tmostak

9 ms·

61.
by tmostak 8y ago
Why not? The Thrust folks have done a lot of good work on implementing highly optimized radix sort on GPU. That said, there is interesting academic work around GPU sort that achieves even higher performance than Thrust in many scenarios, an
62.
by tmostak 8y ago
The MapD (now OmniSci) execution engine is actually built around a JIT compiler that translates SQL to LLVM IR for most operations, but it does use Thrust for a few things like sort. One performance advantage of using a JIT is that the syst
63.
AWS adds 32GB Nvidia V100 GPUs(news.developer.nvidia.com)
2 points by tmostak 8y ago|0 comments
64.
by tmostak 8y ago
If by GPU bandwidth you mean GPU memory bandwidth, being able to operate at memory bandwidth is typically a huge win over CPU, not only because of the big difference between GPU and CPU bandwidth (900 GB/sec vs 150-200 GB/sec), bu
65.
by tmostak 8y ago
I've definitely been hearing about the project but didn't know it was being ported to GPU, that's great news. How does this compare to things like Gunrock?
66.
by tmostak 8y ago
And fwiw MapD is now OmniSci.
67.
by tmostak 8y ago
I was about to share this with the team at OmniSci (GPU analytics platform) and realized the authors included our very own Saman Ashkiani and adviser John Owens. Having fast dynamic data structures on the GPU is of huge utility. People thin
68.
Scaling Pandas to the Billions with Ibis and MapD(mapd.com)
4 points by tmostak 8y ago|0 comments
69.
by tmostak 9y ago
It very much scales! For example, see here for a demo of MapD on 11.6B rows of ship location data. https://www.mapd.com/demos/ships/#/dashboard?_k=8de1f8 We are offering a enterprise Cloud X plan with greater
70.
Announcing the MapD Cloud: self-service GPU-accelerated analytics(mapd.com)
60 points by tmostak 9y ago|7 comments
71.
How VW predicts churn with GPU-accelerated ML and visual analytics(mapd.com)
8 points by tmostak 9y ago|0 comments
72.
Exploring Google Analytics Data with MapD(mapd.com)
4 points by tmostak 9y ago|0 comments
73.
by tmostak 9y ago
Note that MapD can run on CPU as well and is generally quite fast. I haven’t benchmarked but on a dataset of this size the numbers might be similar. I’d also say that while this is a great example of getting up and running with MapD the sys
74.
Getting Started with MapD(randyzwitch.com)
3 points by tmostak 9y ago|0 comments
75.
Connecting Pandas to a GPU Database with Apache Arrow(mapd.com)
3 points by tmostak 9y ago|0 comments
76.
Connecting pandas to a GPU database with Apache Arrow(mapd.com)
8 points by tmostak 9y ago|0 comments
77.
by tmostak 9y ago
Congrats Rodrigo and team! We at MapD look forward to continued collaboration with you guys on the GOAI project and elsewhere!
78.
by tmostak 9y ago
MapD | San Francisco (city), | Backend Developer, Frontend Visualization Developer, Developer Advocate (ONSITE/REMOTE) MapD ( https://www.mapd.com ) is a NEA/Google Ventures/Nvidia/Verizon Ventures/Vanedg
79.
by tmostak 9y ago
Have you tried the native visual analytics client that comes with MapD, MapD Immerse? ( https://www.mapd.com/platform/immerse/ ). It’s not a full BI tool but is very good for interactively slicing and dicing dataset
80.
End-to-end machine learning on the GPU with GOAI(mapd.com)
2 points by tmostak 9y ago|0 comments
81.
Exploring 13 Seasons (7,588,492 Plays) of the NBA in Real Time(mapd.com)
2 points by tmostak 9y ago|0 comments
82.
Interactively explore 11.6 billion rows of US shipping data with a GPU database(mapd.com)
2 points by tmostak 9y ago|0 comments
83.
An interactive GPU-powered deep dive into 11.6 billion rows of US shipping data(mapd.com)
6 points by tmostak 9y ago|0 comments
84.
by tmostak 9y ago
We do support real-time ingest (helped by the fact that we do not need to index on insert). The only example we have online of that now is our Tweetmap demo ( https://www.mapd.com/demos/tweetmap/ )
85.
by tmostak 9y ago
We've considered video rendering using the H264 encoders on the GPU, another approach might be to create a format that has all info needed for interactions and that could be deciphered on the frontend (i.e in WebGL). So it's some
86.
by tmostak 9y ago
MapD predates good virtual (unified) memory support on GPUs and so we built our own caching mechanism where each GPU has its own buffer pool and pulls from a CPU buffer pool (i.e a network of buffer pools). This approach still has the advan
87.
by tmostak 9y ago
We can already do simple histograms (heatmaps). More complicated features such as Gaussian weighting and hexagonal bins are coming.
88.
by tmostak 9y ago
Sorry, as the OP that wasn't the intention (the Vega rendering API has been around for some time and predates our porting it to GPUs).
89.
by tmostak 9y ago
Note that although we can run on CPU, CPUs do not have the graphics pipeline and the memory bandwidth necessary for interactive server-side visualizations like this.
90.
by tmostak 9y ago
MapD customers typically run our product on multiple servers with multiple GPUs per node. So 4 servers with 8 Nvidia P40s each has 4X192GB = 768GB of VRAM. Note MapD compresses data and also keeps data in CPU RAM as needed. Even two servers

More ›