puzzle

34 ms·

301.
by puzzle 8y ago
AWS might be larger than GCP, but the rest of Google (Search, Adwords, Adsense, YT, Gmail, Maps, Apps, etc.) is larger than all of AWS+Amazon and that's the kind of resources over which Google can amortize things. https://el
302.
by puzzle 8y ago
Last time I checked, Google's public renewable energy footprint was 2x Amazon's, as a proximate for the actual footprint, so it's not that crazy to imagine that there are plenty of slack resources to run small VMs or even Bor
303.
by puzzle 8y ago
The rumor at Kubecon (in December, right after re:Invent) was that at the time there were only five people working on EKS, so you shouldn't have expected much. Thus, even if the team grew in the meantime, and it must have, you can&#x27
304.
by puzzle 8y ago
Especially the screen! Nothing else running Windows at the same price and resolution.
305.
by puzzle 8y ago
It's not a novelty, but they aim to bring traditional costs down. One of the founders is Bob Wyman, who was one of the most interesting characters at Google. He could talk for hours about prospective search (he had founded PubSub), his
306.
by puzzle 8y ago
If you're running Kubernetes, you probably have multiple services running (or it's not really worth the complexity). Then whatever slack you have in the cluster can be amortized over all your services, if they all share the same r
307.
by puzzle 8y ago
Not 100%, but running them at least 50% is not a bad idea. In these days of containers and QoS scheduling, it's easier to make good use of slack resources. Letting memory sit unused is an even bigger crime! Someone told me that the tea
308.
by puzzle 8y ago
Now you have another problem: the load balancer needs to know accurately when the backend is ready. This can be prone to race conditions, too, especially for services that have very short response times, unless you manage to keep the LB&#x2
309.
by puzzle 8y ago
At the end of the day, even a non-production GAE environment runs on Borg, talks to GFS and Bigtable, etc. They're not going to run the entire stack on bare metal. The environments will have different configurations, will be backed by
310.
by puzzle 8y ago
Production code has no unauthenticated access to source. In all the cases I know, there was a data push for dynamic configurations from source to production, usually adding access control, auditing, validation, etc. Static configurations w
311.
by puzzle 8y ago
I didn't work on GAE, but I know a fair amount about how it ran and, as I read, too, I could think of quite a few things I would have done, had I been a malicious actor. Denial of service is the first to come to mind. I'm sure a r
312.
by puzzle 8y ago
I agree that there are multiple layers of security in place, which is why I said he would have been caught fairly quickly. I'm not going to go into details, but, at the very least, he had access to quite a bit of proprietary informatio
313.
by puzzle 8y ago
Yes, if he had known more about Google infrastructure, he could have done some damage, both active and passive. He had access to a lot of internals. I was surprised by them at first, too, but they all make sense. On the other hand, he would
314.
by puzzle 8y ago
Well, it's interesting on the human and historical side, especially the part that it was all self-taught, but on the technical side, it's not clear from this interview that he's using GCR and how ( https://en.wikipe
315.
by puzzle 8y ago
They never acknowledged Gary, but removed him anyway. The original said "Scalar floating-point instructions operate on single 32-bit operands that reside in the lower 32-bits of the GPRs. These instructions are considered a proper subs
316.
by puzzle 8y ago
Well, purists would have told you that you had to align your data. Yes, later chips did support misaligned accesses, but that came at a cost: extra bus cycles.
317.
by puzzle 8y ago
I'm still proud of my shell sort implementation in 10 or 12 68K instructions. I was young, so it felt like a huge achievement. :-) That and helping a couple of demo coders shave an instruction or two from core loops. These days I yell
318.
by puzzle 8y ago
They never charged for manuals, as far as I know. I have a whole bookshelf of 68K and PPC books, datasheets and references.
319.
by puzzle 8y ago
It was definitely among the very best, but more than 15 years ago I found an error in the E500/Book E reference manual that repeated itself for more than 200 pages. It only was relevant to you if you cared about the actual bit encoding
320.
by puzzle 8y ago
The most concise, yet illuminating description I have heard about ongoing relationships with Jobs is something similar to "dealing with Khrushchev, but without the shoe antics", i.e. that, to him, everything was a zero-sum game, a
321.
by puzzle 8y ago
It's not clear to me if e.g. the Container Linux system will be still built using Gentoo's emerge or the RH tools. You seem to discount the latter, but I guess it depends on what they really mean with "based on Fedora and Red
322.
by puzzle 8y ago
Security.
323.
by puzzle 8y ago
It's corporate lingo. More specifically, PM/UX speak. "I'll ping you on gChat the slides on what to surface when users tap on the map. Then we can ideate together." Only a slight exaggeration, based on my observatio
324.
by puzzle 8y ago
Very interesting, thanks. Does the map show lane information? I don't know how accurate that is outside of the US. Or maybe in this case it's overestimating the impact of live traffic data for that red light. Sometimes it feels to
325.
by puzzle 8y ago
Where in the world are you? If you think Maps has complex routes obsessed with shaving seconds, you should try Waze just for a day. :-)
326.
by puzzle 8y ago
Are you sure that's not Waze? Google Maps is deliberately a lot like the latter, even if both products are from the same company. Maybe there is something in the road network (e.g. a segment with bad information) along the route you ar
327.
by puzzle 8y ago
Sounds more like Google is turning it into an actual business. I assumed that MapBox was making some money, given that it relies on OSM and can cut corners, but it isn't.
328.
by puzzle 8y ago
Kubernetes is actually the third system. In-between, there's Omega. Yegge must have never set up a GSLB or GFE service to appreciate the extra stuff that Kubernetes features.
329.
by puzzle 8y ago
I don't know that for a fact, but I have seen Apple people at KubeCon, for what it's worth.
330.
by puzzle 8y ago
Well, they probably have some kind of tile "lint", need to precompute a bunch of routing data (depending on what their actual algorithm is), sync all their rendering/prerendering pipelines at a common revision, propagate outp

More ›