Top all-time · #1434
pl.atyp.us
Personal blog · pl.atyp.ussystemsdataprogramming
pl.atyp.us is a personal blog at pl.atyp.us about systems, data and programming. Since 2007, 52 different Hacker News readers have linked it 93 times, most often “High performance server architecture”. No feed was found, so read it on the site itself.
Most-linked posts?
- High performance server architecture7 people · 7 mentions · first linked 2007-12-16
- Fixing Fsync3 people · 3 mentions · first linked 2013-12-02
- It’s Faster Because It’s C3 people · 3 mentions · first linked 2010-07-13
- Leaning Out2 people · 3 mentions · first linked 2013-10-10
- Updating Posix2 people · 3 mentions · first linked 2016-06-07
- /wordpress/index.php/2010/10/when-partitions-attack/2 people · 2 mentions · first linked 2013-11-25
- Levels of Tech-Familiarity2 people · 2 mentions · first linked 2014-09-10
- Real Parallel Filesystems2 people · 2 mentions · first linked 2010-02-12
- Is Eventual Consistency Useful?2 people · 2 mentions · first linked 2013-04-03
- Unpopular Thoughts About Network Neutrality2 people · 2 mentions · first linked 2017-12-28
What Hacker News says?
Not to beat a dead horse or keep a totally unproductive thread alive longer than it has to be, but, if this is you: http://pl.atyp.us/ (from your profile) you seem like an interesting person with a lot of experience and perspective to contribute. So with that said: why not be specific about the selective over-application you're talking about? I couldn't find it in the comment history. People probably won't agree wit…
Comment linking /
on “New Hacker News Guideline: Avoid Gratuitous Negativity”Submitted It’s Faster Because It’s C as a story · 200 points
HN discussionInstead of repeating myself, I'll just cite the blog post I wrote about this, and hope the rest of HN doesn't mind. http://pl.atyp.us/2013-10-leaning-out.html Things have changed only a bit since then. A couple of years' worth of still-increasing income have brought me even closer to permanent financial security. I'm still pretty unimpressed by where this industry is headed technically, socially, and financially. I…
Comment linking /2013-10-leaning-out.html
on “Valuing time over money is associated with greater happiness”Submitted A Gluster developer's thoughts on Torus as a story · 164 points
HN discussionYou couldn't be more wrong. I was writing about one specific incident early in my career, quite likely (judging by your lack of maturity) while you were still in kindergarten. The whole point of the exercise was to put in the extra effort once when it really mattered, so that I wouldn't be subjected to the "work hard all the time" grind that has become such a dysfunctional norm in the startup world. It worked. Since…
Comment linking /2013-10-leaning-out.html
on “A culture of beer and overtime”Submitted Availability and partition tolerance explained and diagrammed as a story · 40 points
HN discussionI did something very much like this for a while. I'd use WordPress to create content on my laptop, then I'd crawl my own site using wget to collect the results and push those up to the real website. Turns out there are a few gotchas that simply-static didn't address, but I wrote a blog post about my experiences here. http://pl.atyp.us/wordpress/index.php/2013/04/using-wordpres... The older posts on my site were all…
Comment linking /wordpress/index.php/2013/04/using-wordpress-to-generate-static-pages/
on “Static Website Generators Are the Next Big Thing”Submitted Real Parallel Filesystems as a story · 20 points
HN discussionThey managed to write an entire article about POSIX and never mention the filesystem parts? Even the section about "asynchronous I/O" seems to have more to do with locking than actual I/O. Nothing about directory operations, permission models, consistency/durability requirements, and all the other things that make that part of POSIX the most egregiously out of date. I wrote a bit about this ( http://pl.atyp.us/2016-…
Comment linking /2016-05-updating-posix.html
on “Posix Has Become Outdated (2016) [pdf]”Submitted Thinking at Cloud Scale as a story · 15 points
HN discussiontl;dr it's not nearly as cool as it could have been. I already posted a more detailed explanation here: http://pl.atyp.us/2013-10-comedic-open-storage.html I tried to post a comment on the NSOP (Not So...), but first I got "HTTP internal error" and then I got "duplicate comment" but it still hasn't shown up, so I'll post it here. "The “private” bit is important; although various techniques have been created for shar…
Comment linking /2013-10-comedic-open-storage.html
on “Seagate just reinvented the disk interface using Ethernet”Submitted Getting Silly About NoSQL as a story · 11 points
HN discussionChess teaches concentration, but also other skills as well. As I wrote in 2002[1], it also teaches things like strategic planning and resource management, ability to persevere despite setbacks, time/focus management, etc. It's certainly not unique in this, but it's a pretty good mix. [1] http://pl.atyp.us/wordpress/index.php/2002/05/what-chess-has...
