arnon

39 ms·

481.
by arnon 10y ago
GPUs have over 3000 cores nowadays ( http://images.nvidia.com/content/tesla/pdf/nvidia-tesla-p100... )
482.
by arnon 10y ago
SQream DB ( http://www.sqream.com ) is an SQL analytics database which uses GPUs. It is on both AWS P2 and Azure NC machines, and it can do very fast analytics on hundreds of raw terabytes on those. Having said that, on-premise or
483.
by arnon 10y ago
I've heard that before. I'll believe it when I see it :)
484.
by arnon 10y ago
Takes 10 minutes to install nv-docker correctly, and you're golden.
485.
by arnon 10y ago
Are the cards physically in the same machine, or are they in a remote chassis and connected through the network? I think on P2, Amazon's cards are remote, and there is pretty significant latency when using them for time-sensitive compu
486.
by arnon 10y ago
The only problem is GeForce 1080s (for example) burn out much quicker under heavy load. They're not designed to sit inside an enterprise chassis
487.
by arnon 10y ago
Amazon Linux AMI for their GPU instances also does not include the CUDA driver.
488.
by arnon 10y ago
Terrifying
489.
by arnon 10y ago
That's not a fair comparison
490.
by arnon 10y ago
They all operate the same way give or take
491.
by arnon 10y ago
There's very little actual data here... Oracle has a bit more data on their website.. https://docs.oracle.com/database/121/TGSQL/tgsql_optcncpt.ht...
492.
by arnon 10y ago
Oh god, /another/ Apache fast data analytics platform. Apache is like a powerhouse in creating these short-lived projects that force companies to get the most expensive, esoteric talents.
493.
by arnon 10y ago
Maybe, but that's a very specific situation. You don't typically create a full RDBMS for a specific data model
494.
by arnon 10y ago
1. Not everything needs to be pushed up to the GPU. Some things are better left in RAM. 2. What if you only push indexes or similar up to the GPU, like an AB-tree index? You're keeping all of the 'heavy' stuff down, and only
495.
by arnon 10y ago
I don't think the development of pg-strom is still active, as far as I can tell
496.
by arnon 10y ago
You're thinking about transactional databases, and you're right. Transactional databases will probably not benefit hugely from a GPU. That's not saying it's impossible, but probably not worth the effort. However, there a
497.
by arnon 10y ago
This example is very straight forward, but it's probably an outlier. The alternative might be too costly in terms of storage or computation...
498.
Why would you even build a database?(sqream.com)
2 points by arnon 10y ago|0 comments
499.
by arnon 10y ago
On the grand scheme of things, TPCH-1 is not a very compute intense query, and it doesn't really reflect on most real life scenarios. I'd love to see some other benchmarks for this kind of operation.
500.
by arnon 10y ago
Unfortunately know some bottom-right quadrant people. We got rid of some of those not too long ago - wouldn't want them on my team again. Super hard to manage and work with.