HPCwire: Woven Demos Low-Latency Ethernet for Market Trading

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

Leading HPC
Solution Providers

HPCwire >> Industry >> Financial Services

April 04, 2008

Woven Demos Low-Latency Ethernet for Market Trading

SANTA CLARA, Calif., April 1 -- Woven Systems Inc., the leading innovator of Ethernet Fabric switching solutions for datacenters and high-performance computing clusters, today announced that its TRX 100 Ethernet Switch and EFX 1000 10 Gigabit Ethernet Fabric Switch provided the scalable, ultra-low latency network infrastructure enabling the world's first large scale market data messaging test of over 10 million messages per second. The benchmarks, which were performed at the AMD Developer Center in Sunnyvale, Calif., included the Tervela TMX Message Switch which intelligently routed market data to over 1500 subscribers. Test results demonstrated performance that was millions of messages-per-second higher than previous solutions and a mean roundtrip message latency of just 58 microseconds through the message switching infrastructure. The jitter performance for the combined Woven-Tervela solution was less than 25 microseconds which translates to much more consistent delivery of market data messages as traffic loads increase.

Market data message networks process and deliver up-to-the-millisecond financial information from data feeds to thousands of subscribers to facilitate market trades. Rapid processing and multicasting of market information from publishers to thousands of subscribers requires extremely low latency to facilitate profitable market transactions. Even more important is consistent, predictable performance as information volume varies throughout the trading day. Market opening and closing times typically drive huge spikes in information volume where predictable network throughput and latency as well as reliable data delivery is needed the most.

"As global markets become increasingly integrated, the volume of information to process and broadcast requires ever-increasing server and network resources," said Dave Passmore, research director with Burton Group. "Compute clusters supporting market data applications have now reached teraflop proportions. This is certainly one of the key drivers of 10 Gigabit Ethernet Fabrics that must be able to deliver ultra-low latency and very predictable performance, especially under peak loads."

The Woven Ethernet Fabric employed in the testing connected to 130 dual socket, dual-core AMD Opteron processor-based servers -- a total of 576 CPU cores with the ability to process up to 3,225 billion floating-point operations per second (gigaFLOPS). Forty servers were connected to each of four Woven TRX 100 server aggregation switches with 10 Gigabit Ethernet uplinks to an EFX 1000 Ethernet Fabric Switch to create a non-blocking fabric.

"Delivering market data as quickly and reliably as possible is what gives our financial services customers a competitive edge," said Barry Thompson, founder and CTO of Tervela. "The Woven switch infrastructure used during the landmark AMD test not only enabled us to scale the benchmark, but allowed us to truly measure the performance of the message network without having to worry about the underlying Ethernet infrastructure."

Ultra-low latency cut-through switching combined with Woven's patented Active Congestion Management algorithms provide the critical features that enable the Ethernet Fabric Switch to handle volatile market data traffic patterns with consistent throughput. Market data applications typically rely on the User Datagram Protocol (UDP) which is often employed for time-sensitive data transmission. However, UDP offers no provision for re-transmission of dropped data. Woven's Active Congestion Management feature continuously monitors network congestion and dynamically re-routes traffic to less busy paths in the fabric to prevent data loss. The use of industry-standard Ethernet also affords the opportunity to easily integrate and consolidate market data applications with the datacenter's storage and system backbone networks, resulting in additional cost savings.

The EFX 1000 Ethernet Fabric Switch is Woven's flagship offering for compute cluster interconnects in financial services markets as well as for large-scale enterprise and Internet datacenters. A single EFX 1000 switch, configurable with up to 144 10 Gbps Ethernet ports, delivers lossless throughput with a maximum port-to-port latency of 1.5 microseconds. Multiple switches can be deployed in a resilient, multi-path Ethernet Fabric supporting up to 4000 non-blocking edge ports. In such multi-switch configurations, the combination of latency-minimizing Active Congestion Management and cut-through switching results in just 4 microseconds between any two pairs of edge ports.

About Woven Systems

Woven Systems is an innovative network infrastructure provider that offers the industry's first massively scalable Ethernet Fabric switching solutions for datacenters. Fully compliant with Ethernet standards, Woven's patented vSCALE packet processing technology redefines network performance and efficiency with Active Congestion Management for balancing traffic, operational simplicity, and a significantly lower power consumption. The Woven solutions deliver the scalability of InfiniBand, the reliability of Fibre Channel and the simplicity of Ethernet. For more information, visit Woven Systems on the Web at www.wovensystems.com.

-----

Source: Woven Systems Inc.

Article Tools

Share & Save Options

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 Linux HPC Empire Strikes Back

While the Microsoft juggernaut has been touting the joys of its new Windows HPC Server 2008, the Linux HPC contingent has been somewhat less vocal of late. But now Red Hat has come up with its version of an integrated cluster solution.
Read More...

Nexsan Looks to Scare Up HPC Customers With Storage Beast

Even though the cost of servers still dominates the datacenter budget, storage is actually on a steeper growth curve. HPC storage, in particular, is being singled out as high-growth opportunity. Vendors are scrambling to keep up.
Read More...

The Week in Review

Google datacenters most energy efficient; Cluster Resources to demo Moab Hybrid Cluster; Red Hat Linux releases HPC distro. John West recaps those stories and more in our weekly wrap-up.
Read More...

Read more HPCwire features...

Top Headlines

Oracle and HP's Database Machine Predicated on Voltaire

Oct 06 | The Register | Does the HP Oracle Database Machine represent InfiniBand's big chance to break out its HPC niche? Read more...

3D Imaging Spreads to Fashion and Beyond

Oct 06 | BusinessWeek | A body scan can save a lot of time in the fitting room, and fields from medicine to architecture are adopting 3D computing applications. Read more...

Structural Engineers and Computer Scientists Hope to Integrate Disciplines to 'Revolutionize Building Construction'

Oct 03 | UCSD News | Despite the evolution of computer science over the past 30 years, structural engineering -- hindered by a reluctance to adapt to digital innovations -- has remained relatively unchanged as a discipline. Read more...

Credit Crisis Spreads a Pall Over Silicon Valley

Oct 02 | New York Times | Silcon Valley is starting to feel the effects of the credit crunch. Read more...

Google: 'The World's Most Efficient Data Centers'

Oct 01 | Data Center Knowledge | Google today disclosed details of its data center energy usage, confirming that it operates some of the most efficient facilities in the world. Read more...

Read more headlines...

Featured Whitepapers

Panasas® Tiered Parity™ Architecture

Sep 04 | | Disk drives are approximately 250 times denser today than a decade ago. This is good news for users who are creating, manipulating and storing more data than ever before. It gives them an opportunity to derive more value from their stored data and lowers the capital acquisition and operating expense associated with that data.

View the White Paper Library

Multimedia

Video White Paper: Architecting a Better Network Storage Solution

BlueArc's Titan architecture represents an evolutionary step in file servers by creating a hardware-based file system that can scale bandwidth, IOPS, and overall data capacity well beyond conventional software-based devices. With its ability to virtualize a massive storage pool of up to four usable petabytes of tiered storage, Titan can scale with growing data requirements, offering a competitive advantage for businesses, researchers, or other enterprises seeking to better manage data growth while still ensuring optimal performance.