Archives

Posts from Taylor Kidd (Intel) RSS

Taylor Kidd (Intel)

I'm an AE (Application Engineer) at Intel. My area of interest is optimization, HPC and something called "scale enabling". My background is system software, embedded systems, distributed computing, and teaching. Way back when, I got my university education specializing in EM propagation and Systems Theory. And I have way too much education.

Use cases for Power Management

By Taylor Kidd (Intel) (18 posts) on June 30, 2010 at 4:39 pm
Comments (2)

I’m collecting power management (PM) use cases in a broad sense, meaning I’m not just limiting myself to client use cases. I figure that trying to limit this investigation to client is foolhardy at this point given the early state of PM technology.

Continued ›

Category: Uncategorized

Performance-power in client applications: looking at JouleSort

By Taylor Kidd (Intel) (18 posts) on February 3, 2010 at 3:44 pm
Comments (1)

Below, I've included my usual summary of any paper I read. Yep, I go through this trouble with everything, no matter how trivial. Of course, usually, this summary is private, sequestered away behind vales of security, not the least of which is the fact that no one in their right mind would be interested in [...]

Continued ›

Category: Game Development, Graphics & Media, Mobility, Parallel Programming
Tags: ,

A useful power-performance metric (Part IIa, the goal)

By Taylor Kidd (Intel) (18 posts) on December 30, 2009 at 2:22 pm
Comments (12)

So what is our ultimate goal here in this series of articles? We want to come up with a useful measure of power-performance for an application, such as Joules per instructions for a given app. But as we will see, finding this answer is much less clear cut than it first appears.

Continued ›

Category: Game Development, Graphics & Media, Mobility, Parallel Programming
Tags: , ,

Question: Estimating power performance on newer architectures

By Taylor Kidd (Intel) (18 posts) on December 18, 2009 at 1:01 pm
Comments (2)

How must these methodologies for estimating application energy usage be modified to account for newer architectures? Must we select a different set of kernels that will more accurately characterize power consumption? Is it even possible to drive the energy consumption from such a set of kernels?

Continued ›

Category: Game Development, Graphics & Media, Mobility, Parallel Programming
Tags: ,

Performance per Watt: Hey, I already know it’s important, don’t I? (The intro, part I)

By Taylor Kidd (Intel) (18 posts) on October 22, 2009 at 11:41 am
Comments (1)

What is performance per Watt? Performance per Watt is pretty straight forward when you first look at it. Then you begin to sink in the quicksand you’ve blithely walked into. The panic sets in as you sink lower and lower. Eventually you decide to ignore the whole complicated mess and go back to saying to [...]

Continued ›

Category: Game Development, Graphics & Media, Mobility, Parallel Programming
Tags: , , ,

Performance per Watt: Hey, I already know it’s important, don’t I? (The preface)

By Taylor Kidd (Intel) (18 posts) on October 14, 2009 at 3:23 pm
Comments (2)

One of the big advantages of blogging is that I can write about anything I want, as long is it doesn’t violate any Intel conduct rules, refer to any competitors, reveal any confidential information, insult anyone, compliment people (yeah, I said compliment), and so on and so on and so on. Outside of that, I [...]

Continued ›

Category: Game Development, Graphics & Media, Mobility, Parallel Programming
Tags: , , ,

Why P scales as C*V^2*f is so obvious (pt 2)

By Taylor Kidd (Intel) (18 posts) on August 25, 2009 at 3:06 pm
Comments (5)

THE GORY DETAILS Let’s continue from where we left off last time. Let’s figure out the why of the equation, P = C * V^2 * (a * f) To do this, we’re going to have to look at what is going on in one of the fundamental building blocks (a CMOS inverter) of an [...]

Continued ›

Category: Game Development, Graphics & Media, Mobility, Parallel Programming
Tags: ,

Why P scales as C*V^2*f is so obvious (pt 1)

By Taylor Kidd (Intel) (18 posts) on August 25, 2009 at 9:42 am
Comments (0)

[apologies gentle readers if you've seen this twice. I've moved this over from my mysterious twin account (taylor-kidd-2) over to this account (taylor-kidd)] [Warning: Math and physics alert! Math and physics alert!] I think that you've all seen this equation before: P = a * C * V2 * f Where P is power, C [...]

Continued ›

Category: Game Development, Graphics & Media, Mobility, Parallel Programming, Uncategorized
Tags: , ,

Energy / Power measurement wish list

By Taylor Kidd (Intel) (18 posts) on September 5, 2008 at 8:34 am
Comments (4)

I got such a good response from my previous post, that I decided to pose another question to my select and invisible audience. If you could come up with a wish list, what power related measurements would you like to get from the (computer) platform, as well as from the processor itself? Now, let's be [...]

