HPCwire: Hardware Providers Set Latency Standard

The Leading Source for Global News and Information Covering the Ecosystem of High Productivity Computing

HPCwire >> Topic >> Interconnects

November 11, 2008

Hardware Providers Set Latency Standard

Page: 1 of 2
1 | 2 All »

Solace, Arista and NetEffect publish hardware architecture and demonstrate market data performance gain

OTTAWA, Nov. 11 -- Solace Systems, the leading supplier of high-speed, low-latency content networking systems, today announced a collaboration with Arista Networks and NetEffect, an Intel company, to demonstrate how specialty hardware platforms can optimize both the peak performance and consistency of ultra-low latency data delivery under the most demanding conditions.

Using Arista Networks low latency switching technology in 1GigE mode, NetEffect Gigabit Ethernet adapters on client computers and a Solace Systems 3260 Content Router for ultra-low latency messaging, tests measured end-to-end market data delivery from the publishing feed infrastructure routed to the appropriate receiving applications. Tests used 100 byte messages at rates of 500,000 messages per second to simulate peak US market data rates for all equities, commodities and FX instruments. Testing resulted in an average latency of 31 microseconds and a 99.9th percentile latency of 47 microseconds with a standard deviation of just 6 microseconds.

A second test with 100 byte messages at 1,000,000 messages per second was run to simulate rates greater than peak OPRA market data activity. This yielded an average latency of 35 microseconds and a 99.9th percentile latency of 55 microseconds, with a standard deviation of 7 microseconds. The combination of these three technologies demonstrated that tens to hundreds of microseconds can be removed from high-volume market data configurations when compared with standard software and Ethernet switching technology.

In financial services, low latency translates directly to greater profits as firms race to feed their algorithmic trading infrastructures with more and more market data. As a result, firms are increasingly looking to replace strained legacy software infrastructure with purpose-built hardware to route, filter and accelerate messaging thus improving trading results.

"Executing the network stacks and messaging functionality in hardware makes a world of difference in the throughput, latency reduction, predictability and elimination of outliers," said Ralph Frankel, Solace's chief technology officer. "Specialty hardware is a hot topic on Wall Street specifically because it offers tangible benefits in terms of speed and ROI, especially in algorithmic trading environments."

All testing was completed in a configuration common to any production customer environment: one Solace 3260 Content Router, one Arista Networks 7124S Ethernet Switch in 1GigE mode and one NetEffect Gigabit Ethernet adapter per client. Financial firms with ultra-low latency requirements should expect to obtain very similar results in their own networks using this configuration. The test criteria were chosen to represent common, real world configurations and do not represent the maximum capabilities of each product. For example, in a two blade configuration, a Solace content router can handle 10 million messages per second (5 million in and 5 million out).

Ultra-low latency trading architects view latency consistency as just as important as absolute values. Most messaging benchmarks only quote minimum or median results, but fail to mention that results at the upper extremes are unacceptably high. The results published here include both the 99.9th percentile data and standard deviation results that affirm the narrow latency bounds of hardware messaging and network acceleration.

"Arista Networks switches support 1GigE and 10GigE speeds on each and every port, and offer the lowest latency in the market in both 1GigE and 10GigE mode. As such, they are ideal for a market data delivery platform," said Mansour Karam, director of marketing and business development at Arista Networks. "By replacing legacy 1GigE switches with Arista Networks switches in either 1GigE or 10GigE mode, customers will save tens of microseconds and ensure consistent ultra-low latency across the market data delivery platform."

For more information including office locations and contact details, visit www.solacesystems.com or www.aristanetworks.com.

Page: 1 of 2
1 | 2 All »

Article Tools

Share Options

(Digg, Technorati, more)

Subscribe

Subscribe to HPCwire

Discussion

There are 0 discussion items posted.

Join the Discussion

You must be a registered user in order to comment on HPCwire stories.

