HPCwire

Since 1986 - Covering the Fastest Computers
in the World and the People Who Run Them

Language Flags

Visit additional Tabor Communication Publications

Datanami
Digital Manufacturing Report
HPC in the Cloud
Green Computing Report

Tabor Communications
Corporate Video

Joint Testing Proves Scalable 10GigE for Cloud Networking


Arista Networks, Fulcrum Microsystems and Ixia collaborate on testing of high performance datacenter switches for cloud networking applications

PORTLAND, Ore. Nov. 16 -- SC09 -- Arista Networks, Fulcrum Microsystems and Ixia today announced the results of collaborative performance and scalability testing of 10-Gigabit Ethernet in a cloud computing environment. The results demonstrated high throughput and low latency for Arista Network's 7148SX switch using Fulcrum's FocalPoint FM4224 switch silicon. The 7148SX is one of the densest 10GE switches on the market today.

As the industry moves to Ethernet in the datacenter, questions have arisen about the best interconnect architecture. The results of these tests show that high density, low latency 10GE switches in logically flat network architectures can scale to meet large-scale applications. The European Advanced Network Test Center (EANTC) has validated the latency and throughput tests per RFC 2544, RFC 2889 (for unicast traffic) and RFC 3918 (for Multicast traffic).

Performed at Ixia's renowned iSimCity facility in Santa Clara, Calif., the testing measured the performance of the Arista 7148SX 48-port 10GE SFP+ switch, which features Fulcrum's FocalPoint FM4224 switch silicon. The tests used full line-rate, large-scale traffic running fully meshed and yielded results that validated the switch's ability to provide the low-latency and throughput performance for lossless datacenter Ethernet transmission.

The test results showed "extraordinarily low latency" in the Arista switch -- as low as 600 nanoseconds (min) between any two ports--and lossless performance for all frame sizes using IETF RFC 2544 and 2889 throughput test methodologies. This performance means scale-out environments can now get true non-blocking performance using the Arista 7148SX switches, as shown by the absence of packet drops in the RFC 2889 full-mesh test results. Complete test results can be found at 7100 Series Performance Results.

With multicast traffic growing in importance in cloud networking applications, the tests showed that the Arista switch delivered at line rate per RFC 3918 mixed class and multicast-only throughput test cases with no packet loss. In the multicast-only throughput and latency test case, a minimum of 600 nanoseconds was recorded with average latency of 1273 nanoseconds and maximum latency of 1520 nanoseconds. Low-latency multicast performance is essential to many financial networks that need "fair" multicast transmissions where even a microsecond of early data delivery can mean an unfair trading advantage.

To deliver SAN-quality performance, lossless transmission is vital even when there is drastic oversubscription of a switch port. In an additional demonstration where EANTC was not present, the Arista 7148SX switch showed the PAUSE and Priority Flow Control (PFC) features necessary to maintain lossless transmission on ports oversubscribed as much as 47-to-1 with streams of converged IP and storage traffic at rate and at scale.

"As high performance datacenters move to flat network architectures in order to accommodate the increase in traffic associated with moving into the cloud, it becomes absolutely critical that these port-dense switches have the proper features to ensure reliable transmission of full line rate traffic," said Mike Zeile, president and COO of Fulcrum Microsystems. "This test data gives a real-world validation of the virtues of FocalPoint silicon, clearly demonstrating the features key to enabling dense, low-latency switches that scale without performance degradation."

"Today's cloud networking designs require ultra-low latency and non-blocking performance to deliver a robust platform that works for all applications," said Anshul Sadana, Vice president of Systems Engineering of Arista." With the Arista platform, our customers are no longer worried about the network being a bottleneck."

"10GE continues to penetrate datacenters as the prime technology to support low latency and high performance applications such as server virtualization, cloud computing and market data feeds," said Victor Alston, Senior VP of Product Development of Ixia. "We are proud to be able to provide superior 10GE test solutions with the performance, density and reliability required to validate the high benchmark set during this test."

Arista Networks will be showcasing its technology at booth # 249. Fulcrum Microsystems solutions will be on display in booth #335. Ixia will demonstrate its 10GE and datacenter Ethernet solutions at the Ethernet Alliance booth, #1259.

About Arista Networks

Arista Networks was founded to deliver cloud networking solutions for large datacenter and computing environments. Arista offers best-of-breed 10 Gigabit Ethernet switches that redefine scalability, robustness, and price-performance. At the core of Arista's platform is the Extensible Operating System (EOS),a pioneering new software architecture with self-healing and live in-service software upgrade capabilities. For more information, visit http://www.aristanetworks.com.

About Fulcrum Microsystems

Fulcrum Microsystems Inc. is a fabless semiconductor company that has developed the FocalPoint family of 10 Gigabit Ethernet (10GbE) switch/router chips for next-generation datacenter fabric designs. FocalPoint devices offer the industry's lowest latency combined with the fine-grained flow control and high throughput required for converged datacenter fabrics. With its ControlPoint network OS, Fulcrum offers a complete solution along with a worldwide ecosystem of partners from which storage and clustering system designers can drive adoption of 10GE as the fabric of choice for a fully converged datacenter. More information can be found at www.fulcrummicro.com.

