The Un(?)fair Advantage of Latency Arbitrage

The Un(?)fair Advantage of Latency Arbitrage
"Technologically advanced traders are giving themselves an advantage that some people feel is just that unfair. Using techniques and technologies I’ll describe below, they squeeze every last microsecond of latency
out of their market data feeds and trading systems to give themselves a sneak peak of market prices that’s measured in milliseconds. Thanks to powerful algorithms and high-speed order executions systems that’s enough time for them to engage in “latency arbitrage” – the buying and selling of equities based on small price changes that have not yet been broadly recognized due to the varying speeds of market data delivery systems."

Tricks of the Trade and Key Technologies:

• Co-location
• Cut-through Switches
• Consolidation of functionality
• Hardware-based Feedhandlers
• Hardware-based Middleware

 

 

Implementing Arithmetic and Other Analytic Operations By Transcriptional Regulation

"The transcriptional regulatory machinery of a gene can be viewed as a computational device, with transcription factor concentrations as inputs and expression level as the output. This view begs the question: what kinds of computations are possible?"

Parallel Computational Subunits in Dentate Granule Cells Generate Multiple Place Fields

"A fundamental question in understanding neuronal computations is how dendritic events influence the output of the neuron. Different forms of integration of neighbouring and distributed synaptic inputs, isolated dendritic spikes and local regulation of synaptic efficacy suggest that individual dendritic branches may function as independent computational subunits. In the present paper, we study how these local computations influence the output of the neuron."

Robust Models for Optic Flow Coding in Natural Scenes Inspired by Insect Biology

Robust Models for Optic Flow Coding in Natural Scenes Inspired by Insect Biology


"The extraction of accurate self-motion information from the visual world is a difficult problem that has been solved very efficiently by biological organisms utilizing non-linear processing. Previous bio-inspired models for motion detection based on a correlation mechanism have been dogged by issues that arise from their sensitivity to undesired properties of the image, such as contrast, which vary widely between images. Here we present a model with multiple levels of non-linear dynamic adaptive components based directly on the known or suspected responses of neurons within the visual motion pathway of the fly brain. "

Technology:Intel launches all-new PC architecture with Core i5/i7 CPUs

Intel launches all-new PC architecture with Core i5/i7 CPUs

"Intel finally brings its Nehalem architecture to its mainstream desktop PC line with today's Core i5/i7 and P55 chipset launch. In contrast to most previous launches, the chipset is actually a big deal this time around, marking as it does Intel's most radical overhaul of basic PC system architecture since the introduction of AGP.
"So with the advent of the P55, Intel's core logic has gone from a two-chip to a one-chip implementation, pushing ahead of the comparable AMD platform. In theory, this very tight, direct coupling of the GPU + GDDR and CPU + DRAM systems should make for a performance boost vs. both earlier topologies."

"Conclusions
The launch of Core i5/i7 and the P55 represent a major step forward for Intel's desktop line, and they bring Nehalem's performance dominance of AMD's offerings onto the mainstream desktop. Intel is now well-positioned against both NVIDIA and AMD/ATI, since the former will soon be squeezed out of the high-volume Intel IGP market entirely (when the GPU goes on-die) and the latter is increasingly forced back into the bargain niche that it had so successfully escaped with the launch of the K8."


Intel's Core i7 870 & i5 750, Lynnfield: Harder, Better, Faster Stronger

"The Lynnfield/P55 launch is huge. Virtually every single motherboard manufacturer has a P55 board available. Prices range from ~$110 - $300 depending on the number of bells and whistles."


"Despite being cheaper, Lynnfield is larger than Bloomfield. The larger die is due to one major addition: an on-die PCIe controller."


"The pink block to the right of the die is the PCIe controller, that's 16 PCIe 2.0 lanes coming right off the chip. Say hello to ultra low latency GPU communication. You'd think that Intel was about to enter the graphics market or something with a design like this."


"Intel made a very important announcement when Nehalem launched last year. Everyone focused on cache sizes, performance or memory latency, but the most important part of Nehalem was far more subtle: the Power Gate Transistor."


