Jayashree Mohan @jayashree2912

Researcher | Microsoft Research India microsoft.com/en-us/research… Bengaluru, India Joined November 2015

Tweets

76
Followers

671
Following

353
Likes

408

kwatra @kwatra

4 months ago

TokenWeave – Efficient Compute-Communication Overlap for Distributed LLM Inference. Why? Even with highspeed NVLink on H100 DGX, communication overhead for distributed LLM inference can be > 20 %! Can we recover this overhead? (1/10)

1 6 18 1K 4

Download Image

Jayashree Mohan @jayashree2912

a year ago

Check out this exciting recent work from our group that shows why legacy eval metrics like accuracy don't completely capture the compressed LLM model behavior and the metrics you should actually look out for! Read details here...

Abhinav Dutta @abhinavdutta555

a year ago

2 7 19 5K 16

0 0 10 2K 0

Jayashree Mohan @jayashree2912

a year ago

Conventional latency and throughput metrics are insufficient in capturing user-facing performance for interactive LLM applications. Introducing our effort : Metron, a benchmark with fluidity metric and fluidity token generation rate that captures user experience!

Amey Agrawal @agrawalamey12

a year ago

2 15 33 6K 14

1 0 11 2K 2

Alexey Tumanov @alsched

a year ago

[proud advisor moment] Happy to see @agrawalamey12 work in close collaboration with MSR-India already enjoying adoption and impact at @anyscalecompute! Please refer to Sarathi-Serve (arxiv.org/abs/2403.02310) for the better reference and coverage of the chunked prefill mechanism.

Robert Nishihara @robertnishihara

a year ago

2 14 89 15K 51

Download Image

0 2 21 1K 2

Jayashree Mohan @jayashree2912

a year ago

Talk to @agrawalamey12 or @nitinkedi at #osdi24 to learn more about our recent work Sarathi-Serve, a scheduler that addresses the latency-throughput tradeoff in LLM serving.

Amey Agrawal @agrawalamey12

a year ago

Talk to @agrawalamey12 or @nitinkedi at #osdi24 to learn more about our recent work Sarathi-Serve, a scheduler that addresses the latency-throughput tradeoff in LLM serving.

1 5 26 7K 12

0 0 11 2K 1

Jayashree Mohan @jayashree2912

2 years ago

Checkout our recent work on speeding up token generation for LLM inference. See the tweet deck below to learn more! Full paper is now up on arXiv - arxiv.org/pdf/2308.16369…

Amey Agrawal @agrawalamey12

2 years ago

Checkout our recent work on speeding up token generation for LLM inference. See the tweet deck below to learn more! Full paper is now up on arXiv - arxiv.org/pdf/2308.16369…

8 47 385 78K 321

0 0 18 3K 3

Jayashree Mohan @jayashree2912

3 years ago

NEW! Spring Deadline : Eurosys 2023 Consider submitting your work to the Spring deadline (May 11 - abstract, May 18- full paper)

ACM SIGOPS @ACMSIGOPS

3 years ago

NEW! Spring Deadline : Eurosys 2023 Consider submitting your work to the Spring deadline (May 11 - abstract, May 18- full paper)

0 8 21 0 0

0 3 18 0 0

Jayashree Mohan @jayashree2912

4 years ago

The @UTSASLab family 🤗 Successfully defended my PhD thesis today. Mixed feelings! Going to miss the enthusiasm and energy of this lab more than anything! Can’t believe time flew by in the blink of an eye! We missed you @SuprShastri !

5 0 94 0 1

Download Image

Jayashree Mohan @jayashree2912

4 years ago

Wohoo. Back after a long break on Twitter! You now know what I was upto 😅 The past 5 years have been a beautiful experience and it’s hard to bid goodbye. So fortunate to have a caring and supportive advisor @vj_chidambaram and group @UTSASLab

13 2 135 0 1

Jayashree Mohan @jayashree2912

5 years ago

I am excited to (virtually) present our work on fine-grained DNN checkpointing - CheckFreq at #FAST21 tomorrow! This is work done in collaboration with @MSFTResearch. Tune in to the Transactions & Deduplication session on Wed 24th, 8:30am PST to learn more about our work..

1 1 44 0 1

Download Image

jsysresearch @jsysresearch

5 years ago

The Student Editorial Board for @jsysresearch is coming together! Featuring @jayashree2912, @iangneal, @SolalPirelli, @Yizhou_Shan, @chinmay, @kkaffes, and more! escholarship.org/uc/jsys/studen…

0 8 26 0 2

Jayashree Mohan @jayashree2912

5 years ago

We were and are very excited by this work. This was long overdue! If you are interested in systems for ML, do give this a read :)

0 1 45 0 2

Jayashree Mohan @jayashree2912

5 years ago

Wrapping up the year on a good note. Stay tuned for the full version of the paper and presentation!

0 0 9 0 0

Jayashree Mohan @jayashree2912

6 years ago

Thank you @HeisenbugConf and @JUGru for an amazing edition of the conference in December! The video of my talk on "Testing file system crash-consistency bugs using CrashMonkey and Ace" is now online! youtu.be/BmhKbGoCyqo

0 3 40 0 3

Supreeth Shastri @SuprShastri

6 years ago

I am on the tenure-track job market! I work at the intersection of systems and disciplines such as economics (PhD) and law (post-doc). I would love to join a dept that values interdisciplinary work! Application material: cs.utexas.edu/~shastri/shast…

2 9 33 0 1

Jayashree Mohan @jayashree2912

6 years ago

I need a job that provides sponsored travel to exotic locations. My PhD life seems to be achieving this target really well.

2 2 55 0 0

Download Image

Jayashree Mohan @jayashree2912

6 years ago

This was a great experience! Such a well organized conference #heisenbug @HeisenbugConf

Heisenbug @HeisenbugConf

6 years ago

This was a great experience! Such a well organized conference #heisenbug @HeisenbugConf

0 0 5 0 0

Download Image

1 2 25 0 0

Jayashree Mohan @jayashree2912

6 years ago

All set for @HeisenbugConf !

0 0 23 0 1

Download Image

Jayashree Mohan @jayashree2912

6 years ago

As I update my CV, I realize that I've spoken about CrashMonkey at 6 different venues so far, including #osdi18 - and there's still no video recording of my talk. I hope at least my next venue @HeisenbugConf does it.