10 ms·

Elegant High Performance Computing # Performance [benchmarks?] Coming soon! I've seen quite a few elegant and within-an-order-of-magnitude-of-semi-optimiz

Elegant High Performance Computing # Performance [benchmarks?] Coming soon! I've seen quite a few elegant and within-an-order-of-magnitude-of-semi-optimized-C CL programs throughout the years, but can one make such a statement without benchmarks? What do we mean by high performance?

thumbuddy 3y ago
I also share this concern. I hope it at least beats out naive implementations of these algorithms, but I wouldn't be surprised if it was far from SOTA or standard tooling.
jackdaniel 3y ago
While I'm not using Petalisp I can attest that Marco is known for delivering high quality software with a strong focus on performance. I suspect that the phrase "Coming soon!" means exactly that - when he'll make them they will be published.
- norir 3y ago
  Regardless of their skill, they only get one chance to make a first impression.
  - CodeSgt 3y ago
    We don’t know if they chose to share it on HN just yet or if this post was made by a fan of the project.
    - phoe-krk 3y ago
      A fan, the author of Petalisp is https://news.ycombinator.com/user?id=heisig https://news.ycombinator.com/user?id=heisig
  - jackdaniel 3y ago
    I don't know what do you try to convey. The link is shared by a third-party.
distcs 3y ago
> I've seen quite a few elegant and within-an-order-of-magnitude-of-semi-optimized-C CL programs throughout the years, but can one make such a statement without benchmarks? What do we mean by high performance? This is someone's project that is still in development. The person who posted this on HN and the person who created this project are not same. The benchmarks will become available when they will become available.
MarioMan 3y ago
In this context, High Performance Computing (HPC) refers to a computing paradigm where high-performance computers work on a distributed problem (e.g., supercomputers), not the actual performance of this code generator. All it means here is that it generates parallel, distributed code intended for use by HPC platforms. See: https://en.wikipedia.org/wiki/High-performance_computing https://en.wikipedia.org/wiki/High-performance_computing
heisig 3y ago
Petalisp author here. I apologize that the README is somewhat lacking, but it wasn't me who posted this on HN. As you may have seen on the commit history, a lot of exciting things have happened over the last few months. However, there are also a few stupid performance bugs left, so I am delaying the release of any performance numbers until those have been fixed. Otherwise, I fear that people will simply misinterpret the results. Nevertheless, I can already state that the single-core performance of Petalisp programs is exactly like that of a C program - simply because Petalisp compiles its programs to C when possible. (Although I also have that long-term agenda of using sb-simd to reach that performance in pure CL some day.) In addition, Petalisp is quite good at automatically parallelizing programs, and we already have most of the infrastructure for distributed and heterogeneous computing in place. I will write a more detailed post for the HN crowd once I have reliable performance numbers and once I finished writing the documentation. Feel free to ask me further questions.
- medo-bear 3y ago
  This would be an exciting read. Im cretainly looking forward to it. Is there a read me for the required infrastructure setup in the works?
  - heisig 3y ago
    I am not sure what you mean by "required infrastructure setup". Installing Petalisp is a single call to (ql:quickload :petalisp) - assuming you have Quicklisp installed. If you also have a C compiler available, and an executable named cc pointing to that compiler, Petalisp will use that to speed up your codes further. About the required hardware - anything that runs SBCL or CCL can also run Petalisp.
    - medo-bear 3y ago
      Sorry I should have been more specific. I meant a distributed computing system setup
      - 3y ago

thumbuddy 3y ago

I also share this concern. I hope it at least beats out naive implementations of these algorithms, but I wouldn't be surprised if it was far from SOTA or standard tooling.

jackdaniel 3y ago

While I'm not using Petalisp I can attest that Marco is known for delivering high quality software with a strong focus on performance. I suspect that the phrase "Coming soon!" means exactly that - when he'll make them they will be published.

norir 3y ago
Regardless of their skill, they only get one chance to make a first impression.
- CodeSgt 3y ago
  We don’t know if they chose to share it on HN just yet or if this post was made by a fan of the project.
  - phoe-krk 3y ago
    A fan, the author of Petalisp is https://news.ycombinator.com/user?id=heisig https://news.ycombinator.com/user?id=heisig
- jackdaniel 3y ago
  I don't know what do you try to convey. The link is shared by a third-party.

distcs 3y ago

> I've seen quite a few elegant and within-an-order-of-magnitude-of-semi-optimized-C CL programs throughout the years, but can one make such a statement without benchmarks? What do we mean by high performance? This is someone's project that is still in development. The person who posted this on HN and the person who created this project are not same. The benchmarks will become available when they will become available.

MarioMan 3y ago

In this context, High Performance Computing (HPC) refers to a computing paradigm where high-performance computers work on a distributed problem (e.g., supercomputers), not the actual performance of this code generator. All it means here is that it generates parallel, distributed code intended for use by HPC platforms. See: https://en.wikipedia.org/wiki/High-performance_computing https://en.wikipedia.org/wiki/High-performance_computing

heisig 3y ago

Petalisp author here. I apologize that the README is somewhat lacking, but it wasn't me who posted this on HN. As you may have seen on the commit history, a lot of exciting things have happened over the last few months. However, there are also a few stupid performance bugs left, so I am delaying the release of any performance numbers until those have been fixed. Otherwise, I fear that people will simply misinterpret the results. Nevertheless, I can already state that the single-core performance of Petalisp programs is exactly like that of a C program - simply because Petalisp compiles its programs to C when possible. (Although I also have that long-term agenda of using sb-simd to reach that performance in pure CL some day.) In addition, Petalisp is quite good at automatically parallelizing programs, and we already have most of the infrastructure for distributed and heterogeneous computing in place. I will write a more detailed post for the HN crowd once I have reliable performance numbers and once I finished writing the documentation. Feel free to ask me further questions.

medo-bear 3y ago
This would be an exciting read. Im cretainly looking forward to it. Is there a read me for the required infrastructure setup in the works?
- heisig 3y ago
  I am not sure what you mean by "required infrastructure setup". Installing Petalisp is a single call to (ql:quickload :petalisp) - assuming you have Quicklisp installed. If you also have a C compiler available, and an executable named cc pointing to that compiler, Petalisp will use that to speed up your codes further. About the required hardware - anything that runs SBCL or CCL can also run Petalisp.
  - medo-bear 3y ago
    Sorry I should have been more specific. I meant a distributed computing system setup
    - 3y ago