home.social

#parallelism — Public Fediverse posts

Live and recent posts from across the Fediverse tagged #parallelism, aggregated by home.social.

fetched live
  1. Join us in Vienna for IWOMP 2026!

    Join us October 7–9, 2026 for the 22nd annual IWOMP event in Austria. IWOMP is the annual workshop dedicated to the advancement of all aspects of parallel programming with OpenMP. It is the premier forum to present issues, trends, recent research ideas, and results related to parallel programming.

    August 14th is the last day with the early-bird pricing -- full pricing starts on August 15

    iwomp.org/registration/
    #OpenMP #concurrence #parallelism #HPC

  2. Join us in Vienna for IWOMP 2026!

    Join us October 7–9, 2026 for the 22nd annual IWOMP event in Austria. IWOMP is the annual workshop dedicated to the advancement of all aspects of parallel programming with OpenMP. It is the premier forum to present issues, trends, recent research ideas, and results related to parallel programming.

    August 14th is the last day with the early-bird pricing -- full pricing starts on August 15

    iwomp.org/registration/
    #OpenMP #concurrence #parallelism #HPC

  3. Двигаем PostgreSQL в сторону OLAP: оптимизация параллельного вычисления агрегатов

    В данной статье я хочу рассказать о проверке одной гипотезы - возможности использования shared memory для ускорения параллельной агрегации методом хэширования. Статья Xu&Marcus, PVLDB, 2025 утверждает, что общая хэш-таблица — незаслуженно списанный со счетов способ параллельной агрегации, если использовать т.н. тикетинг и разделить операции поиска группы и обновления агрегата. Звучит завлекательно. Собираем в кучу свои знания Internals, подписку на Claude и, поскольку в период летних Heat Wave выходные всё-равно проходят дома под кондиционером - разрабатываем идею так глубоко, насколько это получится.

    habr.com/ru/companies/tantor/a

    #postgresql #shared_memory #aggregation #parallelism #performance #query_execution

  4. "Unlock the GPU as a general-purpose engine, not just a rasterizer."

    This series takes you past `vkCmdDispatch` and into how compute actually executes on real hardware — occupancy, latency hiding, the Vulkan memory model, and subgroup operations that let invocations talk to each other without touching global memory.

    Learn more: docs.vulkan.org/tutorial/lates

  5. "Unlock the GPU as a general-purpose engine, not just a rasterizer."

    This series takes you past `vkCmdDispatch` and into how compute actually executes on real hardware — occupancy, latency hiding, the Vulkan memory model, and subgroup operations that let invocations talk to each other without touching global memory.

    Learn more: docs.vulkan.org/tutorial/lates
    #compute #parallelism #HPC #Vulkan #Khronos #programming #AI #GPU

  6. Nebenläufige Programmierung ist ein Begriff, der in der Softwareentwicklung häufig fällt, aber oft unterschiedlich interpretiert wird. Grundsätzlich beschreibt er die Fähigkeit eines Programms, mehrere Aufgaben gleichzeitig oder scheinbar gleichzeitig a...

    magicmarcy.de/nebenlaeufige-pr

    #Nebenläufige_Programmierung #Softwareentwicklung #Aufgaben #gleichzeitig #Nebenläufigkeit #Concurrency #Parallelität #Parallelism #sequentielle_Verarbeitung #Ausführungsreihenfolge

  7. Nebenläufige Programmierung ist ein Begriff, der in der Softwareentwicklung häufig fällt, aber oft unterschiedlich interpretiert wird. Grundsätzlich beschreibt er die Fähigkeit eines Programms, mehrere Aufgaben gleichzeitig oder scheinbar gleichzeitig a...

    magicmarcy.de/nebenlaeufige-pr

    #Nebenläufige_Programmierung #Softwareentwicklung #Aufgaben #gleichzeitig #Nebenläufigkeit #Concurrency #Parallelität #Parallelism #sequentielle_Verarbeitung #Ausführungsreihenfolge

  8. [Перевод] Разница между параллельными и распределёнными вычислениями

    Параллельные и распределённые вычисления часто ставят рядом, но это далеко не одно и то же. В новом переводе от команды Spring АйО разберем, как устроены обе модели, чем отличаются их архитектура, способы обмена данными, масштабируемость и отказоустойчивость. Статья подойдет тем, кто хочет понять, когда достаточно ресурсов одной машины, а когда без сети из нескольких узлов уже не обойтись.

    habr.com/ru/companies/spring_a

    #system_design #consistency #distributed_computing #distributed_systems #distributed #parallels #parallelism #parallel_computing #spring #spring_boot

  9. Пул интерпретаторов в Python 3.14. Что, зачем и почему?

    Как все знают, GIL (Global Interpreter Lock) не позволяет нескольким потокам CPython выполнять CPU-bound задачи параллельно. Глобальная блокировка интерпретатора предоставляет каждому потоку лишь небольшой интервал времени для работы. При этом планирование работы потоков (какому именно потоку из ожидающих предоставить разрешение на выполнение) осуществляется планировщиком операционной системы. Интерпретатор не является полноценным планировщиком работы потоков, он делегирует эту функцию операционной системе. GIL использует мьютексы ОС для блокировки работы потоков так, чтобы в один момент времени мог выполняться только один поток из нескольких.

