Official OpenUCX Twitter account
www.github.com/openucx
Joined September 2018
- Tweets125
- Following0
- Followers207
- Likes40
openucx retweeted
Tomorrow, I’m giving a webinar/Q&A on the current state of @RAPIDSai. nvidia.com/en-us/gtc/session… register to see how we scale to 10TB problems with @dask_dev @blazingsql @openucx & @numba_jit. Plus, we have new integrations with @plotlygraphs @raydistributed & more at #GTC2020!
openucx retweeted
Strongly recommend! Blazing is amazing :)
And a shoutout to some of the amazing infra projects that help make this possible: @openucx @ApacheArrow @dask_dev and @ApacheCalcite
We recently released the BlazingSQL + @RAPIDSai Stack on @awsmarketplace! With a few easy steps, you can launch your own #GPU powered data science workstation at the push of a button.
Find it here: aws.amazon.com/marketplace/p…
Or read our blog post for more: link.medium.com/y9ZyBslMH6?
openucx retweeted
This is the path to Windows support for @rapidsai. Super hype that this is finally publicly announced! Can't wait to play with @blazingsql @ChainerOfficial #cupy @openucx and more in Windows in the future! 😎
To deliver CUDA GPU-acceleration support to Windows users, @microsoft and NVIDIA announce a public preview for GPU in Windows Subsystem for Linux (WSL). WSL, launching later in June, is a layer that enables executing Linux binaries on Microsoft Windows computing systems. #MSBuild
openucx retweeted
Plus, we will only get faster thanks to @openucx and more optimization we have scheduled. Ohh, and so many new partners like @anyscalecompute, @raydistributed, @plotlygraphs, and more integrating @RAPIDSai every day. Join in! The #Python #GPU ecosystem has never been stronger.
openucx retweeted
Crazy amazing work from all involved.
We have tons of optimizations on the @blazingsql side of this too! @openucx integration will supercharge our data shuffle, SQL & CBO optimizations currently in the pipeline, and many many more.
openucx retweeted
TPCx-BB at 10TB run with @rapidsai @openucx @dask_dev @numba_jit @ChainerOfficial #cupy @blazingsql is open source here: github.com/rapidsai/tpcx-bb. Beyond proud of the team that delivered these results in a short period of time with the @PyData ecosystem.
openucx retweeted
This is a key to scaling #cuDF on @dask_dev. Communication is a huge bottleneck in distributed computing and @openucx is a path to acceleration without giving up the resilience and scalability expected.
Distributed #DataScience processing on @rapidsai keeps getting faster. See the latest benchmarks from our high-performance #Python communication library UCX-Py. nvda.ws/3eteqEu
openucx retweeted
@quasiben finishes off the performance season by talking about @openucx for high performance networking in Python
docs.google.com/presentation…
openucx retweeted
Congrats!
I think @openucx has some of the highest awesomeness:awareness ratio of any project. It’s like air traffic control. Easy not to think about, but instrumental to the present and future of speed!
openucx retweeted
Christmas Eve, talking gifts w/ the fam, I want to thank the @RAPIDSai team for @openucx-py, a #GPU #datascience community gift. We hope many use it to accelerate multi-GPU libraries. For those C++ users, here's a distributed join example w/ UCX & #cuDF github.com/rapidsai/distribu…
What a year! We view @RAPIDSai as the #GPU ecosystem connector; #NVDashboard, @datashader integration, & @dask_dev #cloudprovider, are continuations of work we've done w/ @ApacheArrow, @numba_jit, @PyTorch, @ChainerOfficial & more. Enjoy the holidays & get ready for an epic 2020!
openucx retweeted
0.11 is here...more speed/scale/reliability. libcudf modernization, more #cuML MNMG models, @openucx-Py package, RAPIDS & @datashader, @dask_dev cloudprovider, #NVDashboard for GPU monitoring, new @blazingsql logo & more! Cheers to 2020 from @rapidsai. medium.com/rapids-ai/rapids-…
openucx retweeted
Woot! @RAPIDSai + @datashader, bringing @NVIDIA #GPU back to #Datavisualization 🙂 Now let’s scale this with @dask_dev and @openucx.
Datashader Release 0.9.0 is out, now supporting GPUs at last! Short story is that in our testing a NVIDIA GPU was 40X faster than single-core Pandas and 4.5x faster than multi-core Dask for rendering 100 million points. See github.com/holoviz/datashade… for the details.
openucx retweeted
Python, Performance, and @nvidiaai #GPUs! Learn how @dask_dev, @numba_jit, @openucx, @RAPIDSai, and more come together to give #HPC speed with Python ease. lnkd.in/gDRf8Aj
openucx retweeted
Slides on four libraries for high performance Python and how they play together, delivered at @PASC_Conference
- @numba_jit
- @dask_dev
- @rapidsai
- @openucx
docs.google.com/presentation…
openucx retweeted
An update on recent work on high performance networking with @dask_dev and @openucx . blog.dask.org/2019/06/09/ucx…
This also includes a discussion on hardware with heterogeneous networks
openucx retweeted
0.7 is live! @RAPIDSai is moving on up in the best of ways. More bridges fewer walls between RAPIDS & @XGBoostProject, @PyTorch, @numba_jit, @GCPcloud, & so many more! @Dask_dev & #GPU keeps improving (@openucx integration coming in 0.8/0.9)... just WOOT! medium.com/rapids-ai/rapids-…
openucx retweeted
“@NERSC and @NVIDIA will develop a set of #GPU-based high performance data analytic tools using #Python, the primary language used for #DataAnalytics at NERSC” Woot! Bridging #HPC & enterprise w/ @RAPIDSai @ApacheArrow @dask_dev @openucx-py and more from #PyData. I love my job!
.@NERSC and @nvidia have signed a contract to enhance GPU compiler capabilities for @BerkeleyLab's next-generation Perlmutter supercomputer: ow.ly/Gtj030o8LqG https://teapawt.net/t.co/85njEYFlcb #HPC @ENERGY
