SIGARCH Blog

The article mentions a student taking his own life. I remember this was huge news a couple years ago at ISCA '19 and something that really shook my decision to pursue academia. After that event, SIGARCH launched an investigation. After a couple years, here were the results of that investigation. https://www.sigarch.org/other-announcements/isca-19-joint-investigativ… Worth noting is that the investigation actually initially found __no__ misconduct. Imagine that? A student kills himself, and you conclude it was the victim's fault, and not the environment that drove him. It was only until this post [1] emerged that they relaunched the investigation. [1] https://huixiangvoice.medium.com/evidence-put-doubts-on-the-ieee-acms-…

I just wanna leave here some classic papers on sexism and harassment in tech, which are relevant even now: [1] Why are There so Few Female Computer Scientists (Ellen Spertus, 1991) https://dspace.mit.edu/handle/1721.1/7040 [2] How to Encourage Women in Linux (Val Henson, 2002) https://tldp.org/HOWTO/Encourage-Women-Linux-HOWTO/ [3] What Happens to Us Does Not Happen to Most of You (Kathryn S. McKinley, 2018) https://www.sigarch.org/what-happens-to-us-does-not-happen-to-most-of-… [4] Unlocking the Clubhouse: Women in Computing (Jane Margolis and Allan Fisher, 2001) https://www.amazon.com/Unlocking-Clubhouse-Women-Computing-Press/dp/02…

It's sad to see no end to this kind of thinking. Here's some easy to read classic papers on sexism in tech: [1] Why are There so Few Female Computer Scientists (Ellen Spertus, 1991) https://dspace.mit.edu/handle/1721.1/7040 [2] How to Encourage Women in Linux (Val Henson, 2002) https://tldp.org/HOWTO/Encourage-Women-Linux-HOWTO/ [3] What Happens to Us Does Not Happen to Most of You (Kathryn S. McKinley, 2018) https://www.sigarch.org/what-happens-to-us-does-not-happen-to-most-of-… [4] Unlocking the Clubhouse: Women in Computing (Jane Margolis and Allan Fisher, 2001) https://www.amazon.com/Unlocking-Clubhouse-Women-Computing-Press/dp/02…

The RISC V V-extension follows a Cray-style model, and builds on the work with Hwacha to do mixed-precision vectors much more cleanly: https://pdfs.semanticscholar.org/b9e8/fcf11b662a31cecd5a08d5e022dcae77… https://www.sigarch.org/simd-instructions-considered-harmful/

I'm gonna throw some cold water on this and say this is not a new paradigm by any means. https://www.sigarch.org/the-unreasonable-ineffectiveness-of-machine-le… However, it's certainly possible that the time for this idea has come. Google is probably in the best position to apply it. I will say that after having worked at Google for over a decade, some of it on data center performance, there are plenty of inefficiencies that have nothing to do with algorithms or code, but have more to do with very coarse-grained resource allocation. In other words, they are "boring management problems" rather than sexy CS problems. I should also add that you're expanding the claims made in the paper, which appears to be a common mistake. If you look at the related work and citations, you'll see that it's mostly about B-Trees, bloom filters, and indices. But I have seen people generalizing this to "a revolution in systems research", which I don't believe is justified. Indexed data structures are important but a revolution in them is not going to cause a revolution in all of systems. Off the top of my head, the general area of "scheduling" (in time and space) is probably more important and more potentially impactful.

I think it's a good thing to not bake IEEE 754 too deeply into RISC-V. I really want to see hardware built around posits. This is not because they're necessarily superior to floats (they aren't in all use cases), but just because we need some diversity. Too much standardization is bad for innovation, and not everything that's settled should remain that way. https://www.sigarch.org/posit-a-potential-replacement-for-ieee-754/

"A Brief Guide of xPU for AI Accelerators" https://www.sigarch.org/a-brief-guide-of-xpu-for-ai-accelerators/ NPU: Neural Processing Unit: https://en.wikipedia.org/wiki/Neural_processing_unit Coprocessor: https://en.wikipedia.org/wiki/Coprocessor

Wow, that number requires STRONG caveats, lest it be called out as completely false. Take away the tensor cores (unless you only do matmuls?), and an H100 has roughly 2x as many f32 flops as a Zen5 CPU, which is considerably cheaper. I suspect brute force HW/algorithms are not going to age well: https://www.sigarch.org/dont-put-all-your-tensors-in-one-basket-hardwa… (/personal opinion)

The first link in the article is specifically about the fact that men don't experience and aren't aware of how common these issues are for women. Its first anecdote is from a female assistant professor, who is probably similar to the female lecturers you think no one treated differently. https://www.sigarch.org/what-happens-to-us-does-not-happen-to-most-of-…

Probably referring to library operating systems and making a unikernel instead of a shared kernel container. https://www.sigarch.org/leave-your-os-at-home-the-rise-of-library-oper…

ACM article: https://www.sigarch.org/posit-a-potential-replacement-for-ieee-754/ Original paper: https://dl.acm.org/doi/10.14529/jsfi170206

Part 2: https://www.sigarch.org/a-brief-and-biased-history-of-computer-archite…

> Seems like a mess of a function It has to deal with an arbitrary n, so it's going to have some messy bits to deal with the "fringe": https://www.sigarch.org/simd-instructions-considered-harmful/

