Mercury 7410 User Manual

AltiVec Parallel

Vector Technology

Scalable from Two

to Hundreds of

Processors

1K CFFT in 20 

µs 

on Each Processor

125 MHz Memory

System with

Prefetch and ECC

Advanced DMA

Engine for Chained

Submatrix Moves

267 MB/s RACE++

Switch Fabric

Interconnect

Fast L2 Cache 

(250 MHz)

Embedded computing 

reaches a new level of

performance with

RACE++

®

Series

PowerPC

®

7410 daugh-

tercards from Mercury

Computer Systems.

Each PowerPC 7410 

daughtercard contains two 

500 MHz MPC7410 microprocessors

with AltiVec™ technology. These unique
microprocessors combine a modern super-
scalar RISC architecture with an AltiVec
parallel vector execution unit.

The AltiVec vector processing unit 

revolutionizes the performance of
computationally intensive applications such
as image and signal processing. Each 
vector unit can operate in parallel on up to
four floating-point numbers or up to 
sixteen 8-bit integers. This dramatically
accelerates vector arithmetic and provides
greater application performance on smaller,
less power-hungry processors.

AltiVec technology also represents a leap 

in simplifying the programming required 
to achieve high performance. Whereas 
previous DSP-based systems required 
handcrafted assembly language code for
optimal performance, easy-to-use exten-
sions to the C language provide a direct
mapping to AltiVec instructions. This 
permits developers to program more 
productively in a higher-level language,
even for critical sections of code.

Optimized Performance

With the huge increase in processing 

performance brought by AltiVec, most 
applications are no longer CPU-limited.

To keep the processor fed with ample data,
increased emphasis is placed on the 
memory system and communications fabric
that delivers data to the processor. Each
compute node on the 500 MHz PowerPC
7410 daughtercard has a dedicated fabric
interface at 267 MB/s and maximum mem-
ory speed of 125 MHz. By maximizing the
performance of the memory and the fabric
interface to the processor, Mercury has 
optimized RACE++ compute nodes for 
processing continuous streams of data.

AltiVec in RACE++ Computers

The computational power of RACE++

Series systems is built from compute nodes
comprised of processors, memory, and
interfaces to the RACE++ interconnect.
PowerPC 7410 daughtercards each contain
two compute nodes.

Each compute node (CN) consists of an

MPC7410 microprocessor with AltiVec
technology, level 2 (L2) cache, synchronous
DRAM (SDRAM), and a Mercury-designed
ASIC. This CN ASIC contains architectural
advancements that enhance concurrency
between arithmetic and I/O operations.

500 MHz PowerPC 

7410 Daughtercard