fendale

Are you saying that on IOS 18 if you enable developer mode then each time you forgot the network it gets a new Mac? But without developer mode it does not get a new Mac each time you forget it? The Apple docs linked elsewhere in this thread suggest it only gets a new Mac once per 24 hours when you forget the network normally. I’m going on a long boat trip in the next week where this trick might work for me if so!

Ah ok - so googling a bit on this, you do specify the size when creating the namespace. So if you have multiple namespaces, they appear as separate devices on the OS, and then you can mkfs and mount each as if its a different disk. Then you get the different IO queues at the hardware level, unlike with traditional partitioning.

local enterprise-grade SSDs support multiple namespaces (with their own internal queues)

What do you mean by namespaces here? Are they created by having different partitions or LVM volumes? As you mentioned consumer grade SSDs only have a single namespace, I am guessing this is something that needs some config when mounting the drive?

Over the last few months I have been seeing a push toward using Sqlite in production, where appropriate. Some of this has been coming from 37 Signals with their "Once" products, which use Rails and a SqliteDB.

Sqlite can go pretty far if you have fast SSD disk. The biggest problem is that your app is constrained to a single host. For many apps, with backups and a failover plan, that may be OK. For other its a non starter.

Where are you seeing these sort of roles? Are they labeled as "devops" or something else?

One reason we have so many poor engineering managers is because the top engineers get promoted to management. The skills for the two jobs are very different.

I’m sure many good engineers have tried management and found it to be not for them. Have an open discussion with your manager and see where it goes, as if you don’t you will likely end up applying for an engineering role elsewhere and leave anyway.

I was interest in a position at Aiven some time back. Did a recruiter screen and then a manager screen and then got asked to do a 20 hour take home assignment in Python that was to be completed in a single week. I politely declined at that point. Is that still part of the hiring pipeline?

I have a Shelly EM on my main grid circuit and solar array and find it to be an excellent product. They have lots of other automation products too.

I’m kind of similar. I write up how I set something up or some strange problem I solved mainly for my own reference but if it also helps someone else, great. Once or twice I’ve Googled a problem and landed on the solution in my own blog or a stack overflow post I made years ago!

For a job that needs to access 100's of thousands of small files, the ability to read the meta data quickly is very important.

This is the wider issue with small files. On HDFS each file uses up some namenode memory, but if there are jobs that need to touch 100k+ files (which I have seen plenty of), that puts a real strain on the Namenode too.

I have no experience with S3 to know how it would behave in terms of metadata queries for lots of small objects.

Apache Ozone https://hadoop.apache.org/ozone/ is an attempt to make a more scalable (for small files / metadata) HDFS compatible object store with a S3 interface. Solving the meta data problem in the HDFS namenode will probably never happen now. Too much of the Namenode code expects all the meta data to be in memory. Efforts to overcome the NN scalability have been around "read from standby", which offers impressive results.

The meta data is not the only problem with small files. Massive parallel jobs that need to read tiny files will always be slower than if the files were larger. The overhead of getting the metadata for the file, setting up a connection to do the read is quite large to read only a few 100kb or a few MB.

The other issue with the HDFS namenode, is that it has a single read/write lock to protect all the in memory data. Breaking that lock into a more fine grained set of locks would be a big win, but quite tricky at this stage.

The biggest part of the EU cost difference is sales tax which is 20% in the UK and only 8.5% in California I think.

I got my current job through LinkedIn too. Hadoop / big data position. I've got quite a few recruiter spam type messages though, but the one I got the job through was an in house recruiter for the company. I figured they were just doing key word searches as my profile isn't anything special. In my profile I put a good paragraph about each of my past positions highlighting experience with the technologies I wanted to work with. 5 or 6 months later this job came up.

I live in Belfast so I can either use a UK airport (Belfast), or Dublin. In my previous job, we flew business class from the UK to San Fran fairly regularly. It was often £500 - £1000 cheaper to fly Dublin to SFO connecting through London than Belfast -> London -> SFO. Infact, even pricing the London to SFO on the same flight (leaving out the Belfast leg), it was £500 cheaper to start in Dublin.

The price difference was all tax - and lots of it!

From the research I have been doing over the past year or two, the UK stock market has shown about 5% real return (after inflation) over the long term. I think the USA stock market is slightly better.

I am a total novice, and have fairly little invested, but those are the sort of numbers that pension projects are based on.

If you can keep you personal burn rate low, and over time invest a significant amount of money at 5% return for 10 years, the interest really starts to snowball.

1 in 4 likely means you are on call 24 hours a day for 1 week out of each 4 weeks. The support is then shared between 4 people. That's a fairly typical arrangement in largish companies.

I really enjoyed the Honor Harrington series by David Webber. I've also read most of the Safehold series, also by Webber, which has been excellent so far.

Currently half way through the Lost Fleet Series, which is pretty good too.

Quite different from the 3 series above, the Rho Agenda was very good.