Disappointing that ACM (and IEEE presumably) shut down the initial investigation [1] after finding no "tangible evidence of wrongdoing" and dismissing "rumors shared by anonymous sources" and only opened it again [2] in the face of extraordinary and overwhelming specific evidence [3] that was impossible to ignore. [1]: https://www.sigarch.org/wp-content/uploads/2021/02/JIC-Public-Announce… [2]: https://www.sigarch.org/questions-about-policies-processes-in-the-wake… [3]: https://huixiangvoice.medium.com

This is such a sad story. Nothing can bring back the dead, but we can try to prevent a repeat of the events, and ensure that their death was not in vain. It is heartening to see that ACM took the investigation seriously, and actually did discover and condemn serious academic fraud. However, while I can guess who is has been banned from submitting to ACM conferences for the next 15 years, apparently it is ACMs policy not to announce publicly! Just look at the report[0], it's completely anonymous. Am I mistaken? What is the justification for this? At a practical level, how am I as a co-chair of an ACM event supposed to enforce a ban if I don't know who is banned? [0]: https://www.sigarch.org/wp-content/uploads/2021/02/JIC-Public-Announce…

Did he also design AArch64? Because I've noticed it can be used to design very fast processors, yet abandons most decisions in ARM7 and Thumb. I was just curious about how SIMD worked in RISC-V so I looked it up… and apparently RISC-V doesn't have SIMD, instead it brings back microcoded large vector operations from the 70s, and the RISC-V designers called SIMD bad and compared it to the opioid crisis because it means there are too many instruction opcodes? https://www.sigarch.org/simd-instructions-considered-harmful/ So I'm now even less impressed and also slightly offended. This is not realistic, short vectors are important and easy to do in hardware.

for the curious https://www.sigarch.org/posit-a-potential-replacement-for-ieee-754/

You should report it to SIGARCH executive. More information on this blog post https://www.sigarch.org/other-announcements/joint-investigative-commit…

There seem to be a few indications that there is opportunity for change, and RISC-V could make gains, eg https://www.nextplatform.com/2019/02/05/the-era-of-general-purpose-com… https://www.sigarch.org/simd-instructions-considered-harmful/

> I would say that, if you make such a claim, you should state why it is a mistake and what should be done instead. Others have explained the problem with the von Neumann bottleneck,[1][2] and ways to overcome it or even redefine it.[3] The industry is evolving towards programming paradigms less affected by that original computation model such as functional, functional-reactive, or agent-based programming(*), yet most of those are still converted down to Von-Neumann-style runtimes to be processed. Now that machine learning is increasingly being used to solve new problems, many computations may be done with dedicated non-VN hardware like tensor processors or analog computers (which are coming back with a vengeance). (*) Listed in increasing order of real-world expressivity, although their computational expressivity in all them is mathematically equivalent to Brainfuck. [1] https://dl.acm.org/doi/10.1145/1283920.1283933 [2] https://www.techtarget.com/whatis/definition/von-Neumann-bottleneck [3] https://www.sigarch.org/the-von-neumann-bottleneck-revisited/\*

Seems like RMS's Achilles' heel is harassment towards women, over the years, even after being made aware of it politely by others. In that light, these are classic reads on the topic (copy pasting another comment): [1] Why are There so Few Female Computer Scientists (Ellen Spertus, 1991) https://dspace.mit.edu/handle/1721.1/7040 [2] How to Encourage Women in Linux (Val Henson, 2002) https://tldp.org/HOWTO/Encourage-Women-Linux-HOWTO/ [3] What Happens to Us Does Not Happen to Most of You (Kathryn S. McKinley, 2018) https://www.sigarch.org/what-happens-to-us-does-not-happen-to-most-of-… [4] Unlocking the Clubhouse: Women in Computing (Jane Margolis and Allan Fisher, 2001) https://www.amazon.com/Unlocking-Clubhouse-Women-Computing-Press/dp/02… [5] The Elephant in the Valley (by Michele Madansky and Trae Vassallo, 2015) https://www.elephantinthevalley.com/ Also the news of RMS coming back seems to have been hidden from even FSF members. https://twitter.com/fsf/status/1374399897558917128

Here's some readings on why many don't want RMS in a leadership position. [1] https://news.ycombinator.com/item?id=21287006 [2] https://news.ycombinator.com/item?id=20994216 [3] https://news.ycombinator.com/item?id=26545420 [4] https://news.ycombinator.com/item?id=26535789 [5] https://guix.gnu.org/blog/2019/joint-statement-on-the-gnu-project/ . . Some well-regarded papers on sexism in CS, which are still relevant as is evident from some comments in this thread. [1] Why are There so Few Female Computer Scientists (Ellen Spertus, 1991) https://dspace.mit.edu/handle/1721.1/7040 [2] How to Encourage Women in Linux (Val Henson, 2002) https://tldp.org/HOWTO/Encourage-Women-Linux-HOWTO/ [3] What Happens to Us Does Not Happen to Most of You (Kathryn S. McKinley, 2018) https://www.sigarch.org/what-happens-to-us-does-not-happen-to-most-of-… [4] Unlocking the Clubhouse: Women in Computing (Jane Margolis and Allan Fisher, 2001) https://www.amazon.com/Unlocking-Clubhouse-Women-Computing-Press/dp/02… [5] The Elephant in the Valley (by Michele Madansky and Trae Vassallo, 2015) https://www.elephantinthevalley.com/