The Tolly Group tested both 10GbE members of Blade’s RackSwitch G8100 Series with Cisco’s Catalyst 4900M. The BLADE RackSwitch platforms came out on top, with more than 8X better latency performance, 2X greater throughput, 65% lower power consumption, and more than 5X better price/performance.
Comments are welcome: feedback@fulcrummicro.com
Fulcrum is pleased to announce that Aricent, a global innovation, technology and services company focused exclusively on communications, has become a member of Fulcrum’s ControlPoint Developer Alliance. This is great news for Alliance members and Fulcrum’s customer community. Not only has Aricent already assisted Fulcrum customers in integrating customer-owned software with our FocalPoint silicon, Aricent has also ported their full-featured ISS L2/L3 networking stacks to FocalPoint and our Monaco, Reno, and Vegas reference platforms. With Aricent’s support, customers have three proven options for software: (1) use their own; (2) use Fulcrum’s ControlPoint; or (3) use Aricent ISS.
Comments are welcome: feedback@fulcrummicro.com
Comments Off | 
Fulcrum News | 
Permalink
Posted by mzeile
Arista expanded its product line with two new members (7120T and 7140T) that support the 10GBASE-T interface standard and auto-negotiate down to 1-Gigabit rate. The platforms are positioned for data center server access, and offer the industry’s highest 10GBASE-T density in a 1U form factor. Arista will have the new switches on display at Interop 2009, in their booth (#1651).
Read the Arista press release…
Comments are welcome: feedback@fulcrummicro.com
Comments Off | Customers, Industry News | Permalink
Posted by mzeile
At Interop 2009 in Las Vegas (May 17-21), Fulcrum Microsystems (booth #2727) will be demonstrating Enhanced Transmission Selection (ETS), which is another IEEE Data Center Bridging feature supported by our FocalPoint switch silicon. This is a follow-on to our demonstration of IEEE Priority Flow Control (PFC) and IEEE Quantized Congestion Notification (QCN) at SC08. ETS provides special traffic classes such as storage or video with minimum bandwidth guarantees at congested egress ports. When combined with the lossless operation available with PFC and QCN, FocalPoint can classify special traffic such as storage and provide it with bounded latency.
Comments are welcome: feedback@fulcrummicro.com
Comments Off | Fulcrum News | Permalink
Posted by glee
The holy grail in switch fabric design is an output queued architecture. This has been difficult to achieve in the past due to the high bandwidth required between the switch inputs and the output queues. Because of this, most vendors implement a combined input-output queued (CIOQ) architecture which needs less core switch bandwidth, but requires extra features to avoid blocking. One way to minimize blocking is to provide virtual output queues at each ingress port, but for an N-port switch, this means N*N input queues and associated schedulers which add significant complexity. Most vendors compromise somewhere between core bandwidth and ingress complexity, but corner case blocking still remains. They also must store the packet at both ingress and egress, adding to latency and memory requirements.
The FocalPoint RapidArray memory and Nexus crossbar technology provide, for the first time, the capability to support a fully non-blocking output queued, shared memory architecture with extremely low latency. By providing a high-bandwidth core, the switch architecture can be made simpler than competing devices. This eliminates the complexity of ingress VoQs and the extra memory they require. In addition, multicast packets are only stored once, further reducing on-chip memory requirements. This efficient FocalPoint memory architecture means competing devices need larger internal packet memory to compensate, and even then cannot provide low latency solutions.
Comments are welcome: feedback@fulcrummicro.com
Comments Off | FocalPoint Architecture | Permalink
Posted by glee