January 22, 2013
SUNNYVALE, Calif., Jan. 22 – Chelsio Communications, Inc., a leading provider of Ethernet Unified Wire Adapters and ASICs, today announced Terminator 5, the fifth generation of its high performance Ethernet silicon technology. The Terminator 5 (T5) ASIC is built upon the latest iteration of Chelsio’s protocol-rich high speed network processing architecture, and leverages an industry-proven design, which has been widely deployed with more than 200 OEM platform wins and more than 400,000 ports shipped worldwide. Launching two years after the T4 ASIC, the T5 moves the architecture into 40GbE speeds, and prepares the transition to 100GbE expected by 2015.
T5 is a highly integrated, hyper-virtualized 10/40GbE controller with full offload support of a complete Unified Wire solution comprising NIC, TOE, iWARP RDMA, iSCSI, FCoE and NAT. T5 provides no-compromise performance with both low latency (sub 1usec through hardware) and high bandwidth, limited only by the PCI bus. Furthermore, it scales to true 40 Gigabit line rate operation from a single TCP connection to thousands of connections, and allows simultaneous low latency and high bandwidth operation thanks to multiple physical channels through the ASIC.
Designed for high performance clustering, storage and data networking, the T5 enables fabric consolidation by simultaneously supporting TCP/IP and UDP/IP socket applications, RDMA applications and SCSI applications at wire speed, thereby allowing InfiniBand and FibreChannel applications to run unmodified and concurrently over standard Ethernet. This will result in tremendous savings in the data center by avoiding the need for InfiniBand or FibreChannel adapters, cabling, switches and gateways. The API used for the complete software suite (Linux, Windows and FBSD) in current T4 installations is the same for the T5 chip and upcoming 100Gb capable versions, leveraging all the software investment made in T4 deployments.
T5 establishes a new milestone for SAN performance, finally moving past FC performance. With T5, true 40Gb SAN throughput, more than 3x the latest FC speeds, is now enabled via offloaded iSCSI without waiting for specialized FCoE switches. T5 is further expected to improve storage IO performance from the current record of 2.1M IOPS held by the T4 silicon.
T5 also establishes a new milestone for Ethernet clustering performance. While recent comparisons of IB-FDR and T4 iWARP RDMA show the two at parity (LAMMPS, LS-DYNA, HPL and WRF on iWARP vs InfiniBand FDR) with half the latency and 4x the bandwidth of T4, T5 is expected to exceed IB-FDR performance. Thanks to the reliability provided by hardware offloaded TCP/IP, the benefits of T5 RDMA include eliminating the scalability and cabling and noise issues associated with IB-FDR.
With a specific design focus on low latency and small packet processing performance, T5 also enables a new performance milestone for high frequency trading and other latency sensitive applications.
Given the high integration of T5, the ability to concurrently run all the different protocols at industry leading performance levels, and the rich feature set that addresses all of Ethernet’s market segments, this silicon enables OEMs to converge on a single vendor for all their connectivity needs.
The market for 40Gb Ethernet is expected to ramp quickly. Crehan Research forecasts 40Gb Ethernet server adapters and LOMs will exceed $600 million by 2017.
“The Terminator 5 ASIC is an important evolutionary step for Chelsio, bringing all the offload, virtualization, and switching capabilities of the existing T4 chip to 40Gbps performance levels,” said Bob Wheeler, senior analyst at The Linley Group. “The arrival of the T5 should accelerate the convergence of networking, storage, and clustering around 40G Ethernet.”
T5 Architectural Features
The T5 ASIC is built around a highly scalable and programmable protocol-processing engine. Much of the processing of the offloaded protocols is implemented in microcode running on a proprietary pipelined data-flow engine. The pipeline supports cut-through operation for both transmit and receive paths for minimum latency, and the transport processor is designed for wire-speed operation at small packet sizes, regardless of the number of TCP connections.
Some key features of the T5:
Availability and Pricing
The T5 is designed in 45nm SOI CMOS process technology and packaged in a 31x31mm, 899-pin FCBGA. Customer samples are available in Q1 and volume quantities in Q2. Driver development and reference design kits are available now.
About Chelsio Communications
Chelsio is a leading technology company focused on solving high performance networking and storage challenges for virtualized enterprise data centers, cloud service installations, and cluster computing environments. With its fifth generation protocol acceleration technology, Chelsio is enabling hardware and software solutions including Unified Wire Ethernet network adapter cards, unified storage software, high performance storage gateways, unified management software, bypass cards, and other solutions focused on specialized applications.
Source: Chelsio Communications
10/30/2013 | Cray, DDN, Mellanox, NetApp, ScaleMP, Supermicro, Xyratex | Creating data is easy… the challenge is getting it to the right place to make use of it. This paper discusses fresh solutions that can directly increase I/O efficiency, and the applications of these solutions to current, and new technology infrastructures.
10/01/2013 | IBM | A new trend is developing in the HPC space that is also affecting enterprise computing productivity with the arrival of “ultra-dense” hyper-scale servers.
Ken Claffey, SVP and General Manager at Xyratex, presents ClusterStor at the Vendor Showdown at ISC13 in Leipzig, Germany.
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?