Fulcrum Microsystems Blog

Happy New Year from all your friends here at Fulcrum Microsystems!

As 2009 draws to a close, we have taken a minute to reflect on our accomplishments from this calendar year and look forward to the next.  Perhaps most indicative of our success has been The Los Angeles Business Journal story naming the company as the area’s fastest growing private company.  Despite the adverse economic environment, Fulcrum continued to stay true to its mission of innovation and maintained its success.

With a number of design and customer wins set to come to fruition in 2010, along with new product announcements, we look forward to the year ahead.  The sales team has been out on the front lines driving the discussions of solutions that will enable networks to bring the next generation of real-world applications both in and out of the datacenter.

Keep up to date on all things Fulcrum right here at our blog.  Thanks and, again, Happy New Year to all!

The IEEE is defining an important new specification for the cloud data center that defines how to connect multiple virtual machines to a data center bridging (DCB) fabric through a single high bandwidth port. An earlier proposal by the PCI-SIG for single root complex IO virtualization (SR-IOV) has not gained much traction in the industry. Although Cisco has released the Nexus 1000 as a virtualized bridge within the blade server rack, it is in effect a proprietary solution that forces the use of Cisco equipment throughout the data center if unified management software and network policies are desired. This can be an expensive and limiting solution.

A new proposal called Virtual Ethernet Port Aggregator (VEPA) is being developed by the IEEE, as described in this article from Network World. This standard moves the virtualization function from the blade servers into the DCB fabric. Many companies are attracted to this proposal because the management and network policies can be easily applied to both the physical and virtual fabrics using a single software stack running on an Ethernet bridge. With the Cisco proposal, part of the fabric is inside the blade server system, which of course, Cisco wants to supply as well. It looks like Cisco has now acquiesced, and will join the VEPA standards effort.

FocalPoint switches have several features to support the VEPA standard. A key feature is the use of Global Resource Tags (GloRTs), which can be used to identify the locations of thousands of virtual machines connected to the FocalPoint fabric. Translation between VEPA tags and GloRTs at the ingress and egress of the fabric using a Q-in-Q type approach allows easy address translation across blade server domains while maintaining support for all other DCB traffic such as FCoE. This highly scalable solution can be achieved while maintaining the low cut-through latency pioneered by Fulcrum.

For more details on DCB and CEE support in FocalPoint, download this PDF whitepaper.

Comments are welcome: feedback@fulcrummicro.com

At the recent SC09 show, the Ethernet Alliance brought together equipment from 14 vendors to demonstrate Data Center Bridging (DCB) using a number of different storage protocols that leverage cost-efficient, ubiquitous Ethernet as a converged fabric.

The interconnect was a major element of this unified fabric showcase, as it incorporated different technologies such as 10GBase-T and SFP+ 10GbE Direct Attach Cables into the demonstration. It also demonstrated how network convergence takes advantage of high speed Ethernet to deliver client messaging, storage and server application communications over a single network.

Fulcrum’s Monaco reference design, which features the FocalPoint FM 4224 24-port 10GbE switch chip, was the 10GbE switch used for all of the tests. The switch was one of the few available that supports the data center bridging (DCB) features required for converged fabrics including support for the Fibre Channel over Ethernet (FCoE) standard.

For more information on the demo, visit the Ethernet Alliance Website and download the SC09 Whitepaper.

Comments welcome: feedback@fulcrummicro.com

In the face of a difficult economy, growing isn’t a topic you would think we would be writing about, but that’s exactly what Fulcrum has done over the past two years—at a rate higher than any other private company in the greater Los Angeles Area, according to the recent “Fastest Growing Private Companies” listing from the Los Angeles Business Journal.

The article cites Fulcrum’s culture of innovation driven by a refusal to bend to corporate norms.  Instead, the company focuses on the holistic wellbeing of its employees, recognizing the value of maintaining an environment in which engineers are comfortable and can continue to drive the innovation that has put Fulcrum in its current leadership position in the 10GbE switch semiconductor market.

“You have very big competitors and as a startup you have to be differentiated from them,” said CEO Bob Nunn.  “Our entire product line is based on this technology advantage. What’s really unique about (the company is it relies) on being the engineering brains behind designing high-tech semiconductor chip solutions.”

To read the full article, click here.

Comments welcome: feedback@fulcrummicro.com

It has been clear for some time now that SONET/SDH transport will run its course, and that future carrier transport systems will be completely packet-based. Until that time, though, these systems need to support both TDM and packet flows across the backplane. This is not an easy task.

