dap

46 ms·

391.
by dap 13y ago
For a read to fail as a result of storage node failure, all replicas would have to be down. You may see increased latency on successful reads if some but not all of the replicas are down. Recall that writes always create new objects. The
392.
by dap 13y ago
It depends on what "goes down" means. The vast, vast majority of such issues are transient: the node panics or loses power and comes back up. There's no impact to data availability or durability unless all nodes storing a c
393.
by dap 13y ago
It's not "for" any particular system. It's a Unix environment.
394.
Rough Road Ahead, but a Robot Driver Takes It in Stride(nytimes.com)
1 points by dap 13y ago|0 comments
395.
by dap 13y ago
Yeah, that's a more plausible reading. Thanks. But that undermines the original point about big companies even more. Having friends at mostly big companies is not only not enough to generalize about big companies, but it provides no
396.
by dap 13y ago
> I have friends in almost all big companies and I discuss them about these issues a lot. Almost all of them agree that they are in a similar situation. Really? You have friends in almost all big companies? There's a lot of truth i
397.
by dap 13y ago
> Can you just learn from his experience that he describes, and get over an inconsequential phrase at a tiny part of the post? Actually, no. The post wasn't phrased as "here was my experience", and rightly so: there are l
398.
by dap 13y ago
I bought my first iPhone in 2008 primarily because it gave me access to live bus arrival times and an interactive map. Other reasons that were far less important were the ability to look up random stuff on the internet, the fact that I hat
399.
by dap 13y ago
I wish there were more options at the low end of this poll, even if those options don't represent economically viable options. I want to know how people value the product given what they know today. I know it's still early, but it's hard t
400.
by dap 14y ago
It's an interesting presentation, but it neglects an obvious solution to the nested callbacks problem, which are the various "async"-like control-flow libraries out there. More importantly, it neglects a critical problem with all of these
401.
by dap 14y ago
This completely misses the fact that many network services are not compute bound, and multi-tenancy (as we get from "the cloud") lets "legacy" code make very efficient use of CPU resources, even a larger number of relatively slow cores.
402.
by dap 14y ago
I agree that make's syntax is hard to pick up and often unintuitive, but in my experience, if you reject make, it's a big, red burning flag that you don't fully understand the problems that it solves. The ability to define arbitrary depend
403.
by dap 14y ago
The problem is thinking of it as the cost of implementing the feature vs. doing manual surgery on the production database, without realizing that if you choose the latter, you're also choosing the risk that you'll spend hours debugging the
404.
by dap 14y ago
Thanks for the detailed explanation. It sounds like everything was done to fix the problem except try to figure out what the problem actually was. Why not use tools to see what the program is doing, form a hypothesis, gather data to conf
405.
by dap 14y ago
At the end of the day, you can't get away from the fact that arguments and return values may be NULL or they may not, and if so, NULL is interpreted in a certain way by the program that depends on the context. Whether these semantics are d
406.
by dap 14y ago
You've got a lot of other responses explaining lots of reasons why sports can be interesting, but I'll add my own. IMO, if you think it's about watching strangers "kick a ball around," you're missing the point. It's about the stories tol
407.
by dap 14y ago
It didn't die. It was forked by the community when Oracle close-sourced it. The community fork (called illumos) is being actively developed by multiple companies, which have done significant new feature work (e.g., http://dtrace.org/blog
408.
by dap 14y ago
Great post, showing admirable dedication to software reliability and a solid understanding of memory issues. One of the suggestions was that the kernel could do more. Solaris-based systems (illumos, SmartOS, OmniOS, etc.) do detect both co
409.
by dap 14y ago
Maybe this was just the bias of the post, but it sounded from the post that most time is spent ruling out actual memory errors. Programming in a higher-level language doesn't make such errors less likely or easier to identify. It's true th
410.
by dap 14y ago
http://joyent.com/blog/network-storage-in-the-cloud-deliciou... http://joyent.com/blog/on-cascading-failures-and-amazons-ela... http://joyent.com/blog/magical-block-store-when-abstractions...
411.
by dap 14y ago
Others have pointed out that they don't scale well, but if you mail a plaintext PDF to a special email address associated with your kindle, they'll reformat it for the kindle and deliver it: http://www.amazon.com/gp/help/customer/display.h
412.
by dap 14y ago
Yes, it's goofy that typeof (NaN) == 'number'. But it's not that unreasonable to have a numeric value that's explicitly invalid. Pointers and references in other languages have null, which is a legal value for any type of object, but is e
413.
by dap 14y ago
Phantom writes are one reason ZFS stores checksums separately from the blocks being written. If you store checksums with the block itself, then failure modes like phantom writes can result in reading data that you think is valid (because t
414.
by dap 14y ago
> So, along comes btrfs that mimics the feature-set of ZFS but brings along many advantages (much thanks to it being developed much later). Just like SystemTap mimics the feature set of DTrace and brings along many advantages due to its
415.
by dap 14y ago
Isn't it absurd that someone would consider ignoring all of those points about technical details and user experience of free software because of license issues with other "free" software? As a member of the illumos community, I long ago
416.
by dap 14y ago
Check out how many HN articles on Sundays come from the NYTimes. They put out an enormous volume of high quality content that a lot of people seem to find relevant.
417.
Corporate Open Source Anti-Patterns(slideshare.net)
10 points by dap 14y ago|0 comments
418.
by dap 14y ago
"You're not wrong, [Craigslist], you're just an asshole." I don't know who's legally right, but I've got no sympathy for Craigslist. They've got a monopoly on the data (whether right or wrong) and refuse to either build a decent UI for con
419.
by dap 14y ago
Posting is not the UI problem. Browsing posts is the problem.
420.
by dap 14y ago
I agree with the overall thrust, though the printf debugging technique is much more suitable for a relatively deterministic program like a compiler or command line tool than for many interesting programs like OSes and network services.

More ›