    habr.com/ru/articles/957058/

    #Python_314 #parallelism #multithreading

  10. Introducing the most riveting tale of all time: the #love #story between a #GPU and its #graphics, sprinkled with just enough #Triton #jargon to make you nod off faster than a PyTorch Profiler. We've got #parallelism, pheromones, and more tangents than a high school geometry class 💤. Pack your bags, folks, because we're going on an #adventure through a sea of terrifying colors and kernels that nobody asked for! 🚀🌈
    ut21.github.io/blog/triton.html #HackerNews #ngated

  11. Introducing the most riveting tale of all time: the #love #story between a #GPU and its #graphics, sprinkled with just enough #Triton #jargon to make you nod off faster than a PyTorch Profiler. We've got #parallelism, pheromones, and more tangents than a high school geometry class 💤. Pack your bags, folks, because we're going on an #adventure through a sea of terrifying colors and kernels that nobody asked for! 🚀🌈
    ut21.github.io/blog/triton.html #HackerNews #ngated

  12. Arendt et al. discuss several metrics of phenotypic parallel evolution including one that does not actually measure parallelism. They suggest use of a partial Eta-squared of the interaction term between Habitat and Replicate as an appropriate metric.

    Read now ahead of print!
    journals.uchicago.edu/doi/10.1

    #evolution #parallelism #habitat #phenotypes #parallelEvolution

  13. Arendt et al. discuss several metrics of phenotypic parallel evolution including one that does not actually measure parallelism. They suggest use of a partial Eta-squared of the interaction term between Habitat and Replicate as an appropriate metric.

    Read now ahead of print!
    journals.uchicago.edu/doi/10.1

    #evolution #parallelism #habitat #phenotypes #parallelEvolution

  14. Concurrency Concept: Say you have two threads trying to access the same bytes.
    Instead of a lock, both threads can reserve a specific time interval, in which they are allowed to access the memory area. So the access timeline would look something like this (t1 and t2 being the timeframe's the respective threads can access it): t1 | t2 | t1 | t2 | ...

    #programming #concurrency #parallel #parallelism

  15. Concurrency Concept: Say you have two threads trying to access the same bytes.
    Instead of a lock, both threads can reserve a specific time interval, in which they are allowed to access the memory area. So the access timeline would look something like this (t1 and t2 being the timeframe's the respective threads can access it): t1 | t2 | t1 | t2 | ...

    #programming #concurrency #parallel #parallelism

  16. kitfucoda.medium.com/concurren

    Concurrency and parallelism are often confused in async programming discussions. Go's goroutines highlighted the difference: concurrency is doing many things at once, while parallelism is doing many things at the same time.

    AsyncIO handles concurrency well for I/O, but CPU-bound tasks need parallelism. Python uses AsyncIO for concurrency, and ProcessPoolExecutor for parallelism, distributing work across CPU cores.

    Process communication is harder than thread communication. AsyncIO's task cancellation differs from ProcessPoolExecutor's, requiring workarounds like event objects for reliable cancellation and shutdown.

    Essentially, ProcessPoolExecutor enables parallelism for CPU-bound tasks, scaling them across multiple cores, while AsyncIO handles I/O concurrently.

    #python #asyncio #concurrency #parallelism #multiprocessing #opentowork #getfedihired #fedihire #opentowork

  17. kitfucoda.medium.com/concurren

    Concurrency and parallelism are often confused in async programming discussions. Go's goroutines highlighted the difference: concurrency is doing many things at once, while parallelism is doing many things at the same time.

    AsyncIO handles concurrency well for I/O, but CPU-bound tasks need parallelism. Python uses AsyncIO for concurrency, and ProcessPoolExecutor for parallelism, distributing work across CPU cores.

    Process communication is harder than thread communication. AsyncIO's task cancellation differs from ProcessPoolExecutor's, requiring workarounds like event objects for reliable cancellation and shutdown.

    Essentially, ProcessPoolExecutor enables parallelism for CPU-bound tasks, scaling them across multiple cores, while AsyncIO handles I/O concurrently.

    #python #asyncio #concurrency #parallelism #multiprocessing #opentowork #getfedihired #fedihire #opentowork

  18. 🏃‍♀️ High-performance computing, with much less code

    (... back in my business development days I was interested in "rules and schemes" that would separate what is to be done with how it should be done)

    news.mit.edu/2025/high-perform

    #computing #software #programming #languages #hpc #parallelism

  19. 🏃‍♀️ High-performance computing, with much less code

    (... back in my business development days I was interested in "rules and schemes" that would separate what is to be done with how it should be done)

    news.mit.edu/2025/high-perform

    #computing #software #programming #languages #hpc #parallelism

  20. Параллелизм в Go тестах: все, что нужно знать о -p, -parallel и t.Parallel()

    В Go есть три способа управления параллельностью тестов: Короткий гайд о -p , -parallel и t.Parallel а также бонус для любителей параллельного программирования

    habr.com/ru/articles/882222/

    #go #golang #testing #concurrency #parallelism #neva #nevalang #programming_languages #visual_programming #compiler

  21. uses the incredible GIF image (lnkd.in/gT2Q9Wyq) to pertinently and vividly describe what's the potential and essential differences between CONCURRENCY and PARALLELISM … I've just decorated it to become a fabulous MP4 file, which has been integrated the special effects and melodious music … 😉

  22. I used to love #Python, but dealing with the honestly kind of scary #multiprocessing library was a reminder of the #GIL #threading situation really hurts working with #parallelism. 😰🫠

    I've started learning #Kotlin after honestly really enjoying the threading library it provides. It's so's easy to work with when you understand it! 😀 (I would prefer to stay away from some of the #Java conventions after working with them for so long. 👍) #JVM #programming
    f.kawa-kun.com/display/881761a…