| All-In-One daughter card for IBM BladeCenter Chelsio 10GbE Unified Wire Adapter Enables TCP, UDP, iSCSI, iWARP, and FCoE Offload over Single Unified Wire with SR-IOV, EVB/VNTag, DCB |
|
Overview
Chelsio’s T420-BCH is a dual port 10 Gigabit Ethernet Unified Wire adapter with PCI Express 2.0 host bus interface for IBM BladeCenter, optimized for cloud computing, HPC, virtualization, storage, and other data center applications.
The Unified Wire Solution
With the T420-BCH, Chelsio is enabling a unified wire for LAN, SAN and cluster traffic. This unified wire was made possible by the high bandwidth and low latency of 10GbE combined with storage and cluster protocols operating over TCP/IP (iSCSI, FCoE and iWARP respectively). In parallel, operating systems and hypervisors have incorporated native support for iSCSI and database applications are now supporting filebased storage protocols such as NFS as an alternative to SANs.
T420-BCH includes a full-fledged integrated Traffic Manager for robust flow control, traffic management and QoS.
Fibre Channel over Ethernet (FCoE) provides a transition path from legacy SANs to converged networks. Expanding its unified wire approach, Chelsio has added FCoE hardware support to the new T420-BCH adapter.
The adapter’s two ports and IEEE 802.3ad link aggregation/failover features are ideal for critical network applications that require redundancy and high-availability capabilities.
T420-BCH Ethernet-only networking reduces the data center’s cost in network adapters, cables, switches, rack space, power, equipment spares, management tools, planning, networking skills, and installation.
The fourth-generation (T4) technology from Chelsio provides the highest 10GbE performance available and dramatically lowers host-system CPU communications overhead with on-board hardware that off-loads TCP/IP, iSCSI, FCoE and iWARP RDMA processing from its host system. T420-BCH frees up host CPU cycles for useful applications. The system achieves increased bandwidth, lower latency, and lower power.
Packet Switching and Routing
T420-BCH integrates a high performance packet switch, which allows switching traffic from any of the input ports to any of the output ports (wire-to-wire) and from any of the output ports to any of the input ports (host-to-host).
Robust, Proven Solution
Subjected to thousands of hours of compatibility testing, over multiple years of stress testing by several OEM test suites and production deployments in servers, storage systems and cluster computing, Chelsio’s robust, stable protocol offload technology delivers proven performance in a wide range of environments. The T420-BCH is generations ahead of competing products.
Complete and Flexible TCP Offload
The T4 has hundreds of programmable registers for protocol configuration, RFCs, and offload control. The T420-BCH can offload TCP processing per connection, per-server, per-interface, and globally and simultaneously tunnel traffic from non-offloaded connections to the host processor for the native TCP/IP stack to process. The T420-BCH provides a flexible zero copy capability for regular TCP connections, requiring no changes to the sender, to deliver line rate performance at minimal CPU utilization.
T4 – Fourth-Generation Protocol Offload Engine
The T4 represents Chelsio’s fourth-generation TCP offload (TOE) design, third-generation iSCSI design, and second-generation iWARP (RDMA) implementation. For the server connection, the T4 chip includes a PCI Express v2.0 ×8 host interface. With support for the 5Gbps Gen2 data rate, the PCIe interface provides up to 32Gbps of bandwidth to the server. T4 also adds support for PCIe I/O virtualization. Most of T4 features are enhanced versions of those found in T3. Features that carry over from T3 include stateless offloads, packet filtering (firewall offload), and traffic shaping (media streaming).
Software Drivers
Chelsio offers a full suite of protocol software and drivers with the T420-BCH adapters. See www.chelsio.com/support for latest info. The software supports operation in both protocol-offload and non-offload modes.
