Twitter | Search | |
Peter Rudenko
Software engineer interested in system engineering, hpc, machine learning, big data processing, distributed systems.
659
Tweets
1,324
Following
507
Followers
Tweets
Peter Rudenko retweeted
Matthew Rocklin Jun 6
An update on recent work on high performance networking with and . This also includes a discussion on hardware with heterogeneous networks
Reply Retweet Like
Peter Rudenko Jun 5
Scaling Distributed Machine Learning with In-Network Aggregation:
Reply Retweet Like
Peter Rudenko retweeted
SPCL@ETH Mar 7
Torsten’s keynote at the Austrian HPC Meeting in Grundlsee on “High-Performance Communication for Deep Learning”, Feb 26 is now online (slides, video):
Reply Retweet Like
Peter Rudenko retweeted
Miles Brundage Feb 20
"Optimizing Network Performance for Distributed DNN Training on GPU Clusters: ImageNet/AlexNet Training in 1.5 Minutes," Sun et al.:
Reply Retweet Like
Peter Rudenko retweeted
Theodore Vasiloudis Oct 25
Great new PhD-level course starting as a cooperation between , Telecom Paris, and , "Fundamentals of Machine Learning over Networks"
Reply Retweet Like
Peter Rudenko retweeted
IBM Research Oct 25
Ocean Tensor Package: General-Purpose Software for Tensor Computation
Reply Retweet Like
Peter Rudenko retweeted
Logical Clocks Oct 22
Why we replaced Horovod (the great) with the new AllReduce implementation in - CollectiveAllReduceStrategy -
Reply Retweet Like
Peter Rudenko retweeted
Yaroslav Bulatov Oct 7
An overview of approaches to communication used in distributed Deep Learning
Reply Retweet Like
Peter Rudenko retweeted
Official ACM Sep 24
recipient David Patterson and software engineer Cliff Young discuss "A Domain-Specific Architecture for Deep Neural Networks," a contributed article in :
Reply Retweet Like
Peter Rudenko retweeted
Romain Ruetschi Oct 1
The Riemann Hypothesis, explained
Reply Retweet Like
Peter Rudenko retweeted
PVLDB Sep 26
Vol:11 No:13 → Everything You Always Wanted to Know About Compiled and Vectorized Queries But Were Afraid to Ask
Reply Retweet Like
Peter Rudenko retweeted
PFN-tech account Sep 27
We released the source code of PFN-Proto, a prototype AllReduce library we introduced in a research blog post. GitHub: Blog post:
Reply Retweet Like
Peter Rudenko Sep 25
The new H-series also will support SparkRDMA () for acceleration of the popular Spark analytics framework.
Reply Retweet Like
Peter Rudenko retweeted
Hot Chips 19 Jan 17
HC28 (2016) Videos Open to Public
Reply Retweet Like
Peter Rudenko retweeted
Claire Voltaire Jul 3
Replying to @Claire_Voltaire
In a democracy, it’s the people who make the country. In authoritarianism, it is the country that makes the people.
Reply Retweet Like
Peter Rudenko 6 Jun 18
Great talk of my colleague Yuval on running SparkRDMA ( - project we've been working on) on cloud.
Reply Retweet Like
Peter Rudenko retweeted
Nadav Rotem 21 May 18
I wrote a short post on register blocking in matrix multiplication.
Reply Retweet Like
Peter Rudenko retweeted
insideHPC.com 10 Apr 18
Video: Demystifying Parallel and Distributed Deep Learning
Reply Retweet Like
Peter Rudenko 20 Mar 18
combines recent advances in machine-learning systems and algorithms in a nested manner to reflect the hierarchical architecture of modern distributed systems.
Reply Retweet Like
Peter Rudenko retweeted
Miles Brundage 27 Feb 18
"Demystifying Parallel and Distributed Deep Learning: An In-Depth Concurrency Analysis," Ben-Nun and Hoefler:
Reply Retweet Like