摘要：
The answer depends on what exactly you're trying to accelerate.GPU is really a "software acceleration" that can accelerate certain class of compute-intensive applications.The only interface to the G... 阅读全文
摘要：
operating system 64 bit systems have several advantages over 32 bit systems: The number of registers is doubled. This makes it possible to store intermediate data and local variables in registers rath... 阅读全文
摘要：
多核和多核异构系统越来越受到超大规模计算机的青睐，如前不久发布的全球高性能计算TOP500排名第一的IBM的Roadrunner就是把自己的Cell和AMD的皓龙处理器混合在了一起；在芯片层面，包括英特尔、AMD和nVIDIA等都在研制异构多核或众核芯片。例如AMD提出的CPU+GPU的Fusion，nVIDIA的Tesla等。 众所周知，在当前的高性能计算系统和应用中，绝大多数是x86架构的处... 阅读全文
摘要：
http://www.agner.org/optimize/非常好的程序优化相关的resource 阅读全文
摘要：
There is a significant runtime cost for position-independent code when using dynamic linking (shared objects) in Linux, BSD and Mac systems. Critical CPU-intensive code should never be stored in a sha... 阅读全文
摘要：
Shared-memory blocks are perhaps the fastest IPC mechanism, especially for transferring large structures between processes. However, they require careful synchronization, or subtle bugs can occur i... 阅读全文
摘要：
FPGA-based system outperforms CPU and hybrid CPU/FPGA trading platforms by consolidating feed handling, trade trigger logic, and order entry into a single FPGA. Seattle, WA – September 14, 20... 阅读全文
摘要：
Time Stamp CounterFrom Wikipedia, the free encyclopediaJump to: navigation, searchThe Time Stamp Counter is a 64-bit register present on all x86 processors since the Pentium. It counts the number of t... 阅读全文
摘要：
Asio Portable networking, including sockets, timers, hostname resolution and socket iostreams. Author(s) Chris Kohlhoff First Release 1.35.0 Standard Build & Link Categories Concurrent Programmi... 阅读全文
摘要：
http://news.ddaaoo.com/Topic/view/id-57255 关于Linux的消息队列: mq_open用于日志记录中，当有新日志添加到队列中，我就mq_send一个消息，日志记录线程就开始记录数据，直至日志队列为空，然后继续等待。但是，遇到了第一个情况：近乎死锁的长时间等待！因为如果设置为阻塞的，mq_send将在发送胜利后之后返回！这很容易导致死锁！试想，如果大量的消息... 阅读全文