Members enjoy:

	Access to Special Content, including Podcasts, Webcasts, and Whitepapers and more!
	Ability to take an active role in the discussions that are shaping the HPC technology discipline!
	Special offers available only to HPCwire members!

Become a Registered User Today!

Registered Users Log in Here to Comment

Email Address:	Password (case sensitive)

Remember me	Forgot Password?

RSS Feeds

Feeds by Topic

Feeds by Industry

Feeds by Content Type

Feature Articles

The Week in Review

C-DAC announces plans for a petaflop system; IBM researchers are working on vertical integration techniques to extend Moore's Law another 15 years. We recap those stories and more in our weekly wrapup.
Read More...

Moscow State University Supercomputer Has Petaflop Aspirations

The Moscow State University supercomputer, Lomonosov, has been selected for a high-performance makeover, with the goal of tripling its processing power to achieve petaflop-level performance in 2010. T-Platforms, who developed and manufactured the supercomputer, is the odds-on favorite to lead the project.
Read More...

Intel Ups Performance Ante with Westmere Server Chips

Right on schedule, Intel has launched its Xeon 5600 processors, codenamed "Westmere EP." The 5600 represents the 32nm sequel to the Xeon 5500 (Nehalem EP) for dual-socket servers. Intel is touting better performance and energy efficiency, along with new security features, as the big selling points of the new Xeons.
Read More...

Read more HPCwire features...

Top Headlines

Australia Commissions Cray Supercomputer

Mar 19 | OfficialWire | New super to support intelligence work Down Under. Read more...

Intel Partners See 'Easy' Upgrade Path With Xeon 5600 Chips

Mar 18 | ChannelWeb | Westmere parts already showing up in HPC machines. Read more...

AMD: OEMs primed for Opteron 6100s

Mar 17 | The Register | But what about the tier ones? Read more...

Arrival of the Desktop Supercomputer

Mar 17 | Cadalyst Magazine | A new generation of workstations is changing the nature of technical computing. Read more...

Scheduling HPC In The Cloud

Mar 17 | Linux Magazine | Latest iteration of Sun Grid Engine able to tap into Cloud. Read more...

Read more headlines...

Featured Whitepapers

Virtualization for Aggregation And The vSMP Architecture™

Jan 12 | | In-depth look at vSMP Foundation server virtualization technology, technical implementation, use cases and capabilities. The technical whitepaper provides an architectural overview and details on the three vSMP Foundation products: vSMP Foundation for SMP, vSMP Foundation for Cluster and vSMP Foundation for Cloud.

Copper Cable Technologies for High Performance Computing

Jan 18 | | This white paper discusses Gore’s copper cable assemblies, and how they continue to exceed the standards for providing reliable, cost-effective solutions for high-performance computer applications.

View the White Paper Library

Multimedia

Webcast: Virtualized Data Center Roundtable

Join this online panel discussion for live Q&A with leading industry experts, analysts, and end-users to discuss the latest innovations, best practices, barriers to implementation, and measurable benefits of server virtualization with a particular focus on today's real world solutions.

Webcast: Watch SC09 Birds of a Feather Video: Scalable Fault-Tolerant HPC Supercomputers

Learn about scalable fault-tolerant architectures and examples of energy efficient and scalable supercomputing clusters using dual QDR InfiniBand to combine capacity computing with network failover capabilities with the help of programming languages such as MPI and a robust Linux cluster management package.

Webcast: High Performance Computing for a Smarter Planet

LIVE@SCO9: The IBM team discusses new innovations in hardware, software and services that help clients better understand their workloads and get insight from their R&D efforts. Technology demonstrations include the soon-to-be-released Power7 HPC processor, the DCS990 system with 2.4 petabytes of storage, the xCAT management tool, secure HPC cloud computing and more. Winners of two HPCwire Readers' and Editors’ Choice Awards! Take the IBM virtual tour at SC09 or more information go online to: http://www-03.ibm.com/systems/deepcomputing/sc09.html