Big Data - Tim Smith

探索"海量数据"的前沿 - 蒂姆 . 史密斯

Translator: Andrea McDonough Reviewer: Jessica Ruby

翻译人员: Hanlin Xu 校对人员: Bighead Ge

Big data is an elusive concept.

“海量资料” 是一个让人难以捉摸的概念。

It represents an amount of digital information,

它表示了巨大的数字信息量，大到难以

which is uncomfortable to store,

存储

transport,

转移

or analyze.

或分析。

Big data is so voluminous

”海量数据“ 非常庞大以至于

that it overwhelms the technologies of the day

它颠覆了目前的科技发展，

and challenges us to create the next generation

并且挑战我们发明新一代

of data storage tools and techniques.

数据存储技术的工具和技术。

So, big data isn't new.

所以，“海量数据”不是新的话题。

In fact, physicists at CERN have been rangling

实际上，物理学家在欧洲粒子物理研究所已经为

with the challenge of their ever-expanding big data for decades.

他们不断扩大的数据库纠结了数十年。

Fifty years ago, CERN's data could be stored

五十年前，欧洲粒子物理研究所的数据可以被存储在

in a single computer.

单单一台电脑上。

OK, so it wasn't your usual computer,

好吧，那台电脑不是你现在用的普通的电脑。

this was a mainframe computer

这台电脑的主机填满了

that filled an entire building.

整个办公楼。

To analyze the data,

想要分析得到的数据，

physicists from around the world traveled to CERN

世界各地的物理学家们就得来欧洲粒子物理研究所

to connect to the enormous machine.

连接上这个巨大的机器。

In the 1970's, our ever-growing big data

在七十年代，这些不断增长的海量数据

was distributed across different sets of computers,

被分配到不同的计算机集上，

which mushroomed at CERN.

这些计算机集在研究所里迅速扩张。

Each set was joined together

每个计算机集连着

in dedicated, homegrown networks.

专属的广播网。

But physicists collaborated without regard

但是物理学家们的合作研究不能受到这些

for the boundaries between sets,

计算机集的束缚，

hence needed to access data on all of these.

他们需要访问所有的数据，

So, we bridged the independent networks together

所以，我们桥接起这些独立的计算机集

in our own CERNET.

创建了欧洲粒子物理研究所内部网络。

In the 1980's, islands of similar networks

在八十年代，说着不同语言的与此相似的网络

speaking different dialects

扩散到了整个欧洲

sprung up all over Europe and the States,

和美国，

making remote access possible but torturous.

使远程访问成为可能但是非常痛苦和麻烦。

To make it easy for our physicists across the world

为了让全球的物理学家们

to access the ever-expanding big data

更容易地拿到

stored at CERN without traveling,

这些数据,

the networks needed to be talking

这些网络必须用

with the same language.

同一种语言。

We adopted the fledgling internet working standard from the States,

我们采用了初出茅庐的美国因特网标准，

followed by the rest of Europe,

欧洲也随之采用，

and we established the principal link at CERN

之后，1989年，我们设立了欧洲和美国的首要链接

between Europe and the States in 1989,

在欧洲粒子物理研究所

and the truly global internet took off!

随后，全球因特网迅速流行起来。

Physicists could easily then access

物理学家们可以轻而易举地

the terabytes of big data

从全世界各地远程获取

remotely from around the world,

海量数据

generate results,

生成结果，

and write papers in their home institutes.

并且在他们自己的研究所里写研究报告。

Then, they wanted to share their findings

之后，他们想和所有的同行们

with all their colleagues.

分享他们的研究成果。

To make this information sharing easy,

为了让数据分享更容易，

we created the web in the early 1990's.

我们在九十年代早起发明了因特网。

Physicists no longer needed to know

物理学家们再也不用需要

where the information was stored

知道数据储存在哪里

in order to find it and access it on the web,

他们只需要上网找就可以了。

an idea which caught on across the world

这个主意被人们广泛接受了，

and has transformed the way we communicate

随之改变了我们日常生活中

in our daily lives.

人们沟通的方式。

During the early 2000's,

在二十一世纪初期，

the continued growth of our big data

“海量数据”的持续增长

outstripped our capability to analyze it at CERN,

超过了欧洲物理研究所的研究能力

despite having buildings full of computers.