I'm pretty sure the Halifax card I mentioned gives the Visa wholesale rate too. I generally get same rate as quoted on FX sites anytime I've checked it.

If you shop around in the UK for a credit card, you will get one with zero fees and an excellent exchange rate on foreign currency transactions.

I use the Halifax Clarity card and I believe The Nationwide and the Post Office do a good card too.

I'd never use a Debit Card for foreign currency transactions though, they always seem to have fees involved.

It would be great if you could set the points threshold on the RSS feed link too. That would allow all the stories that only make it onto the front page for a tiny amount of time to be skipped from the feed.

I agree that rollbacks are hard, but if you think them through, there is usually a way.

For your example, to merge first and last name, you would create a backup first, ie, create a table that contains the current version of first and last name, and if you need to go back, you restore. Now, if you have an application that cannot have any downtime through DB upgrades you could be in trouble - after your backup, more records could come in to the original table. Probably what you need to do is modify the application to start inserting both full_name and first_name, last_name, and then gradually move over all the old data to the new format etc - ie a gradual migration. Every application and example is different, but the problem is usually solvable, often with some amount of effort.

The master schema create script is an idea I've seen in the past, and it works ok-ish up to a point. The problem is you need to update the master script, and also produce scripts to migrate production, so it doubles the effort. For enterprise database developers, I've seen plenty of development approaches that boil down to:

* Create all changes on Dev using GUI tools. * Used TOAD to diff against production and produce delta script. * Apply delta to prod.

Its gets the job done, mostly, so long as someone doesn't drop the dev database which is the only place with the new code - I've saw it happen. Then you have to go and beg the DBA to do a full restore just you get the release back on track!

This is an area I have experimented with in large enterprise environments, usually with Oracle databases.

In one job I created a process where we used Rails style migrations with some Ruby code I put together. As well as schema migrations, I had it handle applying stored procedures to the database too.

Since then I created a little application called dbgeni (database generic installer) and have been using it successfully to create Dev and Test environments in another job. It changes the schema with migrations and applies changed stored procedures too.

One difference with dbgeni and other tools, is that dbgeni uses plain SQL files and applies them using the database CLI - ie sqlplus for Oracle, the mysql command for MySql and isql for Sybase. This means that if you take the tool away, the SQL scripts are still useful - that allows me to use the tool for applying changes to dev and test environments, but since the production DBAs don't want to use it, I can just bundle up the same scripts and pass them over.

The project is on github https://github.com/sodonnel/dbgeni and it installs as a Ruby gem. I have supporting website http://dbgeni.appsintheopen.com with more details.

The MySQL clustered index (and Oracle IOT) are, in my opinion, often over looked powerful features. To pick an example I recently encountered ... I have an Rails app to implement a RSS reader. Each user has many feeds that get added over time. If you cluster the feeds on user_id then all the feeds for a given user will be in a single database block or two, so reading them out is super efficient. If they are in a standard heap table, (or the way Rails creates the table by default, clustered on ID), then reading the list of N feeds could require N database block reads in the extreme case. Same goes for all articles on a given feed - cluster the related articles together and its very efficient to find the most recent articles for a feed.

Oracle does seems to make limit queries much harder than they need to be. I wrote an article about limit and paging queries in oracle ages ago that may interest you, but you probably have it all figured out now.

http://betteratoracle.com/posts/18-limiting-query-results-to...

Quite a while ago now, there was a period where you couldn't get to Stackoverflow on Virgin media. There were some quite long threads in Virgins forums about it, and apparently not all virgin customers were affected! It did eventually resolve itself, but it was very frustrating!

http://community.virginmedia.com/t5/Up-to-120Mb-Setup-Equipm...

Funny, BT was the example I was thinking of too! I've moved on now but I had a good 9 or so years there.

On the we vs I point, there is a class of interview where you get marked down for using we instead of emphasizing your own contribution. I'm thinking of the competency based interview that is typical of internal positions and promotion boards in big companies. If someone has been through a lot of them, they may be inclined to emphasize their own contribution over the team. The feedback I got on my first failed promotion interview mentioned too much we and not enough I!

You get root on the droplet, so I'm pretty sure you can install whatever you want. I run a couple of rails apps on a 512mb droplet and it works great. Very low traffic on my sites.

Is there somewhere Dublin startups tend to advertise openings? I'm in Belfast and I have wondered in the recent past if there would be any opportunities to work semi-remotely for a Dublin based startup. There is plenty of work in Belfast, but it tends to be Java / .net developers for big corporations.

> twitter-bootstrap-rails, therubyracer, and libv8

Those three gems recently caused me a world of pain. I've been in and out of Rails development since the 1.x versions (more hobby projects than professional development), and once getting going was as simple as 'gem install rails'. Getting the asset pipeline to work changed that totally - I get the advantages of the asset pipeline, and once you get it working and all your gem dependencies sorted it is fine, but it took me a while to get there.