uluyol

8 ms·

61.
by uluyol 4y ago
I can see how you might want HTTP/3 across data centers, but you may want to stick to HTTP/2 for intra-DC workloads. There's a ton of work to optimize TCP including hardware offloads that help push higher throughput. Basicall
62.
by uluyol 4y ago
This depends on your university, and also you should compare against a benchmark that's not cushy tech companies. I had spouse + child health care with no premium and only copays, no coinsurance. Didn't need the child care benefit
63.
by uluyol 4y ago
PhD students really don't get paid much. Tuition is waived if you're a TA, and you get a small stipend + good benefits. If you're an RA, you bring money into the university (through being the engine that pulls in grant money
64.
by uluyol 4y ago
https://xkcd.com/882/ In a world where no one reproduces results, test enough hypotheses and one is bound to hold.
65.
by uluyol 4y ago
> Through Mesa and Gallium3D, we benefit from thirty years of OpenGL driver development, with common code translating OpenGL into the much simpler Gallium3D. Thanks to the incredible engineering of NIR, Mesa, and Gallium3D, our ragtag te
66.
by uluyol 4y ago
Much of Android is written in Java/Kotlin. Transitions from the JVM to C/Rust and back are not going to be free due to extra wrappers and the inability to optimize across the boundary. So for some usecases Java will be faster. But
67.
by uluyol 4y ago
I never said that cops have to to pull the trigger. Just that another person's life being at risk is justification for lethal force. To be clear though, that's all I'm saying. I'm not making a statement about my overall
68.
by uluyol 4y ago
Or someone else's life is at risk.
69.
by uluyol 4y ago
That reference is pretty old at this point, especially considering that some of those terminal emulators were very young projects still. I would search for new results before making a statement about performance.
70.
by uluyol 4y ago
I don't have a problem with value based pricing. But I think I'm just going to end up repeating myself now so I'm going to check out of this discussion now.
71.
by uluyol 4y ago
Neither is ethical, because the company has the option to reduce customer costs while increasing margins , and they are doing the opposite. This is especially true for monopolies. But that's actually why, despite my complaints, I&#x27
72.
by uluyol 4y ago
> They're all printed from the same wafers. Not just that. When you fab a design you'll get sometimes get errors here or there and have to disable things. Or it won't be safe to drive the thing to the highest clock speeds.
73.
by uluyol 4y ago
I wonder how big Twitter's infra costs are (capex). Based on the article, they need to build capacity > 2x peak loaf to tolerate one of the DCs failing. But if they instead had 4 and still only tolerated 1 failure, they'd only
74.
by uluyol 4y ago
There are different semantics for these data structures in different languages. These tend to exist for corner cases, but those corner cases constrain your design. So you might not be able to replace the default Java hashmap with absl::flat
75.
by uluyol 4y ago
This won't help you in a future version of Java where you can write List<double>, right?
76.
by uluyol 4y ago
"Call your friend and convince them to hand over their key if you want out"
77.
by uluyol 4y ago
Link? This is upsetting if true.
78.
by uluyol 4y ago
Google is using remote memory accesses rather than TCP for at least some classes of traffic (e.g. a caching system). They've been publishing details about how it all works too. Also, they have a transport (Pony express) developed speci
79.
by uluyol 4y ago
Not necessarily, especially when you factor in limited (human) resources in. Zink has outperformed native GL drivers even though it introduces an extra layer through vulkan. Ideal world, we'd have direct GL to metal or hardware, but im
80.
by uluyol 4y ago
I have had a nightmare dealing with their customer service. Hour and half holds, dropped calls, inconsistent messaging, losing my information (personal address e.g. despite it being on an incident tracking website), lack of communication, d
81.
by uluyol 4y ago
It took much longer than that for the AMD driver's to get to where they are today. Newer drivers get to piggyback off the work done for others (yay open source) but I think it will likely take longer than 5 years before NVK and friends
82.
by uluyol 4y ago
Do they shrink too? How many threads can be created before address space is exhausted (even if the memory isn't backed by pages, the address space is still reserved)?
83.
by uluyol 4y ago
Arch is a rolling release distro. So it gets hardware support faster than other distros that stick to older kernels and userspace. Most users are not on rolling releases. I personally like the rolling approach, but that doesn't reflect
84.
by uluyol 4y ago
Supercomputers? Rack-scale computing? See some of the work being done with RDMA and "far memory".
85.
by uluyol 4y ago
I'm not sure but there are several active parallel efforts to speed up python (faster-cpython, cinder, pyston). The faster-cpython project is doing 100% upstream development, and I'm not sure but it seems like cinder and pyston ar
86.
by uluyol 4y ago
This is a nice article, but one this that's not quite right is that you can go from a histogram to an eCDF (basically view the bucketing as a loss in measurement precision). I mention this because histograms, especially HDR histograms,
87.
by uluyol 4y ago
Google publishes a lot of papers in AI, networking, and kernel-type work (isolation, scheduling, etc.). This seems to attract a lot of talent and helps build relationships with academia. For whatever reason though, the storage and analytics
88.
Jupiter evolving: Reflecting on Google’s data center network transformation(cloud.google.com)
5 points by uluyol 4y ago|0 comments
89.
Lazier Zink Descriptors(supergoodcode.com)
1 points by uluyol 4y ago|0 comments
90.
by uluyol 4y ago
I'm sure it's encrypted, but I doubt that they compress everything. Images and video tend not too compress well since they've typically already been aggressively compressed with specialized algorithms. It would just be throwi

More ›