Mercury 7410 user manual

Summary of the content on the page No. 1

500 MHz PowerPC
7410 Daughtercard
Embedded computing To keep the processor fed with ample data,
reaches a new level of increased emphasis is placed on the
performance with memory system and communications fabric
®
RACE++ Series that delivers data to the processor. Each
®
PowerPC 7410 daugh- compute node on the 500 MHz PowerPC
tercards from Mercury 7410 daughtercard has a dedicated fabric
Computer Systems. interface at 267 MB/s and maximum mem-
ory speed of 125 MHz. By maximizing the
Each Power

Summary of the content on the page No. 2

Mercury can configure systems with hundreds of compute MPC7410 Data nodes, communicating over the second-generation RACE++ and Instruction Flow switch fabric interconnect. Merging RACE++ and AltiVec technology provides embedded computers with unprec- edented computational power. AltiVec Vector Processing Unit The AltiVec vector processing unit operates on 128 bits of data concurrently with the other PowerPC execution units. AltiVec instructions may be interleaved with other PowerPC instructions

Summary of the content on the page No. 3

With chained DMA, the DMA controller works from a the first set of crossbars in the RACEway switch fabric. The linked list in memory so that a complex chain of DMA second set of crossbars is contained on RACEway Interlink requests requires no overhead from remote modules that connect multiple motherboards. This architec- processors. DMA requests also support non-sequential ture assures each compute node direct and unblocked access access to and from local memory. This strided DMA to the RACEway

Summary of the content on the page No. 4

The TATL™ Trace Analysis Tool and Library from Mercury RACEway Broadcast Support is also available to assist during software development and The PowerPC compute node supports RACEway crossbar debugging. TATL is a minimally intrusive software logic broadcast capabilities for enhanced bandwidth utilization analyzer that allows developers to visualize multiprocessor when sending the same data to multiple nodes. This interactions. Using TATL, developers can spot bottlenecks, hardware broadcast als