加载中...

Intel® Many Integrated Core (Intel® MIC) architecture is the latest breakthrough in supercomputing speed, performance, and compatibility.

Revolutionary speed

Intel MIC architecture will help create platforms running at trillions of calculations per second using fast and familiar Intel® Xeon® processors and co-processors based on the new architecture. The leap in processing speed is nothing short of exponential. As supercomputers have just recently broken the petaflop barrier, Intel already foresees many Intel MIC processors being combined to help partners break the next big milestone: the exaflop or 1,000 petaflop barrier.

Processing highly parallel workloads

The first Intel MIC products will target applications in High Performance Computing (HPC), Workstation, and Data Center segments that use highly parallel processing. The architecture utilizes a high degree of parallelism in smaller, lower power, and single threaded performance Int

2011 年 5 月 4 日，美国加州圣克拉拉—— 英特尔公司今天宣布在晶体管发展上取得了革命性的重大突破。晶体管是现代电子设备的微小的元件。自 50 多年前硅晶体管发明以来， 3-D 结构晶体管史无前例将首次投入批量生产。英特尔将推出被称为三栅极（Tri-Gate）的革命性 3-D 晶体管设计（英特尔曾在 2002 年首次披露），并将批量投产研发代号 Ivy Bridge 的 22 纳米英特尔芯片。1 纳米是 1 米的十亿分之一。

这款 3-D 三栅极晶体管代表着从 2-D 平面晶体管结构的根本性转变。几十年来，2-D 晶体管不仅一直在计算机、手机、消费电子产品中得到了广泛应用，还用于汽车、航空、家用电器、医疗设备以及数千种日用设备的电子控制中。

英特尔公司总裁兼首席执行官欧德宁（Paul Otellini）表示：“英特尔的科学家和工程师通过采用 3-D 结构，再一次实现了晶体管的革命。随着我们把摩尔定律推进到新的领域，3-D 结构将帮助我们打造令人惊叹且能改变世界的设备。”

科学家早就意识到 3-D 结构对于延续摩尔定律

英特尔携伙伴南京启动“核芯成长计划”

原文出处：http://www.zhujiangroad.com/news/view1.asp?id=57105

　　6月16日，英特尔“核芯成长计划”正式在南京开启，英特尔携联想、惠普、华硕这三家当地最大的合作伙伴，将基于完美融合了英特尔核芯显卡的第二代智能英特尔酷睿处理器笔记本电脑带到了更多消费者面前，这是继第二代智能英特尔酷睿处理器的笔记本电脑在南京全面上市后，重点针

It is easy to see by formal-logical methods that there exist certain instruction sets that are in abstract adequate to control and cause the execution of any sequence of operation. ...The really decisive  considerations from the present point of view, in selecting an instruction, are more of a practical nature: simplicity of the equipment demanded by the instruction set, and the clarity of its application to the actually important problems together with the speed of its handling of those problems.

Burks, Goldstine, and von Neumann, 1947

AMD APU开启异构计算，实现CPU、GPU彻底融合

http://www.eet-china.com/ART_8800645095_617693_NT_2f0b18b7.HTM

关键字： APU  异构计算  Llano 

微处理器厂商AMD宣布Fusion处理器将彻底融合CPU与GPU。Fusion是AMD的旗舰处理器，整合了x86处理器和图形处理器。AMD公司院士Phil Rogers在华盛顿贝尔维尤举行的Fusion开发者大会的一个主题演讲中宣布了此消息。AMD高级副总裁、产品部门总经理Rick Bergman宣布正式发布A系列APU芯片“Llano”。

(你或许知道双核、四核甚

http://code.google.com/p/mytos/

“mytos(My Tiny OS),一个微小型的内核，集中于实现一个微小型内核，对于开始学习内核或者并没有太多内核编程的人来说阅读此源码是一个不错的选择（虽然写的不是很好，目标在于能运行就可以了），但对于不想阅读庞大的linux或者其他操作系统来说个不错的选择。 对于已经有过比较多内核编程的人来说或者大牛来说这个代码就没有什么学习意义了，推荐阅读minix2.4版本（与之相关的书籍是《操作系统设计与实现》），linux 0.12版本（《linux内核完全剖析》第二版） ”

Fallacy 1: Computers at low utilization use little power.

Power efficiency matters at low utilizations because server workloads vary. CPU utilization for servers at Google, for example, is between 10% and 50% most of the time and at 100% less than 1% of the time. Figure 1 shows power for servers with the best SPECpower results at 100% load, 50% load, 10% load, and idle. Even servers that are only 10% utilized burn about two-third of their peak power.