每秒万亿次浮点运算TFLOPS ？
该设备的处理速度可以达到80~100亿次浮点运算/秒，相当于一台标准PC处理能力的4~8倍。
The device can process between 8 and 10 gigaflops, or about eight to four times the processing power of a standard PC. One gigaflop equals 1 billion floating operations per second.
a quasar系统的浮点运算能力可以达到6t，并且能源利用率高达每瓦特450 m次浮点运算。
The Aquasar system achieves a performance of six Teraflops and has an energy efficiency of about 450 megaflops per watt.
整个系统由10台运行着Linux系统750的IBM服务器电源驱动，使用15万亿字节（约13970G）的内存，2880座处理器内核，可以在80万亿次浮点运算。
The system is powered by 10 racks of IBM POWER 750 servers running Linux, and uses 15 terabytes of RAM, 2, 880 processor cores and can operate at 80 teraflops.
整个系统由10台运行着Linux系统750的IBM服务器电源驱动，使用15万亿字节(约13970g)的内存，2880座处理器内核，可以在80万亿次浮点运算。
The system is powered by 10 racks of IBM power 750 servers running Linux, and USES 15 terabytes of RAM, 2,880 processor cores and can operate at 80 teraflops.
这套新系统的运算能力可达 70TFLOPS（70万亿次浮点运算） ，已经接近全世界前50强超级电脑的水平 ，新系统的运算能力 为旧有Power5系统的4倍 。
The new system has the capability of running at about 70 TFLOPS (trillion calculations per second), which puts it somewhere near the 50th most powerful computer in the world.
Microwulf经测试，运行能力为每秒26.25gigaflops，相当于每秒能进行260多亿次的双精度浮点运算。
Microwulf has been measured to process 26.25 gigaflops, or 26.25 billion double-precision floating point instructions, per second.
每秒多达2千亿次的浮点运算处理能力。
更新于(2005)-IBM的深蓝电脑据说每秒能处理227上兆次的浮点运算。
UPDATE (2005) - IBM's Blue Gene computer can reportedly process 227 trillion flops per second.
“原子”解码算法只通过一次傅氏变换即完成解码；而整数傅氏变换避免了傅氏变换中最耗时的浮点运算。
The "atomic" algorithm performs decoding process using Fourier transformation only once and the integral Fourier transformation works without floating operation.
“原子”解码算法只通过一次傅氏变换即完成解码；而整数傅氏变换避免了傅氏变换中最耗时的浮点运算。
The "atomic" algorithm performs decoding process using Fourier transformation only once and the integral Fourier transformation works without floating operation.
应用推荐