@mosharaf profile picture
Mosharaf Chowdhury @mosharaf

SymbioticLab (symbioticlab.org/)

Joined April 2008
  • Posts 1,210
  • Following 0
  • Followers 561
29 Photos and videos
Mosharaf Chowdhury reposted
Jae-Won Chung
 @jaewon_chung_cs
6 Jul 2023
Which modern LLM is the most energy efficient? ⚡️ Today, we're excited to release the first version of the ML ENERGY leaderboard, providing benchmarks of modern LLMs in terms of energy, model quality, and more! 🎉 ml.energy/leaderboard

The ML.ENERGY Leaderboard

How much time and energy do generative AI models consume?

ml.energy
1
5
17
3,608
Mosharaf Chowdhury reposted
Jae-Won Chung
 @jaewon_chung_cs
6 Jul 2023
The initial version was developed together with @JIACHENLIU8, Zhiyu Wu, and @mosharaf, part of @SymbioticLabUM / @UMichCSE. Also, we would like to thank @huggingface, @lmsysorg, @AiEleuther, and @Gradio for their awesome open source efforts. 🥰
1
3
772
Fantastic work by @jaewon_chung_cs et al. to understand and reduce the energy consumption of AI.
Jae-Won Chung
 @jaewon_chung_cs
18 Apr 2023
Care about Deep Learning energy consumption? Zeus is a practical energy optimization framework for DNN training. Already pulled 100k times on Docker Hub! ml.energy/zeus
1
9
1,555
Mosharaf Chowdhury reposted
Adrian Colyer @adriancolyer
5 May 2017
My pleasure - it seems such an obviously good idea! Very nice piece of work, congrats.
1
Thanks! We have many plans to extend it in coming months in terms of supporting different OS kernels, HA w/o performance drops, w/o RDMA etc x.com/eoinbrazil/status/8672…
Eoin Brazil @eoinbrazil
24 May 2017
Nice to see this work on GH github.com/Infiniswap/infini… HT @mosharaf x.com/adriancolyer/status/86…
Infiniswap is on github now! Let us know what you think!!! github.com/Infiniswap/infini…

GitHub - SymbioticLab/Infiniswap: Infiniswap enables unmodified applications to efficiently use...

Infiniswap enables unmodified applications to efficiently use disaggregated memory. - SymbioticLab/Infiniswap

github.com
1
1
Mosharaf Chowdhury reposted
Adrian Colyer @adriancolyer
5 May 2017
Why page to disk (slow!) when you could page to remote memory in the cluster using RDMA? blog.acolyer.org/2017/05/05/…
2
23
46
Mosharaf Chowdhury reposted
Adrian Colyer @adriancolyer
5 May 2017
"Efficient memory disaggregation with Infiniswap" Gu et al., NSDI'17 blog.acolyer.org/2017/05/05/… #themorningpaper
6
6
Mosharaf Chowdhury reposted
ACM SIGCOMM @ACMSIGCOMM
14 Oct 2016
Congratulations to @vyas_sekar (@CSDatCMU @cylab) for winning the SIGCOMM Rising Star Award 2016 sigcomm.org/awards/sigcomm-r…
2
12
15
Mosharaf Chowdhury reposted
ACM SIGCOMM @ACMSIGCOMM
25 Jun 2016
The winner of the 2015 Doctoral Dissertation Award is Mosharaf Chowdhury. sigcomm.org/awards/dissertat…
6
11
Mosharaf Chowdhury reposted
Databricks
 @databricks
17 Nov 2015
Learn about the Berkeley Data Analytics Stack in AMP Camp 6 this week! Attend in-person or watch the live stream: dbricks.co/1MkiG5h
16
9
Mosharaf Chowdhury reposted
AMPLab @amplab
12 Nov 2015
Join AMPCamp 6 on Nov 19-20 to learn Spark and the Berkeley Data Analytics Stack, at UC Berkeley or via live stream ampcamp.berkeley.edu/6
9
11
Mosharaf Chowdhury reposted
billions of packets @justinesherry
10 Nov 2015
Applying to grad school in networks? @mosharaf is looking for students! mosharaf.com/blog/2015/10/26…
1
Mosharaf Chowdhury reposted
AMPLab @amplab
19 Aug 2015
Mesos making news and vying for "Unicorn" status - amplab.cs.berkeley.edu/news/…
4
7
Mosharaf Chowdhury reposted
Matei Zaharia @matei_zaharia
11 Aug 2015
Teaching or doing research with Spark? @databricks just launched a free academic program: databricks.com/blog/2015/08/…

Get Started With Databricks - Try Databricks for Free

Start your free Databricks trial today. Get credits to build data pipelines, ML models, and AI applications. Compare our trial vs. free community edition options.

databricks.com
39
37
Be well-balanced x.com/Frankmcsherry/status/6…
Thanks @__apf__
1
Mosharaf Chowdhury reposted
Tachyon Nexus @TachyonNexus
7 May 2015
We are hiring! Join our team from UC Berkeley, Google, CMU, VMWare, Stanford etc. to create the future of storage! tachyonnexus.com

Alluxio | Home

Alluxio is a high-performance distributed caching layer built for large-scale AI workloads. Alluxio maximizes I/O throughput and minimizes I/O latency for model training, model deployment, inference...

alluxio.io
2
25
11
Congrats! @jrexnet @michaelfreedman
Load more