About Ixia

Ixia is a leading provider of converged IP performance test systems and service verification platforms for wireless and wired infrastructures and services. Ixia's test systems are used by network and telephony equipment manufacturers, semiconductor manufacturers, service providers, governments and enterprises to validate the performance and reliability of complex networks, devices and applications. Ixia's multiplay test systems address the growing need to test voice, video and data services and network capability under real-world conditions.

-----

Source: Arista Networks; Fulcrum Microsystems; Ixia

June 18, 2013

June 17, 2013

June 14, 2013

June 13, 2013

June 12, 2013

June 11, 2013

June 10, 2013

June 07, 2013

June 06, 2013


Most Read Features

Most Read Around the Web

Most Read This Just In

Asetek

Feature Articles

My Supercomputer is Bigger Than Yours!

Contributing commentator, Andrew Jones, offers a break in the news cycle with an assessment of what the national "size matters" contest means for the U.S. and other nations...
Read more...

Alternatives Emerge as Linpack Loses Ground

Today at the International Supercomputing Conference in Leipzing, Germany, Jack Dongarra presented on a proposed benchmark that could carry a bit more weight than its older Linpack companion. The high performance conjugate gradient (HPCG) concept takes into account new architectures for new applications, while shedding the floating point....
Read more...

Intel Snaps New Grips to HPC Hook

Not content to let the Tianhe-2 announcement ride alone, Intel rolled out a series of announcements around its Knights Corner and Xeon Phi products--all of which are aimed at adding some options and variety for a wider base of potential users across the HPC spectrum. Today at the International Supercomputing Conference, the company's Raj....
Read more...

Short Takes

Supercomputers: Not Always the Best for Big Data

Jun 18, 2013 | The world's largest supercomputers, like Tianhe-2, are great at traditional, compute-intensive HPC workloads, such as simulating atomic decay or modeling tornados. But data-intensive applications--such as mining big data sets for connections--is a different sort of workload, and runs best on a different sort of computer.
Read more...

Gordon Flashes Its Versatility in HPC Workloads

Jun 18, 2013 | Researchers are finding innovative uses for Gordon, the 285 teraflop supercomputer housed at the San Diego Supercomputer Center (SDSC) that has a unique Flash-based storage system. Since going online, researchers have put the incredibly fast I/O to use on a wide variety of workloads, ranging from chemistry to political science.
Read more...

Supercomputers: Still the King of the HPC Hill

Jun 17, 2013 | The advent of low-power mobile processors and cloud delivery models is changing the economics of computing. But just as an economy car is good at different things than a full size truck, an HPC workload still has certain computing demands that neither the fastest smartphone nor the most elastic cloud cluster can fulfill.
Read more...

TACC Longhorn Takes On Natural Language Processing

Jun 14, 2013 | For all the progress we've made in IT over the last 50 years, there's one area of life that has steadfastly eluded the grasp of computers: understanding human language. Now, researchers at the Texas Advanced Computing Center (TACC) are utilizing a Hadoop cluster on its Longhorn supercomputer to move the state of the art of language processing a little bit further.
Read more...

Titan Didn't Redo LINPACK for June Top 500 List

Jun 13, 2013 | Titan, the Cray XK7 at the Oak Ridge National Lab that debuted last fall as the fastest supercomputer in the world with 17.59 petaflops of sustained computing power, will rely on its previous LINPACK test for the upcoming edition of the Top 500 list.
Read more...

Sponsored Whitepapers

Best Practices in Big Data Storage

05/10/2013 | Cleversafe, Cray, DDN, NetApp, & Panasas | From Wall Street to Hollywood, drug discovery to homeland security, companies and organizations of all sizes and stripes are coming face to face with the challenges – and opportunities – afforded by Big Data. Before anyone can utilize these extraordinary data repositories, however, they must first harness and manage their data stores, and do so utilizing technologies that underscore affordability, security, and scalability.

Progress in Parallel: the Bull Parallel Programming Center

04/15/2013 | Bull | “50% of HPC users say their largest jobs scale to 120 cores or less.” How about yours? Are your codes ready to take advantage of today’s and tomorrow’s ultra-parallel HPC systems? Download this White Paper by Analysts Intersect360 Research to see what Bull and Intel’s Center for Excellence in Parallel Programming can do for your codes.

Sponsored Multimedia

HPCwire Live! Atlanta's Big Data Kick Off Week Meets HPC

Join HPCwire Editor Nicole Hemsoth and Dr. David Bader from Georgia Tech as they take center stage on opening night at Atlanta's first Big Data Kick Off Week, filmed in front of a live audience. Nicole and David look at the evolution of HPC, today's big data challenges, discuss real world solutions, and reveal their predictions. Exactly what does the future holds for HPC?

Webinar: Mellanox Virtual Modular Switch, the Most Efficient 40GbE Aggregation Switch Solution

Join our webinar to learn how IT managers can migrate to a more resilient, flexible and scalable solution that grows with the data center. Mellanox VMS is future-proof, efficient and brings significant CAPEX and OPEX savings. The VMS is available today.

Atlanta's Big Data Kick Off Week Meets HPC Cray Exxact

HPC Job Bank


Featured Events






  • November 17, 2013 - November 22, 2013
    SC'13
    Denver, CO
    United States


HPCwire Events