Big Data. Big Decisions
InformationWeek
Special Coverage Series

Commentary


Parallelism at PDC: Day 0 / overview

The pre-conference session on Concurrency was six hours of good information about what Microsoft is rolling out for parallel programming.

The session was titled "Concurrent, Multi-core Programming on Windows and .NET" and done by David Callahan, Joe Duffy and Stephen Toub.



The pre-conference session on Concurrency was six hours of good information about what Microsoft is rolling out for parallel programming.

The session was titled "Concurrent, Multi-core Programming on Windows and .NET" and done by David Callahan, Joe Duffy and Stephen Toub.

There were many shameless plugs for a new book by Joe Duffy - made good by giving away books, and demonstrating more than enough competence  with the subject to make the book worth a look.

The session started with David giving us an excellent introduction toe parallel computing.

He started with “the shift to multicore,” and started this with talking about the first 40 years of Moore’s Law.  Yes, an amazing prediction.  This brings up the question - how do you turn cheap processors in to value? He reference an excellent paper by “Moore’s Dividend” by Jim Larus of Microsoft Research (I blogged about this a few months ago).

David talked about the Power-Wall, Memory Wall and ILP Wall – which all lead to the realization that single core performance has all but stopped. David showed a block diagram of the upcoming Intel Larrabee project .

David discussed parallelism with four key points: there will be latent parallelism for future scaling, you should start by focusing on data – the scalable dimension, program using tasks not threads and there is NO silver bullet – many “right” approaches.

David had a nice example program he showed, and talked about the need to understand if you really need to force ordering on data structure access such as walking a graph.  David advocated relaxed sequential ordering.

David defined Parallelism, Concurrency, “multi-threading” – citing a lack of industry standard, but a need for clarity today.

David proposed definitions of:

Concurrent processing: independent requests (most server applications). [simulating isolation of threads]

Parallel processing: decompose one task to enable concurrent execution. [scheduling tasks]

David stressed the need to program in tasks not threads.  Of course, I couldn’t agree more.

David talked about “structured multi-threading” and cited Intel Threading Building Blocks and Cilk Arts as examples of this.  There are “no threads” here and “no concept of processors” here.  The user is abstracted away from mapping of threads onto processors.

David talked about work-sharing and cited OpenMP as an example.

David talked about Dataflow and showed a strassen multiplication (an example of how to do this in TBB can be found in my book on TBB).

David advocated “over-decomposing for scaling” citing it as the new Free Lunch.

David finished up talking about the challenges of scaling (Rabbit Hole of Performance) and debugging a parallel program (Black Hole of Correctness).

David's general treatment of the subject set things up well for Steve to jump in an show us code.

Steve, interestingly, started by showing a demo using threads.

I was surprised – because, while it matched what attendees expect when you talk about parallelism, it directly conflicts with the advice to program in ‘tasks instead of threads’ that David advocated.

Steve was demonstrated the concepts of concurrency, and the various thread controls such as background, interrupt and abort.

Steve then moved on to talk about ThreadPools with QueueUserWorkItem.

After all this, Steve said we DO want to program in tasks instead of threads.

Steve said that an application should NOT do blocking work in the part of the program that has the message loop for handling window events. His example was for “button click” – and the handling (blocking work) should be put off in a thread (use “BeginInvoke()” to run a delegate on the thread pool associated with the window the event is associated with.

Steve demoed coding for “background work and UIs.”  When asked about Silverlight – and confirmed that all this applies with Silverlight as well.  In fact, Steve pointed out that in a browser – it is even more important to do everything asynchronously – so you don’t freeze up Internet Explorer or Chrome – or your browser of choice the Silverlight code is running in.  He said, "Silverlight wants you to do everything asynchronously."

At this point, we broke for lunch.

More about the afternoon session later. 



Related Reading


More Insights




Currently we allow the following HTML tags in comments:

Single tags

These tags can be used alone and don't need an ending tag.