"Using some clever materials engineering Intel developed a very low resistance, low leakage, transistor that can effectively drop any circuits behind it to near-zero power consumption; a true off switch. This is the Power Gate Transistor."


"On every single Nehalem (Lynnfield included) lies around 1 million transistors (about the complexity of a 486) whose sole task is managing power. It turns cores off, underclocks them and is generally charged with the task of making sure that power usage is kept to a minimum. "


"Final Words


As I see it, LGA-1366 has a few advantages:


1) High-end multi-GPU Performance


2) Stock Voltage Overclocking


3) Future support for 6-core Gulftown CPUs


Speaking of turbo, I'd say that Intel is definitely on to something here. The performance impact was small with Bloomfield, but turbo on Lynnfield is huge. My tests showed up to a 17% increase in performance depending on the workload, with most CPU-influenced scenarios seeing at least 9 or 10%. The turbo mode transitions happen fast enough to accelerate even simple actions like opening a new window. OS and application responsiveness is significantly improved as a result and it's something that you can actually feel when using a Lynnfield machine. It all works so seamlessly, you just always get the best performance you need. It's like Intel crammed the best single, dual and quad-core processors all into one package.


Lynnfield shows us the beginning of how all microprocessors are going to be made in the future. Even AMD is embracing turbo, we'll see it with Fusion in 2011. Extend turbo to its logical conclusion and you end up with something very exciting. Imagine a processor made up of many different cores, large and small, CPU and GPU. Each one turning on/off depending on the type of workload, and each running as fast as possible without dissipating more heat than your system can handle."

Technology: The Future Is Parallel: What's a Programmer to Do?

Talks and Posters

• Guy Steele. Organizing Functional Code for Parallel Execution; or, foldl and foldr Considered Slightly Harmful. Slides for a one-hour invited talk. Some overlap with the earlier talk "The Future Is Parallel" but with many new examples and an emphasis on the use of "big" reduction operators to express catamorphisms. Presented at the ACM SIGPLAN International Conference on Functional Programming (ICFP), Edinburgh, Scotland, August 31, 2009.
  
  
• Guy Steele. The Future Is Parallel: What's a Programmer to Do? Breaking Sequential Habits of Thought. (30-minute version). Shorter version of the talk presented at NEPLS. Presented at the Workshop on Directions in Multicore Programming Education (co-located with ACM ASPLOS), Washington, DC, March 8, 2009.

• Guy Steele. The Future Is Parallel: What's a Programmer to Do? Breaking Sequential Habits of Thought.. Slides on the divide-and-conquer strategy for organizing programs, with some small surprising code examples contrasting it with customary linear sequential strategies. One-hour talk presented at the 5 March 2009 meeting of the New England Programming Languages and Systems (NEPLS) Symposium at Mitre. Previously presented at CMU and at an IEEE workshop. 

Technology:Hardware Hackers Create a Modular Motherboard | Gadget Lab | Wired.com

• Hardware Hackers Create a Modular Motherboard | Gadget Lab | Wired.com

• An ambitious group of hardware hackers have taken the fundamental building blocks of computing and turned them inside out in an attempt to make PCs significantly more efficient.
• The group has created a motherboard prototype that uses separate modules, each of which has its own processor, memory and storage.
• Each square cell in this design serves as a mini-motherboard and network node; the cells can allocate power and decide to accept or reject incoming transmissions and programs independently.
• A modular architecture designed for parallel and distributed processing could help take computing to the next level, say its designers.
• “We are at a point where each computer processor maxes out at 3Ghz (clock speed) so you have to add more cores, but you are still sharing the resource within the system,” says Justin Huynh, one of the key members of the project. “Adding cores the way we are doing now will last about a decade.”
• Computing today is based on the von Neumann architecture: a central processor, and separate memory and data storage. But that design poses a significant problem known as the von Neumann bottleneck. Though processors can get faster, the connection between the memory and the processor can get overloaded.That limits the speed of the computer to the pace at which it can transfer data between the two.