（尽管他们拥有一幢幢全是计算机的大楼）

We had to start distributing the petabytes of data

我们不得不开始散步这些“拍它字节” 数据 （拍字节或拍它字节（Petabyte、PB）是一种资讯计量单位，现今通常在标示网络硬盘总容量，或具有大容量的储存媒介之储存容量时使用。）

to our collaborating partners

给我们的合作伙伴，

in order to employ local computing and storage

从而使用上百各大科学研究院的

at hundreds of different institutes.

地方计算机存储资源。

In order to orchestrate these interconnected resources

为了更好得调配这些

with their diverse technologies,

互相联系的资源

we developed a computing grid,

我们研发了一个计算机网格

enabling the seamless sharing

使无缝的全球数据分享

of computing resources around the globe.

成为可能.

This relies on trust relationships and mutual exchange.

这依赖于相互信赖的关系和互相交流。

But this grid model could not be transferred

但是这个网格模型可以轻易地被转送到这种关系之外，

out of our community so easily,

没有相互信赖的关系和互相交流，

where not everyone has resources to share

每一个人都会对自己的资源表现的很保守，

nor could companies be expected

一些公司也不会有

to have the same level of trust.

同样的信任度。

Instead, an alternative, more business-like approach

取而代之一种商业化方式的获取信息的方式

for accessing on-demand resources

在最近非常

has been flourishing recently,

流行，

called cloud computing,

那就是云技术。

which other communities are now exploiting

云技术被很多其他团体用来

to analyzing their big data.

分析他们的海量数据。

It might seem paradoxical for a place like CERN,

像欧洲粒子物理研究所这样的地方

a lab focused on the study

专注于研究小得无法想象的粒子

of the unimaginably small building blocks of matter,

却可以成为“海量数据" 的源头

to be the source of something as big as big data.

这可能会让人感觉很矛盾

But the way we study the fundamental particles,

然而，我们学习这些基本颗粒的方式

as well as the forces by which they interact,

和这些颗粒作用于彼此的作用力

involves creating them fleetingly,

包含了：短暂地创造它们，

colliding protons in our accelerators

在加速器里使它们碰撞，

and capturing a trace of them

在它们在以接近光速运动时

as they zoom off near light speed.

记录下它们的迹线。

To see those traces,

为了能很好地观察这些轨迹，

our detector, with 150 million sensors,

在探测器里，我们装了1.5亿个感应器，

acts like a really massive 3-D camera,

这些探测器就像硕大的3D照相机，

taking a picture of each collision event -

拍下每一次碰撞-

that's up to 14 millions times per second.

那是每秒钟1400万张。

That makes a lot of data.

这构成了很多数据。

But if big data has been around for so long,

如果”海量数据“已经存在了那么久，

why do we suddenly keep hearing about it now?

我们为什么现在才听说它呢？

Well, as the old metaphor explains,

老话说得好

the whole is greater than the sum of its parts,

”团结力量大“，

and this is no longer just science that is exploiting this.

不仅是科学研究在利用这个。

The fact that we can derive more knowledge

从各种信息中，我们可以通过拼接相关信息和发现关联性

by joining related information together

从而导出更多的信息。

and spotting correlations

这，让我们更消息灵通，

can inform and enrich numerous aspects of everyday life,

也可以丰富我们的日常生活。

either in real time,

无论是在实时，

such as traffic or financial conditions,

（比如信息量或金融信息）

in short-term evolutions,

在短期的演变

such as medical or meteorological,

（比如说医学或气象学）

or in predictive situations,

或者在需要预测的情况下

such as business, crime, or disease trends.

（比如说商业，犯罪，疾病发展趋势）。

Virtually every field is turning to gathering big data,

事实上，每一个领域都需要收集海量数据，

with mobile sensor networks spanning the globe,

比如遍布全球的移动感应网络，

cameras on the ground and in the air,

比如陆地或在空中都有的摄像器，

archives storing information published on the web,

比如网络信息档案集，

and loggers capturing the activities

和捕捉全球网民

of Internet citizens the world over.

网上活动的记录器。

The challenge is on to invent new tools and techniques

我们面临的挑战是去发明新的工具与新的技术

to mine these vast stores,

从而来挖掘这些巨大的存储箱，

to inform decision making,

帮助我们做正确的决定，

to improve medical diagnosis,

提高医学诊断正确率，

and otherwise to answer needs and desires

甚至推满足未来社会

of tomorrow's society in ways that are unimagined today.

尚无法想像的需求和渴望。