PPKE ITK PhD and MPhil Thesis Classes

32 1. INTRODUCTION
1.5.2 Cell Blade Systems
Cell blade systems are built up from two Cell processor chips interconnected with
a broadband interface. They offer extreme performance to accelerate computeintensive
tasks. The IBM Blade Center QS20 (see Figure 1.15) is equipped with
two Cell processor chips, Gigabit Ethernet, and 4x InfiniBand I/O capability. Its
computing power is 400GFLOPS peak. Further technical details are as follows:
• Dual 3.2GHz Cell BE Processor Configuration
• 1GB XDRAM (512MB per processor)
• Blade-mounted 40GB IDE HDD
• Dual Gigabit Ethernet controllers
• Double-wide blade (uses 2 BladeCenter slots)
Several QS20 may be interconnected in a Blade Center house with max.
2.8TFLOPS peak computing power. It can be reached by utilizing maximum
7 Blades per chassis.
The second generation blade system is the IBM Blade Center QS21which
is equipped with two Cell processor chips, 1GB XDRAM (512MB per processor)
memory, Gigabit Ethernet, and 4x InfiniBand I/O capability. Several QS21
may be interconnected in a Blade Center chassis with max. 6.4TFLOPS peak
computing power. The third generation blade system is the IBM Blade Center
QS22 equipped with new generation PowerXCell 8i processors manufactured
using 65nm technology. Double precision performance of the SPEs are significantly
improved providing extraordinary computing density up to 6.4 TFLOPS
single precision and up to 3.0 TFLOPS double precision in a single Blade Center
house. These blades are the main building blocks of the world’s fastest supercomputer
(2009) at Los Alamos National Laboratory which first break through
the ”petaflop barrier” of 1,000 trillion operations per second. The main building
blocks of the world’s fastest supercomputer, besides the AMD Opteron X64 cores,
are the Cell processors. The Cell processors produce 95% computing power of
the entire system (or regarding computational task), while the AMD processors