Some approaches utilize TDM-only implementations in which the packets are segmented into TDM cells across the backplane. Other approaches utilize both a packet fabric and a TDM fabric in the backplane. Both of these solutions are very costly and do not provide a good path to the packet-only systems of the future. Which one should you chose?

During recent customers visits in Europe, I realized that many telecom system vendors are facing this quandary. How do I support legacy TDM services while designing for the packet-based future and competing with the emerging low-cost vendors?

The answer lies in FocalPoint, which can support packet and TDM flows within the same switch. FocalPoint provides a low-cost, high-bandwidth packet based fabric that has several advanced features for supporting TDM flows. This provides a unified fabric solution with a foundation based on the future (packet transport). For more information, see our new TDM Applications Note (PDF download).

Comments welcome: feedback@fulcrummicro.com

I just got back from the SC09 show in Portland, where I saw a number of interesting HPC and data center interconnect-related emerging trends. Here are a few snapshots:

• Fulcrum, Arista Networks and Ixia announced the first ever performance test of a high density (48-port) 10GbE switch, which showed that it had the low latency, throughput and data center bridging features necessary to scale data centers for cloud applications. The FocalPoint silicon inside the Arista 7148SX performed extremely well, demonstrating completely lossless, full line rate throughput. As data centers continue to expand to support cloud-based applications, port densities of 10GbE switches will continue to increase, but must do this while maintaining the very low latency and high throughput proven by this test. Full test results are available on the Arista site. We’ll also have another post with much more detail on the ramifications of this testing.

• One anticipated highlight of all SC shows is the announcement of the Top 500 supercomputer list. This year’s announcement proved dramatic with a new Cray XT5 computer taking the top spot. On the interconnect front, the InfiniBand Trade Association tried to create its own drama by pointing to a rise in the use of the InfiniBand (IB) interconnect. The association said that IB is now the interconnect technology for 23% of the Top 500 and is gaining ground on the leader, 1G Ethernet (GE). Many supercomputer designers used IB because of its latency advantage over the open interconnect standard GE, but IB is almost proprietary, as only two silicon vendors are left (and even they are looking hard at adopting 10GbE). Now, with Fulcrum’s low latency 10GbE switches with built-in data center bridging, supercomputer designers can take advantage of the economies of scale Ethernet brings. With current generation 10GbE switches competitive in latency and scale, it’s only a matter of host adapters catching up in these areas before 10GbE will push IB to the fringes of this list in the coming years.

• Cloud computing is driving many data center and HPC managers to consider whether to rent or buy HPC capabilities to service smaller workgroups or individual high-performance users. This paradigm was on display at SC09, where Cray made a big splash about its CX1 deskside “baby” supercomputer, which comes on wheels so it can be moved from office to office. The company said Dell would be reselling the computer and also that HPCWire gave it a 2009 editor’s choice award. But HPCwire also called HPC cloud technologies a technology to watch, and another industry commentator said that HPC and cloud computing are “twins separated at birth.” Gains in networking technology will continue to make cloud-based HPC a performance competitor with deskside computing and make the rent vs. buy decision even more difficult.

• Another interesting story that points to the broader industry trend of demand for low-latency switching is the emergence of blades featuring graphics processing units (GPUs). For applications like data mining, which require multiple traffic streams, the low latency of FocalPoint silicon can help isolate GPU traffic from other traffic, while still enabling data center architects to benefit from the power and cost savings offered by using GPUs.

And finally, congratulations go out to Fulcrum partner Arista Networks, which was named “Best Interconnect Product/Technology” in HPCwire’s annual Reader’s Choice Awards! Read the full article that includes all award winners in all categories on HPCwire.

Comments welcome: feedback@fulcrummicro.com.

At the Linley Data Center Ethernet Seminar this year, the CTO of Dune and the CTO of Fulcrum presented their cases for the best data center network architecture in the final session of the day. Dune argued for a telecom-like fabric with predictable bandwidth and latency with deep buffering in the core switch using external memory.

Unfortunately, storing packets in external memory can add a significant amount of latency along with added cost to the system. Fulcrum argued for an Ethernet fabric that can scale to hundreds of 10GbE ports while maintaining low cut-through latency using an efficient on-chip memory structure.

Although telecom-based fabrics can find applications in parts of the data center network, when it comes to a single unified data center fabric, Fulcrum Ethernet switches can provide a more cost effective solution—even when slightly over-provisioned to maintain non-blocking operation with corner case traffic patterns.

Comments welcome: feedback@fulcrummicro.com