Показаны сообщения с ярлыком Concurrency. Показать все сообщения
Показаны сообщения с ярлыком Concurrency. Показать все сообщения

пятница, 1 апреля 2011 г.

checkdeadlock.org - первый релиз

В продолжении к предыдущему посту. Т.к. домашний компьютер постоянно включенным оказалось очень неудобно, разместил на hosting.ua проект checkdeadlock.org. На этом хостинге нет контейнера сервлетов, поэтому пришлось написать свой небольшой сервер, который работает на 8080 порту (т.к. я не root, то доступ к 80 мне пока закрыт. Нужно будет поговорить по этому поводу с саппортом. Но пока и так неплохо).

суббота, 26 марта 2011 г.

Нахождение взаимных блокировок потоков при статическом анализе кода

Всем доброго вечера.
Когда Иван читал у нас, на мех-мате в Каразина свой спецкурс у меня появилась навящевая идея по поиску дедлоков без запуска программы, только на основании ее кода. Было сделано несколько попыток найти какой-то внятный алгоритм. К настоящему моменту я достаточно неплохо, как мне кажется, продвинулся в этом направлении.
Чтобы не усложнять до предела работу по парсингу кода на Java или C я написал простой язык для описания потоков.

четверг, 15 октября 2009 г.

JAOO: Data Parallel Haskell

Слайды с презентации Simon Peyton-Jones "Data Parallel Haskell" на JAOO Conference:
http://jaoo.dk/aarhus-2009/file?path=/jaoo-aarhus-2009/slides/SimonPeyton-Jones_DataParallelHaskell.pdf

Target audience: Developers who want to know what is just "over the horizon" for parallel programming.

Abstract:

There are many approaches to exploiting multi-cores, but a particularly promising one is the "data-parallel" paradigm, because it combines massive parallelism (on both shared and distributed memory) with a simple, single control-flow programming model. Indeed, I think that data parallelism is the only way we will be able to exploit tens or hundreds of processors effectively.

Alas, data-parallel programming is usually restricted to "flat" data parallelism, which is good for implementers but bad for programmers. Instead, I'll describe the "nested" data parallel programming model, first developed in the 90's by Blelloch and Sabot. It is great for programmers but much harder to implement; as a result, it's virtually unknown in practice. It's really only feasible to support nested data parallelism in a purely functional language, so we are building a high-performance implementation in Haskell.

In this talk you'll learn what nested data parallelism is, why it's important, and how to use it in Haskell. Fear not: I won't assume you know any Haskell. Yet.


JAOO: The Clojure Concurrency Story

Слайды с презентации Rich Hickey "The Clojure Concurrency Story" на JAOO Conference:
http://jaoo.dk/aarhus-2009/file?path=/jaoo-aarhus-2009/slides/RichHickey_TheClojureConcurrencyStory.pdf

Abstract:

All functional programming languages emphasize working with immutable data as much as possible. How can that be efficient, and what benefits does it bring? And what happens when you need state, to manage changing values over time? One method is to use mutable references with concurrency semantics, such as transactions, coupled with persistent data structures. This combination makes for easy, lock-free designs, well within the comfort zone of those used to imperative programming.

This talk discusses how immutability, state and identity are handled in the Clojure language. You'll learn details about Clojure's managed references, agents and STM.

Target audience: Developers interested in alternative approaches to concurrency.

JAOO: Actors, and the Forgotten Art of Modeling Concurrent Systems

Слайды с презентации Kresten Krab Thorup & Dave Thomas "Actors, and the Forgotten Art of Modeling Concurrent Systems" на JAOO Conference:
http://jaoo.dk/aarhus-2009/file?path=/jaoo-aarhus-2009/slides/DaveThomas_and_KrestenKrabThorup_ActorsAndTheForgottenArtOfModelingConcurrentSystems.pdf

Abstract:

In current talk on programming, programming languages and concurrency (and specifically the intersection of those), much focus is on how to make programs run fast and best utilize the given hardware, multicore computers, and server farms. That is all very good, and understanding how to do that is indeed an important engineering quest. However, ...

It seems that in our haste to be fast, we have forgotten the art of modeling. Some times we need to model a system which is by its very nature concurrent, and we need to express this understanding, so that we - as human beings - can comprehend and reason about its behavior. For these cases we need to intentionally model the concurrency. When the task at hand is not just a matter of parallellizing an otherwise sequential program, but expressing the interactive behaviors of systems that handle many things concurrently.

This is nothing new. In the mid-80's up to the mid-90's there was a lot of active research and development in so-called actor languages; a class of concurrent programming languages that target this problem space. Today, actor programming is starting to be used again, especially in context of the Scala programming language.

In this talk we will review the rationales and thinking behind actor languages, provide examples of actor programming models, languages, and draw the lines up to current programming models.

JAOO: The Concurrency Revolution: The Hardware Story

Слайды с презентации Brian Goetz "The Concurrency Revolution: The Hardware Story" на JAOO Conference:
http://jaoo.dk/aarhus-2009/file?path=/jaoo-aarhus-2009/slides/BrianGoetz_TheConcurrencyRevolutionTheHardwareStory.pdf

Abstract: Everyone is saying we need a revolution in the way we approach concurrency; the availability of cheap multiprocessor systems has brought concurrency, which used to be a topic for experts only, into the foreground. How did we get here? Why are the performance-enhancing techniques that worked for twenty years no longer working, and how did this become such a problem that we have to change our programming models? In this talk, we'll dive under the hood and look at the hardware trends that have forced us to turn away from increasing sequential performance and towards multicore CPUs and chip multi-threading, and look at the implications for software design and performance analysis.

Target audience: Advanced engineers who are interested in performance management and tuning, trends in software design, or anyone who is interesting in understanding the magic under the hood.