Stuff The Internet Says On Scalability For Sep 18th, 2020

http://highscalability.com/blog/2020/9/18/stuff-the-internet-says-on-scalability-for-sep-18th-2020.html

~60% : of organizations run a mix of SQL and NoSQL databases. Only 14% of organizations run exclusively NoSQL databases.

0.81% : BackBlaze’s Annualized Failure Rate (AFR) for Q2 2020. Q1 2020 which was 1.07% One year ago (Q2 2019), the quarterly AFR was 1.8%. Three drive models had 0 drive failures: the Toshiba 4TB, the Seagate 6TB and the HGST 8TB.

5% : of world’s websites hosted on Wix. 700 million uniques per month. 

$734.38 : Joe Emison’s itemized monthly bill for a full-stack insurance company running on serverless.

Interesting, but I get a distinct pro-serverless vibe from this website. I’m curious about specifics here, especially volume.

£200m : British companies paid in ransomware last year.

@TimSweeneyEpic : Two facts about Apple: 1) Apple is #3 in the world in game revenue 2) Apple doesn’t make games

Oof. I wonder how Steam compares.

45nshukla : Moving 25TB data from one S3 bucket to another took 7 engineers, 4 parallel sessions each and 2 full days

@Rainmaker1973 : NASA only uses 15 digits of π for calculating interplanetary travel. At 40 digits, you could calculate the circumference of a circle the size of the visible universe with an accuracy that’d fall off by less than the diameter of a single hydrogen atom

@_ericelliott : 2 months into TDD: Tests are hard to write & brittle. 2 years in: Tests taught me better code patterns, reduce bugs 40%-80% and eliminate fear of change. 10 years in: TDD changed my life.

I agree with this for testing in general, but for TDD specifically I’d reverse the order.

rkangel : We mostly use C rather than C++, but the same two big reasons get in the way of using Rust for everything:* Compiler support * Availability of suitable engineers

Compiler support for?

@nathankpeck : My DynamoDB usage last quarter: - 1 TB of data, >10 billion rows - 1 billion on demand read units per month - 200 million on demand write units per month - Consistent 2.5ms query latency the entire time - Cost is about $6k per month. Having such a worry free DB is priceless

From Twitter:

It is a demo app I’m building, like a light Twitter with a bunch of bots acting like users. All on demand because I wanted to test on demand at scale to see how the cost characteristics are if you go fully on-demand for a largish workload. Still quite affordable imo

the term “latency” has become regurgitated ad nauseam. In performance analysis it is a useless generic word. Successful analysis requires that it immediately be decomposed into service time, waiting time, etc. 

Snowflake, which offers cloud-based data warehousing software, is worth about $70 billion. Teradata, which sells traditional data warehousing hardware and software, is worth $2.5 billion

Bruce Dawson : If everyone on a project spends all of their time heads-down working on the features and known bugs then there are probably some easy bugs hiding in plain sight. Take some time to look through the logs, clean up compiler warnings (although, really, if you have compiler warnings you need to rethink your life choices), and spend a few minutes running a profiler. Extra points if you add custom logging

Great insight!

RustConf 2020 videos are now available .

🔥🔥🔥

Edit