Y
HN Search
Hacker News Search
new
|
comments
|
top
|
jobs
ah-
searching PlanetScale…
1.
▲
2.
▲
3.
▲
4.
▲
5.
▲
6.
▲
8 ms
·
31.
▲
by
ah-
8y ago
Unlikely, IIRC you only had someting in the order of 32 three-instruction slots of memory.
32.
▲
Python 3 Comes of Age
(tech.winton.com)
7 points
by
ah-
8y ago
|
0 comments
33.
▲
by
ah-
8y ago
I'm not sure what else they could do about this. I'm as sad as you are about licensing complications, but I can totally see why they're doing it. And I'd say it's miles better than these parts being closed.
34.
▲
by
ah-
8y ago
I'll need to have a closer look later, but would vaex fit in with somewhat indexed mapped files? E.g. parquet supports column indexes now: https://issues.apache.org/jira/browse/PARQUET-1201
35.
▲
by
ah-
8y ago
Great to see that you're supporting Apache Arrow! That makes it so much easier to gradually switch over.
36.
▲
by
ah-
8y ago
What was your motivation to go with Enterprise? It seems like most features are OSS now.
37.
▲
by
ah-
8y ago
Check out https://pybossa.com/
38.
▲
by
ah-
8y ago
From what I've seen so far it copies every frame via the CPU and that gets quite slow if you want to do it 60 times a second at high resolution. Optimus etc. use DMA, so you can copy directly between the two devices. Hopefully looking
39.
▲
by
ah-
8y ago
If you have two graphics cards and a modern processor you can pass one of them through to a VM that can directly use it with native drivers, so you get native GPU performance in that VM. However that means the picture only goes to the physi
40.
▲
by
ah-
8y ago
What would you use nowadays to actually implement the real time GraphQL subscriptions and event system in practice?
41.
▲
by
ah-
8y ago
I often run into situations where I hope pandas were 50-100x faster. Dask can help, but introduces quite a bit of additional complexity. I'm also looking forward to stricter data models than what pandas currently uses, in particular pr
42.
▲
Linux 4.19-rc4 released, an apology, and a maintainership note
(lkml.iu.edu)
10 points
by
ah-
8y ago
|
1 comments
43.
▲
Tachyum Prodigy CPU Live Blog
(anandtech.com)
1 points
by
ah-
8y ago
|
0 comments
44.
▲
by
ah-
8y ago
Their denoising looks interesting: https://www.youtube.com/watch?v=9yy18s-FHWw
45.
▲
by
ah-
8y ago
kdbs q-sql does improve a lot on that. Especially expressiveness. https://code.kx.com/q/ref/qsql/ http://www.timestored.com/b/kdb-qsql-query-vs-sql/
46.
▲
Introducing the Gandiva Initiative for Apache Arrow
(dremio.com)
2 points
by
ah-
8y ago
|
0 comments
47.
▲
Negative latency
(meanderful.blogspot.com)
3 points
by
ah-
8y ago
|
1 comments
48.
▲
Commercial Suicide – Integration at the Database Level
(dhickey.ie)
3 points
by
ah-
8y ago
|
1 comments
49.
▲
by
ah-
8y ago
Have you seen https://github.com/pangeo-data/pangeo ? It's pretty amazing. There's still a lot left to figure out, even storage formats aren't fully solved.
50.
▲
by
ah-
8y ago
runbox.com
51.
▲
by
ah-
8y ago
https://mitpress.mit.edu/books/machine-learning-1 is great as well.
52.
▲
by
ah-
8y ago
I wonder if this would ever be integrated into Kafka proper. Shipping out historical chunks onto infinite storage seems like a generally sensible thing. This would be even better if it didn't need a modified client.
53.
▲
by
ah-
8y ago
https://meanderful.blogspot.com Worth a read.
54.
▲
by
ah-
8y ago
This is amazing! Is there a roadmap of what is being worked on or is this mainly a dump of their existing codebase? From looking at the commit history it seems like this is pretty actively developed.
55.
▲
by
ah-
9y ago
I do the same. A simple text file that's not in git, structured a bit like markdown. I keep it open in vim, so it's always there when I need it. I've tried using more complex tools like vim outliner and evernote, but simple t
56.
▲
by
ah-
9y ago
Are you planning to open source it? A lot of your competitors have, like http://pipeline.ai/ , https://github.com/pachyderm/pachyderm and recently https://github.com/polyaxon/polyax
57.
▲
by
ah-
9y ago
Seems to be down. What do they have?
58.
▲
by
ah-
9y ago
How are you dealing with software that expect to get aws tokens? Especially with sts AssumeRoles only lasting up to one hour? For example, if you want to run the Athena JDBC driver locally or read something from S3 in a longish (>1h) run
59.
▲
by
ah-
9y ago
I've switched to https://github.com/boz/kail now, as stern seems to somehow miss pods coming up.
60.
▲
by
ah-
9y ago
FWIW, I've done the same recently (deploy ES + Kibana on K8s) and it pretty much just worked. Statefulset, EBS volume claims, official Docker images. Didn't use XPack or anything fancy though, haven't updated it and the only
More ›