sandfox

The banks have nothing to do with the cost of commercial rent around old street roundabout and the shoreditch triangle. It's expensive because all the silly startups now think they need to be here to sound credible / look cool / etc etc so landlords can charge these prices and people will pay. Infrastructure wise, the area is a mess. Fibre coverage is spotty at best. ADSL seems to be pretty crap. Most buildings are poorly converted from storage/etc into office space. On the plus side we have nice places to eat, drink and get coffee so it's not all bad I suppose. Also the commute is pretty good for east london folk.

Also side rant. To all non-londerners, Tech City / Silicon Roundabout is an abstract concept that doesn't actually exist in any concrete form. Unless someone told you, you wouldn't even notice you were passing through it. Infact the only remotely "tech looking" building is satellite company Immarsat's HQ office, and IIRC they were around long before "tech city" had been dreamed up.

Writing code/recipes/etc for chef/ansible/puppet is no different to writing code for an application. Just because you think you have dictated (in code) that things should be a certain way does not mean that what will be executed or run.

nagios is awful :-), active checks are ultimately a dead end. Heartbeat style stuff is where it's at. You should definately check out sensu and riemann, (or hevean forbid if you really like node.js everything Godot)

http://sensuapp.org/ http://riemann.io/ https://github.com/nodejitsu/godot

"mission critical"...

because PHP is "slow" and c is "fast", just about anything thats written in PHP such as FastCGI parser would be daftly slow compared to a counterpart written in c or c++. Most people will not be prepared to sacrifice perfomance just for the sake having everything in PHP

You could (and many) do the same thing with git.

Hadoop cluster of of GPU + SSD'd up machines is not going to be cheap.... (but would be fun)

Why is your logging library handling log rotation? Also, not everyone logs to disk straight from process or at all.

I know microbenchmarks are generally daft and horribly misapplied, but it might be worth showing/measuring the difference between using built-in procedural methods and your OO ones on some some example use cases to stop people speculating and just to make sure you aren't majorly shooting yourself in the foot perfomance wise for the benefit of a nice API.

All that aside, kudos!

They can (and often are), but some of the drones are capable of completely computer controlled take off and landing. I remember seeing somewhere that the human drone pilots caused more crashes when landing than the computer controlled ones.

Why shouldn't a 'devops' company complain about dependency? It's their domain completely. Treating anything as a "fact of life" is not going to make the problem go away... And what about anyone out there trying to use them who isn't a 'devops' thingy. A painful dependency graph is going to be something that influences their decision on what tool they choose.

Just because you have tools ease dealing with dependencies, it does not mean that those dependencies and the cost of dealing with them has gone away. Easier to not have the problem instead of solving it...

only 30 mins of data retention... even AWS gives you 2 weeks at 5min resolution for free (albeit only stuff the hypervisor can measure because they don't run an agent, so no RAM etc).

It should probably be noted that: a) you don't have use node's built in HTTP server (yeah, I know, nearly everyone will), you are more thn free to write your own or use one from it's module repository (npm) b) The entire HTTP module is currently being over-hauled in the 0.11.X development branch and the changes should appear in the stable 0.12.x releases.

Out of interest has anyone seen what other web servers support for these more 'esoteric' verbs is like?

well, they do give a million free push messages a month to start with which slightly skews it.

"1 Million FREE Push messages/month. $0.001 per additional Push message"[1]

[1] https://go.urbanairship.com/accounts/register/

I wonder how long till AWS provides some form of realtime websocket/whatever connection service. (I admit it will be some time before things are stable enough to make it sensible for AWS to get involved).

This marks as interesting expansion nonetheless. Probably need to work out what extra cost of this is like (pretty hard to compare though)

And this is why he has done this and you haven't :-p

Though seriously, when was HN ever about health and safety?

Errmmmm don't think anyone is offended by the word blitz, or ever has been. It's a pretty common term over here (UK) to describe a period of intense activity of any kind. Everyone from soldiers to school teachers use it.

e.g we need to blitz the sink - we need to clean it thoroughly and/or quickly.

Why not just plant lots of explosive devices WITH a trigger mechanism... make the numbers game work for you :-)

Pedant time! ssh is just one of many transports that git can use.

If anyone has ever bothered looking at any Netflix talks / slideshares [1] about APIs, they would have surely seen the various graphs detailing how their API usage started and grew.

TL:DR 3rd party usage of the APIs used to be big, but then became pretty much a statistical irrelevance compared to usage by Netflix's own software / tv apps / etc.

So it would seem the business case to keep developing a 3rd party friendly API is not really there anymore.

[1] http://www.slideshare.net/danieljacobson/techniques-for-scal...

This is fairly key. Alot of people think in a home-centric fashion when you mention remote work and mis the crux of remote work, that you are in fact working remotely with regards to the (main) office/hq/etc.

Define slow? Which particular aspect?

Google twelephone

For a potential reason why this might not make any difference to Amazons revenue/sales/whatever one should read this gem of a past by John Allspaw (Ops whiz kid for those who have no idea who he is) :: http://www.kitchensoap.com/2013/01/03/availability-nuance-as...

Yeah, if you terminate rather than stop the instance then everything that was created as part of the instance will be destroyed by default although you can choose to not kill attached EBS drives. Any EBS drives that were attached after instance creation will be left untouched by default after instance termination.

depends, TL;DR, local disk drives (Ephemeral) do not persist anything that wasn't part of the base image if you power down the machine (doesn't include restarts - data is persisted in this situation). network based drives (EBS) persist all their data across machine shutdown.

Slightly longer version - When you stop a AWS instance, the instance is torn down completely and all local state/data (Ephemeral) on the hypervisor is lost. when you then start the instance again a new hypervisor with spare capacity is found and then booted with your chosen base disk image. Restarts do not cause your instance to torn down so you don't lose local state / data. If your instance uses network drives (EBS) then these drives keep state when your server is torn down, and are re-attached when the server is started back up again,

Or generally possible directly with web requests. Pretty much everything is torn down at the end of each request. (Yeah PHP-FPM slightly muddies the waters)

It turns out that I don't know my left from my right.

Look at the top left of the white area of the page..