Continued ›

Category: Game Development, Graphics & Media, Mobility, Parallel Programming
Tags: , ,

So how are P-states related to power management?

By Taylor Kidd (Intel) (18 posts) on August 15, 2008 at 12:16 pm
Comments (0)

This relationship between P-states, voltage and frequency is well and good, but how does this relate to power management? Power is literally, energy usage per unit of time. To get the total energy usage, you integrate the instantaneous power over the interval you're interested in, i.e. get the area under the curve. If there are [...]

Continued ›

Category: Game Development, Graphics & Media, Mobility, Parallel Programming
Tags: , ,

Can P-states save overall energy?

By Taylor Kidd (Intel) (18 posts) on July 31, 2008 at 12:48 pm
Comments (9)

Part of the reason I've been so silent recently is because I've been really really busy. (Aren’t we all?) But it also has to do with a short segment I had written on P-states and energy savings. This brief segment outlined a relationship between processor voltage, leakage current, frequency and power. My conclusion was that [...]

Continued ›

Category: Game Development, Graphics & Media, Mobility, Parallel Programming
Tags: , , ,

What exactly is a P-state? (Pt. 1)

By Taylor Kidd (Intel) (18 posts) on May 29, 2008 at 8:57 am
Comments (3)

A P-state is a voltage and frequency operating point What is a P-state? When someone refers to a P-state, generally only the frequency is talked about. For example, on my Intel Core Duo, P0 is 2.3 GHz, and P1 is 980 MHz. In truth, a P-state is both a frequency and voltage operating point. Both [...]

Continued ›

Category: Game Development, Graphics & Media, Mobility, Parallel Programming
Tags: , ,

There's got to be a catch

By Taylor Kidd (Intel) (18 posts) on April 29, 2008 at 9:24 am
Comments (0)

I hate moving. Nothing ever goes as it should. It takes 10 times longer than you expected. And that last box is finally unpacked just before you end up moving again. There's got to be a catch There are 5 CC-states and, depending upon how you count, 6 PC-states in the Penryn line of Intel [...]

Continued ›

Category: Graphics & Media, Mobility, Parallel Programming

(update) C-states, C-states and even more C-states

By Taylor Kidd (Intel) (18 posts) on March 27, 2008 at 11:56 am
Comments (8)

As I said before, a C-state is an idle state. The processor isn't doing anything useful, so why not shut some things off? Think of it in terms of your house. If you're not at home, why keep the lights, radio, and those 6 televisions going? Modern processors have several different C-states representing increasing amounts [...]

Continued ›

Category: Game Development, Graphics & Media, Mobility, Parallel Programming
Tags: , , ,

C-states and P-states are very different

By Taylor Kidd (Intel) (18 posts) on March 12, 2008 at 3:46 pm
Comments (10)

C-states are idle states and P-states are operational states. This difference, though obvious once you know, can be initially confusing. With the exception of C0, where the CPU is active and busy doing something, a C-state is an idle state. Since an idle CPU isn't doing anything (i.e. any useful work), why not shut it [...]

Continued ›

Category: Mobility, Parallel Programming
Tags: , ,

Introduction to power management on Intel Processors

By Taylor Kidd (Intel) (18 posts) on March 4, 2008 at 10:55 am
Comments (8)

I'm writing a white paper on how power management works on Intel processors. This means that I need a forum for airing out my preliminary sections. You guys are it. So if you would be so kind as to "constructively" comment. Feel free to point out bad wording, areas that need clarification, even places where [...]

Continued ›

Category: Mobility, Parallel Programming
Tags: , ,

How to write energy efficient apps

By Taylor Kidd (Intel) (18 posts) on December 9, 2007 at 1:55 pm
Comments (2)

T'is the season for holiday, vacations and early Q4 deadlines. 'nough said. Jason: "what can software developers do while they creating the applications to use those hardware components to consume the power? (sic) " There are a variety of things. Intel's focus is on processors, of course. Intel processors, such as the Core Duo, have [...]

Continued ›

Category: Game Development, Graphics & Media, Mobility, Parallel Programming
Tags: , ,

Power: Server vs client energy usage

By Taylor Kidd (Intel) (18 posts) on November 19, 2007 at 9:43 am
Comments (10)

Hello? Is anyone out there? I'm starting up a new blog concerning power issues and application software. I'm Taylor Kidd, an AE (that's Applications Engineer in Intelispeak) in the Intel SSG (Software Solutions Group), more specifically, the PPE (Power Platform Enabling) team. Our mission, since we decided to accept it, is to make venders and [...]

Continued ›

Category: Game Development, Mobility
Tags: , , , ,