three Mini-PCs for redundancy and just-in-case; there are discounts for colocating more than one. Here they are racked up and in action. Of course I labelled the front and rear before shipping because that's how I roll. [endoffice-colocated-2019-mini-pcs] Let's break this down and see what the actual costs of colocating a Mini-PC are versus the cloud. Let's assume a useful life of say, three years? Given the plateauing of CPU speeds, I think five years is more realistic, but let's use the more conservative number to be safe. - $880 mini-pc 32GB RAM, 6 CPUs, 500GB SSD - $120 taxes / shipping / misc - $29 × 12 × 3 = $1,044 That's $2,044 for three years of hosting. How can we do on Digital Ocean? Per [their current pricing page](: - 32GB RAM, 8 vCPUs, 640GB SSD - $160/month - $160 × 12 × 3 = $5,760 This isn't quite apples to apples, as we are getting an extra 140GB of disk and 2 bonus CPUs, but let's assume the CPUs are both slower and partially consumed by multi-tenancy compared to our brand new dedicated, isolated CPUs. (I was curious about this, so I just spun up a new $160/month DO instance for a quick test. The sysbench results are 4086 and 11760 respectively, considerably below the 2019 Mini-PC results, above.) Still, you pay almost three times as much for a cloud server. 🤑 I'm not saying this is for everyone. If you just need to spin up a quick server or two for testing and experimentation, there's absolutely no way you need to go to the trouble and up-front cost of building and then racking colocated mini-pcs. There's no denying that spinning servers up in the cloud offers unparalleled flexibility and redundancy. But if you do have need for dedicated computing resources over a period of years, then building your own small personal cloud, with machines you actually own, is not only one third the cost but also … kinda cool? [your-own-personal-cloud] If you'd also like to embark upon this project, you can get [the same Partaker B19 box I did for $490 from Amazon]( or [$460 direct from China via AliExpress](. Add memory and drive to taste, build it up, then check out [endoffice.com]( who I can enthusiastically recommend for colocation, or the colocation provider of your choice. Get something cool hosted out there; let's do our part to keep the internet [fun and weird!]( You are subscribed to email updates from [Coding Horror](. To stop receiving these emails, you may [unsubscribe now](. Email delivery powered by Google Google, 1600 Amphitheatre Parkway, Mountain View, CA 94043, United States

Marketing emails from codinghorror.com

View More
Sent On

20/04/2020

Sent On

12/09/2019

Sent On

20/08/2019

Sent On

30/05/2019

Sent On

22/10/2018

Sent On

21/09/2018

Email Content Statistics

Subscribe Now

Subject Line Length

Data shows that subject lines with 6 to 10 words generated 21 percent higher open rate.

Subscribe Now

Average in this category

Subscribe Now

Number of Words

The more words in the content, the more time the user will need to spend reading. Get straight to the point with catchy short phrases and interesting photos and graphics.

Subscribe Now

Average in this category

Subscribe Now

Number of Images

More images or large images might cause the email to load slower. Aim for a balance of words and images.

Subscribe Now

Average in this category

Subscribe Now

Time to Read

Longer reading time requires more attention and patience from users. Aim for short phrases and catchy keywords.

Subscribe Now

Average in this category

Subscribe Now

Predicted open rate

Subscribe Now

Spam Score

Spam score is determined by a large number of checks performed on the content of the email. For the best delivery results, it is advised to lower your spam score as much as possible.

Subscribe Now

Flesch reading score

Flesch reading score measures how complex a text is. The lower the score, the more difficult the text is to read. The Flesch readability score uses the average length of your sentences (measured by the number of words) and the average number of syllables per word in an equation to calculate the reading ease. Text with a very high Flesch reading ease score (about 100) is straightforward and easy to read, with short sentences and no words of more than two syllables. Usually, a reading ease score of 60-70 is considered acceptable/normal for web copy.

Subscribe Now

Technologies

What powers this email? Every email we receive is parsed to determine the sending ESP and any additional email technologies used.

Subscribe Now

Email Size (not include images)

Font Used

No. Font Name
Subscribe Now

Copyright © 2019–2025 SimilarMail.