@YuxuanZhang_ Unfortunate I did not see this earlier today - I was in SVL this week (and ran into Steve in MAT3) but am now on the way to the airport.
English
Thomas Wenisch
66 posts
@estark37 Fwiw the process you describe heads in the direction of how journals work - no meeting and one associate editor decides the outcome based on reading the reviews.
English
Program committees (PCs) are the mysterious cabals that review conference papers and decide what to accept/reject. I'm on a hiatus from them, but any PC chair who addresses my complaints might lure me back one of these days... emilymstark.com/2023/09/08/com…
English
@justinesherry One of the folks that reports to me at Google was the original implementer of the first generation of GPUDirect at NVIDIA ~10 years ago. As long as your questions don’t ask anything too proprietary it’s plausible I can help.
English
Do I know anyone who has worked with NVIDIA GPUDirect? Can I talk to you about it?
English
Thomas Wenisch retweetledi
Congratulations to all of the @ACMSIGOPS dissertation award winners!! I enjoyed reading every thesis and it was a very hard choice! sigops.org/2022/dmr-2022/
English
@Soaryn117 @factoriogame Pyanodon’s mods. Especially with Alien Life. It’s the GregTech of Factorio.
English
Thomas Wenisch retweetledi
cloud.google.com/blog/topics/sy…
Pleased to announce sharing of Google workload traces for the broader systems/architecture community as we all work together on extending Moore's Law in new ways. #Google #GoogleCloud @sigarch
English
@achowdhery @sharan0909 @nfiedel @JeffDean @m_isard @ada_rob @bsaeta Congrats @achowdhery and team! So pleased to work with you to move giant tensors around Google’s datacenter networks.
English
Really excited to present the first large-scale use of Pathways system! Joint work with so many of colleagues at Google! @sharan0909 @nfiedel @JeffDean @m_isard @ada_rob @bsaeta .
Google AI@GoogleAI
Introducing the 540 billion parameter Pathways Language Model. Trained on two Cloud #TPU v4 pods, it achieves state-of-the-art performance on benchmarks and shows exciting capabilities like mathematical reasoning, code writing, and even explaining jokes. goo.gle/3j6eMnK
English
Thomas Wenisch retweetledi
New! An intro to emerging persistent memory (PM) technologies that promise the performance of dynamic random-access memory #DRAM with the durability of traditional #storage #media, such as #hard #disks and solid-state drives #SSD . bit.ly/3JfFi9q #persistentmemory
English
Thomas Wenisch retweetledi
The wifi went out for 90 minutes. You would think that we had stepped back into pioneer times. Luckily, no one got dysentery.
GIF
English
@danluu I’m pretty sure Hank Levy created the content for this flyer himself (i.e., not marketing). Disappointed he didn’t include ASPLOS (which I am co-PC chairing with Shan Lu this year.)
English
I find it sort of astounding how, 17 years after Steve Yegge published sites.google.com/site/steveyegg…, almost no companies "get it" when it comes to marketing the company to potential hires, e.g., SOSP flyers:
web.archive.org/web/2021102308…
web.archive.org/web/2021102308…
web.archive.org/web/2021102308…
English
@Foone I am deeply honored to share a birthday with foone. I will pour some floppies on the curb in their honor.
English
@jonmasters Calxeda was a couple years too early. We really needed 64-but ARM before servers were truly viable, else the software enablement story was just too complicated. But the consequence is that I get to work with some awesome calxeda alumni today!
English
Arm servers have always been a giant game of Go. You will lose a lot of (smooth) stones along the way. But eventually, you’ll have a couple eyes (eg Apple, cloud), and suddenly a lot of connected stones will appear 😀
New Castle, NH 🇺🇸 English
I’ve been a believer in @Arm servers for a very, very long time. It was always going to take a decade to build the ecosystem. But now we are going to upend the entire world ❤️
OGAWA, Tadashi@ogawa_tter
=> "World's First Bicycle Powered ARM Server", Jun 27 2012 web.archive.org/web/2012063012… @jonmasters at RedHat Summit '12 Distributed Mandelbrot 32 cores (8 Calxeda EnergyCores) twitter.com/jonmasters/sta… 648x MIPS64s (SiCortex SC648) by ten UCLA Cycling team, 2007 youtube.com/watch?v=DdLCIz…
New Castle, NH 🇺🇸 English
My thanks to the many collaborators who worked with me on the research recognized by the 2021 ACM SIGARCH Maurice Wilkes award. My message of thanks can be viewed here: youtube.com/watch?v=9t0Gyw…
YouTube
English
Thomas Wenisch retweetledi
The Maurice Wilkes award for mid-career faculty goes this year to @ThomasWenisch! “For contributions to memory persistency and energy-efficient systems.” #isca2021
English
Wow, Germany's choice of a costume that's just giving the finger all song is definitely something...
English
English
And just a reminder that my HotCRP now will forever show me this crucial information! lol @xexd
Alexandra Silva@alexandra8silva
Really excited to have my first @ACMSIGCOMM w @tiferrei2000, @LikelyHarrison, and @lorisdanto! Real fun project, during my sabbatical at @madpl1 hosted by the cool @lorisdanto! Really proud of @tiferrei2000, 1st year UG @uclcs, and behind a lot of the cool ideas in the paper! 1/3
English
Thanks @NSF for funding our project on correctly rounded math libraries! nsf.gov/awardsearch/sh…
English
@ratulm @AkshithaSriram1 I didn’t realize University of Wisconsin - Steven’s Point was a top contender!
English
@AkshithaSriram1 It's easy. Is the university in a state that begins with W and in a city that begins with S? That's what I did.
English
As I move on to the hard part of trying to make a decision, I’m curious what factors you considered when you were trying to decide between your options. What parameters made you finally decide on your university?
English
@JAldrichPL This issue is such a no-brainer. Conference and journal organizers should require reviewers to grant irrevocably to authors permission to use the entirety of the author-visible portion of their review for any purpose.
English
This week, the ACM pubs board will discuss the recent SIGCOMM CARES / HotNets 2020 review controversy and in particular the ACM policy on reviews.
Question: what portion of a review should authors be able to publicly share?
English