Tundra Semiconductor Corporation (TSX:TUN) is the global leader in System Interconnect providing world-class customer and technical support, leading-edge semiconductor solutions and design services to the world's foremost communications, networking, storage system, and information technology vendors. Consistently delivering on system level performance promises that reduce time to market, Tundra System Interconnect products ensure market advantage in wireless infrastructure, storage networking, network access, military, industrial automation, and information technology applications. Silicon Logic Engineering, Inc. (SLE), Tundra's semiconductor design services division, offers industry-leading ASIC and FPGA design services, semiconductor intellectual property and product development consulting.
HyperTransport-enabled products from Tundra Semiconductor Corp.
Product Type
|
Product Detail
|
I/O Bridge
|
Tundra Tsi301 HyperTransport to PCI Bridge
The Tundra Tsi301" is a performance optimized HyperTransport" bridge designed for use with MIPS processors focusing on power and cost.
The Tsi301 supports 64-bit, 66-MHz PCI operations and implements two, bidirectional 8-bit HT interfaces that provide 800 Mbyte/s of bandwidth in each direction. Up to 31 devices can be daisy-chained to build higher capacity systems with multiple PCI busses and HyperTransport-based peripherals.
|
I/O Bridge
|
Tundra Tsi308 HyperTransport to PCI-X Bridge
The Tundra Tsi308" is a HyperTransport" to PCI-X Bridge designed for cost and power sensitive MIPS processor based applications in communications, networking, and entry-level storage systems.
Each HT port operates at a frequency of up to 600 MHz DDR for both transmit and receive directions and sustains a total aggregate bandwidth up to 2.4 Gbit/s per 8-bit bidirectional HyperTransport port. Each Tsi308 HyperTransport port can be 2, 4, or 8 bits wide in both transmit and receive directions.
For increased flexibility, the Tsi308 supports one 64-bit, PCI/X, or 2-32 bit PCI/X configurable ports. A fairness algorithm allocates bandwidth among devices, thereby eliminating starvation of bridges at the end of the chain. Up to 31 devices can be daisy-chained to build higher capacity systems with multiple PCI/X buses and HT-based peripherals.
|
|