Applications
Data-Center Networking
Scale up servers and NAS systems
Link servers in multiple facilities to synchronize data centers
Consolidate LAN, SAN, and cluster networks
Cloud Computing
Virtualization features to maximize cloud scaling and utilization
Runs InfiniBand, Fibre Channel apps unmodified on Ethernet
Cloud ready functional and management features
QoS and Traffic management
Networked Storage
Enable high performance NAS systems and Ethernet-based IP SANs
Develop shared-storage systems providing both file- and block-level services
High Performance Computing
Very low latency Ethernet
Increase cluster fabric bandwidth
Deploy Ethernet-only networking for cluster fabric, LAN, and SAN
Highlights
PCI Express Gen2 x8
Low Latency
Supports Up to 1M connections
Full TCP & UDP offload
TCP Chimney
Full iSCSI, FCoE offload
Full iWARP RDMA offload
EVB, VEPA, Flex10, VNTag
PCI-SIG SR-IOV
Integrated media streaming offload
HW based firewall in the cloud
Traffic filtering & management
Specifications
Host Interface
PCI Express Gen2 x8
MSI-X, MSI and support for legacy pin interrupts
Physical and Environmental
Dimensions without bracket: 4.92″ x 6.26″ or 12.49 cm x 15.90 cm
Operating Temp: 0 to 55ºC or 32 to 131ºF
Operating Humidity: 5 to 95%
Typical power consumption: 12W
Ethernet
IEEE 802.3ae (10 GbE)
IEEE 802.3z (1 GbE) compliant
IEEE 802.IP priority
IEEE 802.IQ VLAN tagging
IEEE 802.1Qbg EVB/VEPA
IEEE 802.1Qbh VNTag
IEEE 802.3x flow control
IEEE 802.3ad load-balancing and failover
Ether II and 802.3 encapsulated frames
Multiple MAC addresses per interface
Jumbo Frames up to 9.6Kbytes
Stateless Offloads
TCP/UDP checksum offload for IPv4 & IPv6
TSO, LSO, and GSO for IPv4 & IPv6
VLAN filtering, insertion & extraction
Line rate packet filtering and attack protection
Fine granularity time stamping (down to 4ns)
Ethernet Routing (packet header rewrite)
Packet Tracing and Packet Sniffing
UDP & Multicast Offload
UDP Sockets API
Low user-to-user latency
Multicast replication on ingress or egress
High Performance RDMA
Low latency and line rate bandwidth
Enhanced RDMA primitives
iWARP support on Linux OFED
Microsoft Network Direct support
iSCSI Acceleration
Full iSCSI initiator and target mode stack
CRC32 offload generation & verification
iSCSI proxy switching based on SCSI CDB
Full HBA offload
FCoE
Full HBA FCoE (Initiator or Target)
Open-FCoE Offload (Initiator)
CRC32 offload generation & verification
Ingress & Egress ACL (Access Control List)
TCP/IP Full Offload
Full TCP implementation including IPv4 & IPv6
Microsoft Chimney support
Extensive RFC compliance, fully featured stack
Full TCP Proxy between any set of connections
VLAN support up to 4096 different VLAN IDs
Load balancing and Fail-over capabilities
Virtualization
PCI-SIG SR-IOV
128 Virtual Functions (VF), 8 PF
144 port virtual switch
EVB, VEPA, Flex10, VNTag
336 virtual MAC addresses
Ordering info
Model: T420-BCH
Daughter card (CFFh) for IBM BladeCenter
Physical interface: PCIe x8, dual XAUI ports
INFORMATION IN THIS DOCUMENT IS PROVIDED IN CONNECTION WITH CHELSIO PRODUCTS. NO LICENSE, EXPRESS OR IMPLIED, BY ESTOPPEL OR OTHERWISE, TO ANY INTELLECTUAL PROPERTY RIGHTS IS GRANTED BY THIS DOCUMENT. EXCEPT AS PROVIDED IN CHELSIO’S TERMS AND CONDITIONS OF SALE FOR SUCH PRODUCTS, CHELSIO ASSUMES NO LIABILITY WHATSOEVER, AND CHELSIO DISCLAIMS ANY EXPRESS OR IMPLIED WARRANTY, RELATING TO SALE AND/OR USE OF CHELSIO PRODUCTS INCLUDING LIABILITY OR WARRANTIES RELATING TO FITNESS FOR A PARTICULAR PURPOSE, MERCHANTABILITY, OR INFRINGEMENT OF ANY PATENT, COPYRIGHT OR OTHER INTELLECTUAL PROPERTY RIGHT. CHELSIO PRODUCTS ARE NOT INTENDED FOR USE IN MEDICAL, LIFE SAVING, OR LIFE SUSTAINING APPLICATIONS. CHELSIO MAY MAKE CHANGES TO SPECIFICATIONS AND PRODUCT DESCRIPTIONS AT ANY TIME, WITHOUT NOTICE.