mjb

38 ms·

301.
by mjb 6y ago
This is a classic paper, and well worth reading if you use relational databases for anything serious. Using 'phenomena' is a nice framing, and one that I found particularly effective for understanding how to write correct database
302.
by mjb 6y ago
"What Good is Temporal Logic?" ( https://lamport.azurewebsites.net/pubs/pubs.html#what-good ) is in the running for my favorite CS paper. Well worth reading, and understandable even if you have a passing famili
303.
by mjb 6y ago
For example, one of the things that computers do a lot in cloud "control planes" is placement optimization. I've got this compute job, or data, or packet, where should I put it? Computers can solve these kinds of problems muc
304.
by mjb 6y ago
This is one of my favorite papers. The core point that I think should get a lot more attention is this one: > When manual take-over is needed there is likely to be something wrong with the process, so that unusual actions will be needed
305.
by mjb 6y ago
Yes, that’s what you need to do if you want to approach the theoretical durability of any storage system. If you’ve got durability ‘budget’ elsewhere in your design, you might not have to do this.
306.
by mjb 6y ago
Thanks for writing it! I've shared that with a lot of people, because it's a great introduction to stepping through the abstraction of QEMU (or libvirt or whatever) and seeing what KVM can do. There's some good articles that
307.
by mjb 7y ago
I've never had crust issues with this technique, although I do use a steam injection oven which makes it easier. It's doesn't seem to have the same effect as enriched dough in making the crust soft.
308.
by mjb 7y ago
> The Tangzhong roux technique was developed in Asia around 2000. The technique was first mentioned by Yvonne Chen in her book, “Bread Doctor”, published in Taiwan in about 2003. Tangzhong means "soup" in Chinese. From http:&#
309.
by mjb 7y ago
That's been my experience. I often make ciabatta with 95% hydration, of which 1/3rd comes from a water roux. Without tangzhong the dough is much sticker and harder to handle, which makes it harder to both form the loaves nicely, a
310.
by mjb 7y ago
As a very minor issue, reference [4] is cited as "P. A. Bernstein, P. A. Bernstein, and N. Goodman". That seems like an excessive amount of Philip Bernstein.
311.
by mjb 7y ago
This is really cool work, very exciting to see if published in paper form (I saw your talk about some of this last year)! I had a question about your completeness argument (in 4.3.1), you say "we typically observe enough of a history t
312.
by mjb 7y ago
Hi, I'm the post author. As you guessed, I'm South African. I worked at Amazon in Cape Town in the early days of EC2, and live in Seattle now. A lot of core development on EC2 (and other AWS products) still happens in Cape Town.
313.
by mjb 7y ago
> Well, when it comes to performance - you can’t use the average if you don’t know the distribution. This is frankly wrong. Performance comes in multiple flavors. Latency is one of those, and there we know that percentiles really matter
314.
by mjb 7y ago
No, that's not true. The problem isn't there protocol, it's the security argument. Bitcoin works with three people with one computer each, as long as none of those people has access to two more computers. As soon as one of th
315.
by mjb 7y ago
The IO operations amplification for 64 of 96 is pretty brutal, and particularly unfavorable in a world where capacity-per-IOPS keeps trending up. I wonder how they'll deal with that.
316.
by mjb 7y ago
Speaking only for myself, because we don’t have anything approaching a standard here, I do about 60% PlusCal and 40% straight TLA+. Mostly the tradeoffs are the ones you mentioned. If I was the only audience of what I was writing, I’d pick
317.
by mjb 7y ago
> it's a huge breath of fresh air as a formal documentation language. Yeah! When I started with TLA+ I was mostly enamored with model checking and proofs. Those turned out to be useful, but the unexpected use of being a really great
318.
by mjb 7y ago
I'm one of the authors of this paper, along with Fan and Tao. I'm also a huge fan and religious reader of The Morning Paper, so it's really cool to see Adrian feature our work. If you don't know The Morning Paper, be su
319.
How AWS’s Firecracker virtual machines work(amazon.science)
8 points by mjb 7y ago|0 comments
320.
by mjb 7y ago
Thanks for pointing out rust-vmm, Steve. Andreea (one of my co-authors on this paper) wrote an article about it a while back, which explains how it fits into the bigger picture: https://opensource.com/article/19/3&
321.
by mjb 7y ago
> QEMU is a complex program, as demonstrated by the complexity of measuring it accurately For sure. Our goal was a fair comparison, but any comparison by one set of criteria is going to paint an incomplete picture, especially of somethin
322.
by mjb 7y ago
Very interesting, I hadn't seen SLOG before. It seems like there's a fairly similar core insight: placing data near where it's used can help with some system properties. They appear to be more latency-focused and we were (pri
323.
by mjb 7y ago
This is my paper (along with Tao and Fan). It's a great feeling to have this published and available, and I'm super proud of the team behind Physalia. There's a lighter-weight introduction to the work here: https:/&#x2F
324.
by mjb 7y ago
It's not really about the design of S3, but if you're interested in some of the philosophy and thinking behind S3 you might enjoy "Beyond eleven nines: Lessons from Amazon S3 culture of durability" https://www
325.
by mjb 7y ago
This is a grab-bag of uncited bunk. pH! Conspiracies! Misreading of studies! Bizarre cures! It's all here.
326.
by mjb 7y ago
Thanks for the useful summary. > All engineers, from entry level to principal. I'd phrase that as "some engineers of all levels". It's certainly not true that all engineers at AWS use TLA+ :)
327.
by mjb 7y ago
Practical TLA+ is indeed a great place to start, and Hillel Wayne's writing out formal methods is all worth reading. Lamport's TLA+ book, and "Specifying Systems" are both available online ( https://lamport.azu
328.
by mjb 7y ago
Yes, since we published this (I'm one of the authors), the Automated Reasoning Group have done amazing work applying formal methods to a broad range of problems at AWS. We're also still using TLA+ in many places for the reasons th
329.
by mjb 7y ago
Lots of work, in both academia and industry. For example, check out Daniel Bittman's work at UCSC, and Tim Kraska's at MIT. Two interesting things are happening here: memory becoming non volatile, and storage and network bandwidth
330.
by mjb 7y ago
> Whenever a server failed and restarted, it would pick up processing from the last successful commit. If a server did not restart, the worker count would be updated and remaining workers would redistribute the partitions among themselve

More ›