Fast, low energy required, low heat performance, multiple high performance core, and reduced noise, its all the concepts of new generation of today's advanced supercomputer. IBM already announced its next generation supercomputing project, Blue Gene/Q, which is designed as an advanced technology supercomputer, to solve the most challenging problems facing engineers and scientists at faster, more energy efficient, and more reliable rates than before.
courtesy image of www.deepcomputing.com
The system, named “Sequoia”, is expected to achieve 20 petaflops at peak performance, marking it as one of the fastest supercomputers in the world. Moreover, Blue Gene/Q is expected to become the world’s most power-efficient supercomputer, churning out 2 gigaflops per watt. Blue Gene/Q is expected to predict the path of hurricanes, analyze the ocean floor to discover oil, set the automatic control for advanced space airplane, simulate nuclear weapon performance, and decode gene sequences.
The third generation in the Blue Gene family of supercomputer, Blue Gene/Q operates at an order of magnitude faster than previous systems, deploying 16 multi-processing core technology and a scalable peak performance up to 100 petaflops—a massive leap forward in parallel computing power. Applicable to a growing set of computationally intensive workloads within the scientific community, Blue Gene/Q is the ideal platform for highly complex projects in a broad range of areas from nuclear energy to climate modeling.
The new IBM PowerPC® A2 processing architecture plays a key role in delivering performance. Each processor includes 16 compute cores (up from four used with Blue Gene/P, the previous system) plus a core allocated to operating system administrative functions and a redundant spare core. Blue Gene/Q incorporates architectural advances that contribute to the system’s outstanding performance and help simplify programming
In addition, hardware-based transactional memory helps programmers avoid the potentially complex integration of locks and helps eliminate bottlenecks caused by deadlocking. Hardware-based transactional memory technology helps to deliver efficient and effective multi-threading while reducing the need for complicated programming.
