Cloud computing systems today, whether open-source or used inside companies, are built using a common set of core techniques, algorithms, and design philosophies – all centered around distributed systems. Learn about such fundamental distributed computing "concepts" for cloud computing. Some of these concepts include: clouds, MapReduce, key-value/NoSQL stores, classical distributed algorithms, widely-used distributed algorithms, scalability, trending areas, and much, much more! Know how these systems work from the inside out. Get your hands dirty using these concepts with provided homework exercises. In the programming assignments, implement some of these concepts in template code (programs) provided in the C++ programming language. Prior experience with C++ is required. The course also features interviews with leading researchers and managers, from both industry and academia.
1.1. Why Clouds?
Loading...
来自 University of Illinois at Urbana-Champaign 的课程
Cloud Computing Concepts, Part 1
631 个评分
从本节课中
Week 1: Orientation, Introduction to Clouds, MapReduce
This course is oriented towards learners with similar backgrounds as juniors and seniors in a CS undergraduate curriculum. Since learners come from various backgrounds, it is critical you view this lecture AND pass the prerequisite test. This will ensure you have many of the assumed prerequisite pieces of knowledge required to enjoy this course.
与讲师见面
Indranil GuptaProfessor
Department of Computer Science
Welcome.
In this, uh, lecture series, uh, we'll be looking at,
uh, an introduction to cloud computing,
and we'll focus a little bit on,
uh, the aspects that distinguish,
uh, cloud computing from previous generations
of, uh, distributed, uh, systems.
So, uh, first of all
there's a lot of hype around cloud computing.
Uh, several, uh, forecasting agencies which,
uh, forecast mark-market trends for different, uh, markets
have said that cloud computing is large
and it's only going to grow, uh, perhaps exponentially.
Gartner, which is one of the most, uh,
popular forecasting firms,
said in 2009 that cloud computing revenue
would exceed $150 billion by 2013,
and that it would represent
a significant fraction of IT spending, uh, by 2015.
IDC in 2009 said that, uh,
in, um, the, uh, next 5 years, uh,
spending on IT cloud services would triple
and would reach several tens of billions of dollars.
Forrester said that cloud computing would go
to $240 billion as an industry by 2020.
And it's not just in industry.
Several governments
have also started relying on cloud computing,
including fedbizopps.gov, which was set up, uh,
set up many, uh, years ago by the federal, uh, government.
So obviously there is a lot of hype,
uh, but then the question is what exactly is cloud computing?
Well, there are many cloud providers,
as I'm sure, uh, many of you, uh, might have heard of.
One of the most, uh, popular cloud providers is AWS,
or Amazon Web Services.
There are also other, uh, providers here,
Microsoft Azure, uh, Google Compute Engine,
and there are a whole bunch of other companies,
including Rightscale, Salesforce, EMC, Gigaspaces,
10gen, Datastax, Oracle, VMWare, Yahoo, Cloudera,
and hundreds of more companies
that are too many to, uh, mention here.
So let's just focus on the, um, uh, AWS, uh, portion.
Uh, AWS offers a variety of services;
three of their most popular services are called
EC2, S3 and EBS.
EC2 is an elastic compute cloud,
which provides the computing services,
S3 is a simple storage service,
which provides you the ability to store data
so that you can access it from anywhere in the world,
and EBS is Elastic Block Storage,
which is essentially volume storage that
the EC2 instances can access while they are running.
So there're two categories of clouds, uh, today.
Uh, they are either public cloud or a private cloud.
A private cloud is one that is accessible only to,
uh, special privileged employees or personnel.
For instance, if you're working in a company X,
and the company X has a data center
or a cluster or a cloud
which is accessible
only to the employees of the company,
that's a private cloud; it is not openly accessible.
On the other hand, uh,
a-a public cloud is one that can be accessed by anyone,
anywhere in the world, as long as they have
a credit card and are willing to pay.
Uh, private clouds, for instance, also include, uh,
the first category also includes, uh,
you buying a data center or a cloud
and running it yourself;
that's your own private cloud, obviously.
So among public clouds,
the ones that are openly accessible, uh, um,
as I mentioned, uh,
there are a couple that are, uh, fairly popular,
Amazon AWS services, uh, Google Compute Engine,
and, uh, the, uh, Microsoft Azure tend to be popular.
Uh, the AWS and Azure share a lot of, uh, traits,
so I'll just focus on AWS here.
Uh, the S3 Simple Storage Service
allows you to store arbitrary data sets,
and you pay per gigabyte month that you store;
there are no upfront costs.
The Elastic Compute Cloud, uh, once again allows you
to upload and run arbitrary OS images.
Essentially they are- these are virtual machines,
or VMs, and you pay per CPU hours that you use.
So if you use, say,
5 VMs for 1 hour, uh, each, all at the same time,
you'd be paying, uh, for 5 CPU hours.
And similarly, the other services that,
uh, AWS or Microsoft Azure might offer, uh,
are also, uh, priced accordingly.
Uh, there are a variants of this of course.
Google AppEngine or Compute Engine
offers the ability to develop applications
within, uh, their AppEngine,
uh, and then you upload the data
that would be imported into their format,
and you run your application.
So for instance,
you could use these to host your own web services,
you could use these, um,
to host your, uh, data so that it is served out,
or you could host these to run your own computation,
uh, that you might want to run.
So, cloud computing is attractive to customers because,
uh, it allows them to save both time as well as money.
Uh, Eli Lilly, which is one of the largest, uh, um,
customers of AWS, reputedly,
uh, for instance, uh, er, uh, has been, uh, benefiting
from the ability to bring up new servers fairly quickly.
Earlier they reported that it used to take, uh,
seven and a half weeks for them to deploy a new server.
This would, uh, involve, uh, them, uh, um, uh, uh,
setting the specs for the server,
then actually purchasing it,
and then going through the purchasing process, uh,
the server comes in and they set it up, uh,
and install the necessary s- operating system and software,
and all of this takes, uh, took several weeks.
Now they can rig up a new server in, uh, around 3 minutes,
and a 64-node Linux cluster in 5 minutes,
uh, while it was taking months, uh, earlier.
This of course saves them time; it also saves them money,
because, uh, they can save the costs
that they might have otherwise had to pay the sysadmins.
GlaxoSmithKline reported that,
uh, they saved about 30% by moving to the cloud.
Uh, Jim Swartz, uh, from CI- the CIO Sybase said that,
uh, they've been saving $2 million since 2006,
and this is only going to increase.
And as a result, it's not surprising that
hundreds of startups across Silicon Valley,
uh, are leveraging, uh,
some of the cloud providers, who I mentioned,
and a variety of other cloud providers as well,
rather than buying their own machines.
But once again, this brings us back to the question
of what exactly is a cloud for us, uh, technical folks,
who want to understand the deep, detailed, uh,
aspects of cloud computing.
We'll discuss that in, uh, the next lecture.
[electronic music]