Comment linking /wordpress/index.php/2002/05/what-chess-has-taught-me/
on “Playing chess is a life lesson in concentration”Submitted Distributed File System SBFAQ as a story · 9 points
HN discussionI enjoyed this thought piece on technical debt vs technical rsik: http://pl.atyp.us/2015-01-technical-risk.html
Comment linking /2015-01-technical-risk.html
on “Organizational debt is like technical debt, but worse”Submitted Fixing Fsync as a story · 9 points
HN discussionIt's easy to criticize when you've never worked on the kind of infrastructure that Gluster has. Yeah, for a project I'm doing all by myself it would take all of five minutes to update the web page with that sort of information. Hey look, that's exactly how it happened when this was part of a project I was doing myself. http://pl.atyp.us/hekafs.org/ Now, on a huge project with the website and docs with separate owner…
Comment linking /hekafs.org/
on “Introducing the Infinit file system”Submitted A response to NuoDB's "patent" as a story · 9 points
HN discussionThis is cool but I think it's important to remember that they're comparing algorithmic benchmarks but modern software has many different bounds besides for raw, algorithmic processing. For example, I think I/O and database interactions are the bottleneck for most web apps. I really liked the way this article broke it down: http://pl.atyp.us/wordpress/?p=2947
Comment linking /wordpress/?p=2947
on “Google paper comparing performance of C++, Java, Scala, and Go [PDF]”Submitted Why Buy Red Hat Enterprise Linux? as a story · 7 points
HN discussionThe author brings up a lot of good points, but they're mostly good within the context of HPC. Out in the broader universe where most data lives, many people do require the full suite of POSIX namespace, metadata, and permission semantics. Yes, even locking, no matter how many times we tell them that relying on that in a distributed system is Doing It Wrong. I know because I support their crazy antics at one of the b…
Comment linking /2016-05-updating-posix.html
on “What's So Bad About Posix I/O?”Submitted High performance server architecture as a story · 7 points
HN discussionThere's also an excellent writeup by GlusterFS contributor Jeff Darcy: http://pl.atyp.us/wordpress/index.php/2012/10/rackspace-bloc...
Comment linking /wordpress/index.php/2012/10/rackspace-block-storage/
on “Rackspace Launches Developer Discount”Submitted High-Performance Server Architecture, Software Scaling as a story · 5 points
HN discussionAh, the myth of the sufficiently redundant network. http://pl.atyp.us/wordpress/index.php/2010/10/when-partition... http://kellabyte.com/2013/11/04/the-network-partitions-are-r...
Comment linking /wordpress/index.php/2010/10/when-partitions-attack/
on “How To Make An Infinitely Scalable RDBMS”Submitted High-Performance Server Architecture as a story · 5 points
HN discussion> A trivial system with these properties is one that replicates every update to all available nodes and requires reads to get matching results from at least half of the nodes. Nope. You haven't thought this through. One problem is that "all available nodes" is one of those fuzzy concepts that sounds reasonable but is actually impossible to actually do. Say you send a write to 3 replicas, all of which you think are u…
Comment linking /wordpress/?p=2521
on “A Dismal Guide to Concurrency”Submitted Key-Value Stores: Beyond Filesystems (Jeff Darcy) as a story · 5 points
HN discussionIn principle, yes. I believe most modern filesystems use (kernel-level) write barriers to ensure filesystem consistency, and those do depend on having hardware barriers as well. The kernel is not set up to export this interface, however, and there doesn't appear to be any serious work being done to fix that. EDIT: http://pl.atyp.us/2013-11-fixing-fsync.html has more details.
Comment linking /2013-11-fixing-fsync.html
on “Linux's fsync() woes are getting some attention”Submitted Introduction to Distributed Filesystems as a story · 4 points
HN discussion