<br> Defines a single line break

<hr> Defines a horizontal line

Matching tags

These require an ending tag - e.g. <i>italic text</i>

<a> Defines an anchor

<b> Defines bold text

<big> Defines big text

<blockquote> Defines a long quotation

<caption> Defines a table caption

<cite> Defines a citation

<code> Defines computer code text

<em> Defines emphasized text

<fieldset> Defines a border around elements in a form

<h1> This is heading 1

<h2> This is heading 2

<h3> This is heading 3

<h4> This is heading 4

<h5> This is heading 5

<h6> This is heading 6

<i> Defines italic text

<p> Defines a paragraph

<pre> Defines preformatted text

<q> Defines a short quotation

<samp> Defines sample computer code text

<small> Defines small text

<span> Defines a section in a document

<s> Defines strikethrough text

<strike> Defines strikethrough text

<strong> Defines strong text

<sub> Defines subscripted text

<sup> Defines superscripted text

<u> Defines underlined text

BYTE encourages readers to engage in spirited, healthy debate, including taking us to task. However, BYTE moderates all comments posted to our site, and reserves the right to modify or remove any content that it determines to be derogatory, offensive, inflammatory, vulgar, irrelevant/off-topic, racist or obvious marketing/SPAM. BYTE further reserves the right to disable the profile of any commenter participating in said activities.

Disqus Tips To upload an avatar photo, first complete your Disqus profile. | View the list of supported HTML tags you can use to style comments. | Please read our commenting policy.

Follow InformationWeek

By The Numbers

What Are Your Primary Concerns About Using Big Data Software?

Base: 417 respondents at organizations using or planning to deploy data analytics, BI or statistical analysis software
Data: InformationWeek 2013 Analytics, Business Intelligence and Information Management Survey of 541 business technology professionals, October 2012

What Do You Think?

What's your attitude about SQL analysis on top of Hadoop?
We want fast, standard SQL analysis capabilities on Hadoop ASAP
Hadoop is for unstructured data; SQL is for relational databases
We'll give SQL on Hadoop a try, but relational DBs will remain the mainstay
Given strong SQL support on Hadoop, we'd nix the data warehouse
We're not interested in Hadoop
No opinion



Related Content

From Our Sponsor

Five Big Data Challenges and How to Overcome Them with Visual Analytics

Five Big Data Challenges and How to Overcome Them with Visual Analytics

Business leaders often need a visual snapshot of data to quickly grasp and use it. This paper identifies five challenges in presenting data and how visual analytics can resolve them. Solutions are suggested to overcome the challenges of: speed, data clarity, data quality, displaying meaningful results, and dealing with outliers.

Game-Changing Analytics: How IT Executives Can Use Analytics to Create Innovation and Business Success

Game-Changing Analytics: How IT Executives Can Use Analytics to Create Innovation and Business Success

Today's competitive advantage requires a deeper understanding of your business, your market and your customers. As an IT executive, you can drive that knowledge transformation. In this white paper, learn how to make decisions as a strategic business leader and three steps to begin an analytics initiative within your enterprise.

Data Visualization Techniques: From Basics to Big Data with SAS Visual Analytics

Data Visualization Techniques: From Basics to Big Data with SAS Visual Analytics

High-performance data visualization turns sophisticated analyses into meaningful graphics, leading to faster and smarter decision making. In this white paper, learn how visual analytics can transform big data, with additional features such as real-time functionality, mobile compatibility, robust applications for technical groups and accessibility for nontechnical users.

Big Data: Lessons from the Leaders

Big Data: Lessons from the Leaders

Financial performance, competitive advantage, operational efficiency, strategic decision making - every business goal can extract value from big data, and the time for doubt or inaction has long passed. In this Economist Intelligence Unit report, in-depth interviews with data pioneers reveal the link between the effective use of big data and the bottom line among other results.

Decision-Driven Data Management: A Strategy for Better Decisions with Better Data

Decision-Driven Data Management: A Strategy for Better Decisions with Better Data

Which came first, the data or the decision? This white paper makes the case for having a decision in mind, then tailoring big data's volume, variety and velocity to achieve business results such as overcoming customer dissatisfaction or creating well-informed strategies in real time.

Informationweek Reports

Research: The Big Data Management Challenge

Research: The Big Data Management Challenge

The challenge of big data is real, but most organizations don't differentiate 'big data' from traditional data, and nearly 90% of respondents to our survey use conventional databases as the primary means of handling data. We'll help you understand what constitutes big data (it's not just size) and the numerous management challenges it poses.