Multicore Processors

In 1965, when he first set out what we now call Moore’s Law, Gordon Moore (who later co-founded Intel Corp.) said the number of components that could be packed onto an integrated circuit would double every year or so (later amended to 18 months).

In 1971, Intel’s 4004 CPU had 2,300 transistors. In 1982, the 80286 debuted with 134,000 transistors. Now, run-of-the-mill CPUs count upward of 200 million transistors, and Intel is scheduled to release a processor with 1.7 billion transistors for later this year.

For years, such progress in CPUs was clearly predictable: Successive generations of semiconductor technology gave us bigger, more powerful processors on ever-thinner silicon substrates operating at increasing clock speeds. These smaller, faster transistors use less electricity, too.

But there’s a catch. It turns out that as operating voltages get lower, a significant amount of electricity simply leaks away and ends up generating excessive heat, requiring much more attention to processor cooling and limiting the potential speed advance——think of this as a thermal barrier.

To break through that barrier, processor makers are adopting a new strategy, packing two or more complete, independent processor cores, or CPUs, onto a single chip. This multicore processor plugs directly into a single socket on the motherboard, and the operating system sees each of the execution cores as a discrete logical processor that is independently controllable. Having two separate CPUs allows each one to run somewhat slower, and thus cooler, and still improve overall throughput for the machine in most cases.

From one perspective, this is merely an extension of the design thinking that has for several years given us n-way servers using two or more standard CPUs; we’re simply making the packaging smaller and the integration more complete. In practice, however, this multicore strategy represents a major shift in processor architecture that will quickly pervade the computing industry. Having two CPUs on the same chip rather than plugged into two separate sockets greatly speeds communication between them and cuts waiting time.

The first multicore CPU from Intel is already on the market. By the end of 2006, Intel expects multicore processors to make up 40% of new desktops, 70% of mobile CPUs and a whopping 85% of all server processors that it ships. Intel has said that all of its future CPU designs will be multicore. Intel’s major competitors——including Advanced Micro Devices Inc., Sun Microsystems Inc. and IBM——each appear to be betting the farm on multicore processors.

Besides running cooler and faster, multicore processors are especially well suited to tasks that have operations that can be divided up into separate threads and run in parallel. On a dual-core CPU, software that can use multiple threads, such as database queries and graphics rendering, can run almost 100% faster than it can on a single-CPU chip.

However, many applications that process in a linear fashion, including communications, backup and some types of numerical computation, won’t benefit as much and might even run slower on a dual-core processor than on a faster single-core CPU.

多內(nèi)核處理器

1965年，Gordon Moore首次提出了今天我們所說(shuō)的摩爾定律。他(后來(lái)與人共同籌建了英特爾公司)說(shuō)，能夠封裝進(jìn)集成電路的元器件數(shù)目每年(后來(lái)修改成每十八個(gè)月)約翻一番。

1971年，英特爾的4004處理器有2300個(gè)晶體管。1982問(wèn)世的80286有134000晶體管。今天，一般的處理器有高達(dá)2億只晶體管，英特爾預(yù)定在今年晚些時(shí)候推出有17億只晶體管的處理器。

多年來(lái)，處理器的這種進(jìn)步是完全可以預(yù)測(cè)的: 一代接一代的半導(dǎo)體技術(shù)給我們帶來(lái)了在更薄的硅襯底上、工作在更高時(shí)鐘速度上的更大、更強(qiáng)的處理器。那些更小、更快的晶體管耗電也更少。

但總是有盡頭的。隨著工作電壓更低，漏電就更多，產(chǎn)生更多的熱量，就需要對(duì)處理器的冷卻給予更多的關(guān)注，這就限制了潛在的速度提高——可以把它當(dāng)作熱障。

為了突破熱障，處理器生產(chǎn)廠家采用了一個(gè)新的策略，將兩個(gè)或更多完整的獨(dú)立處理器內(nèi)核(即CPU)封裝在一個(gè)芯片上。這種多內(nèi)核處理器能直接插入主板的單個(gè)插座上，而操作系統(tǒng)把每個(gè)執(zhí)行的內(nèi)核看作一個(gè)分立的、可獨(dú)立控制的邏輯處理器。有了兩個(gè)獨(dú)立的CPU就允許每個(gè)CPU稍微運(yùn)行得慢些，從而溫度就低一些，但在多數(shù)情況下，仍能改進(jìn)機(jī)器整體的吞吐量。

從某個(gè)角度看，這種多內(nèi)核處理器只是已沿用多年的、采用兩個(gè)或更多標(biāo)準(zhǔn)CPU的多路服務(wù)器設(shè)計(jì)思想的延伸，我們只是簡(jiǎn)單地使之封裝得更小、集成更多的元器件。然而，在實(shí)踐中，多內(nèi)核策略代表著處理器架構(gòu)的重大轉(zhuǎn)變，將會(huì)在計(jì)算行業(yè)中快速流行。在同一芯片中有兩個(gè)CPU，而不是插入兩個(gè)分開的插座，極大地提高了CPU之間通信的速度，降低了等待時(shí)間。

來(lái)自英特爾的第一個(gè)多內(nèi)核CPU已經(jīng)上市。英特爾希望到2006年底，多內(nèi)核處理器在新銷售的臺(tái)式機(jī)中達(dá)到40%、在移動(dòng)CPU中達(dá)到70%、服務(wù)器中達(dá)到85%。英特爾已經(jīng)說(shuō)過(guò)，將來(lái)所有的CPU設(shè)計(jì)都將是多內(nèi)核的。英特爾的主要競(jìng)爭(zhēng)對(duì)手，包括AMD、Sun和IBM，也都把寶押在了多內(nèi)核處理器上。

多內(nèi)核處理器除了運(yùn)行溫度低、速度快，還非常適合那些操作可以分成不同線程以及并行運(yùn)行的任務(wù)。在一個(gè)雙內(nèi)核的CPU上，可以使用多線程的軟件(同時(shí)運(yùn)行數(shù)據(jù)庫(kù)查詢和圖形生成)運(yùn)行速度幾乎比單CPU芯片快了一倍。

但是，很多以線性方式處理的應(yīng)用程序，如通信、備份和某些類型的數(shù)值計(jì)算，在速度稍微慢一些的雙內(nèi)核處理器上并不能比速度更快一些的單內(nèi)核CPU上獲得更大的優(yōu)勢(shì)。