tmostak

8 ms·

1.
GPU-Accelerated Presto(prestodb.io)
3 points by tmostak 3mo ago|0 comments
2.
by tmostak 4mo ago
GPU-accelerated databases have a long history. I founded HeavyAI (previously MapD/OmniSci) in 2013, but there are or have been many other startups in this space, such as Voltron Data, Kinetica, Sqream, etc. And now you have major playe
3.
by tmostak 8mo ago
Evidence (preferably with recent Teslas/HW4)?
4.
by tmostak 8mo ago
Evidence of this? I own a Tesla (HW4, latest FSD) as well as have taken many Waymo rides, and have found both to react well to unpredictable situations (i.e. a car unexpectedly turning in front of you), far more quickly than I would expect
5.
by tmostak 8mo ago
Do you have data to back this claim up, specifically with HW4 (most recent hardware) and FSD software releases?
6.
by tmostak 11mo ago
Even without NVLink C2C, on a GPU with 16XPCIe 5.0 lanes to host, you have 128GB/sec in theory and 100+ GB/sec in practice bidirectional bandwidth (half that in each direction), so still come out ahead with pipelining. Of course p
7.
by tmostak 1y ago
We've made extensive use of perfect hashing in HeavyDB (formerly MapD/OmniSciDB), and it has definitely been a core part of achieving strong group by and join performance. You can use perfect hashes not only the usual suspects of
8.
Fast geospatial aggregation and visualization with Uber H3(heavy.ai)
1 points by tmostak 1y ago|0 comments
9.
Benchmarking geospatial join performance on GPU vs. CPU(heavy.ai)
1 points by tmostak 1y ago|0 comments
10.
Benchmarking a GPU database against CPU data warehouses(heavy.ai)
1 points by tmostak 1y ago|0 comments
11.
Interactively Explore 20 Billion Records of Ship AIS Data Using a Single GPU(heavy.ai)
1 points by tmostak 2y ago|0 comments
12.
by tmostak 2y ago
This looks amazing! Just looking through the code a bit, it seems that the model both supports a (custom) attention mechanism between features and between rows (code uses the term items)? If so, does the attention between rows help improve
13.
by tmostak 2y ago
You should be able to train/full-fine-tune (i.e. full weight updates, not LoRA) a much larger model with 96GB of VRAM. I generally have been able to do a full fine-tune (which is equivalent to training a model from scratch) of 34B para
14.
Benchmarking a GPU-Accelerated Database Against CPU Data Warehouses(heavy.ai)
1 points by tmostak 2y ago|0 comments
15.
Benchmarking GPU-Accelerated HeavyDB Against CPU Data Warehouses(heavy.ai)
4 points by tmostak 2y ago|0 comments
16.
Embed AI fuzzy logic into your SQL(heavy.ai)
2 points by tmostak 2y ago|0 comments
17.
by tmostak 2y ago
This assumes that you can linearly scale up the number of TPUs to get equal performance to Nvidia cards for less cost. Like most things distributed, this is unlikely to be the case.
18.
Exploring 36 years of FAA flight data with AI and a GPU Database(tech.marksblogg.com)
1 points by tmostak 2y ago|1 comments
19.
Exploring 36 years of FAA flight data with AI and a GPU Database(tech.marksblogg.com)
9 points by tmostak 2y ago|1 comments
20.
by tmostak 2y ago
Are you measuring tokens/sec or words per second? The difference matters as generally in my experience, Llama 3, by virtue of its giant vocabulary, generally tokenizes text with 20-25% less tokens than something like Mistral. So even i
21.
by tmostak 2y ago
But it's likely to be much slower than what you'd get with a backend like llama.cpp on CPU (particularly if you're running on a Mac, but I think on Linux as well), as well as not supporting features like CPU offloading.
22.
by tmostak 3y ago
Thank you, it's been a major team effort!
23.
Show HN: Use natural language to query and visualize 400M tweets(demo-heavyiq.heavy.ai)
11 points by tmostak 3y ago|3 comments
24.
Explore 400M tweets with LLM-powered conversational analytics(demo-heavyiq.heavy.ai)
2 points by tmostak 3y ago|1 comments
25.
by tmostak 3y ago
More info can be found here: https://www.heavy.ai/heavyiq/overview
26.
by tmostak 3y ago
HEAVY.AI | SQL Analyst/Wrangler | Part-time or Full-time | Remote HEAVY.AI builds a GPU-accelerated analytics platform that allows users to interactively query and visualize billions of records of data in milliseconds. We’re looking fo
27.
by tmostak 3y ago
These are awesome!
28.
by tmostak 3y ago
I assume if MSFT/Satya are supportive it won't be an issue.
29.
by tmostak 3y ago
It wasn't clear to me what evaluation method was being used, the chart in the blog says Execution Accuracy, but the numbers that seem to be used appear to correlate with "Exact Set Match" (comparing on SQL) instead of the &qu
30.
by tmostak 3y ago
I agree that Spider queries are not necessarily representative of the SQL you might see in the wild from real users, but looking at some analysis I did of the dataset around 43% of the queries had joins, and a number had 3, 4, or 5-way join

More ›