CS 525
Advanced Distributed
Systems
Spring 2011
Yeah! That’s what
I’d like to know.
Indranil Gupta (Indy)
Lecture 2
What(’s in) the Cloud?
January 20, 2011
1
All Slides © IG
Clouds are Water Vapor
• Oracle has a Cloud Computing Center.
• And yet…
• Larry Ellison’s Rant on Cloud Computing
2
The Hype!
•
•
•
•
•
Gartner - Cloud computing revenue
will soar faster than expected and will
exceed $150 billion within five years.
Forrester - Cloud-Based Email Is
Often Cheaper Than On-Premise
Email
Vivek Kundra, CTO of Obama
Government: “Growing adoption of
cloud computing could improve data
sharing and promote collaboration
among federal, state and local
governments.” E.g: fedbizopps.gov
Merrill Lynch: “By 2011 the volume of
cloud computing market opportunity
would amount to $160bn, including
$95bn in business and productivity
apps (email, office, CRM, etc.) and
$65bn in online advertising.”
IDC: “Spending on IT cloud services
will triple in the next 5 years, reaching
$42 billion and capturing 25% of IT
spending growth in 2012.”
3
Sources: http://www.infosysblogs.com/cloudcomputing/2009/08/the_cloud_computing_quotes.htm and http://www.mytestbox.com
Ha ha hype! It’s a
bunch of tripe,
since no one is
probably making or
saving money.
4
$$$
•
•
•
Ingo Elfering, Vice President of
Information Technology Strategy,
GlaxoSmithKline:
“With Online Services, we are able to reduce
our IT operational costs by roughly 30% of
what we’re spending now and introduce a
variable cost subscription model for these
technologies that allows us to more rapidly
scale or divest our investment as necessary
as we undergo a transformational change in
the pharmaceutical industry”
Jim Swartz, CIO, Sybase: “At Sybase, a
private cloud of virtual servers inside its data
centre has saved nearly $US2 million
annually since 2006, Swartz says, because
the company can share computing power
and storage resources across servers.”
Dave Power, Associate Information
Consultant at Eli Lilly and Company:
“With AWS, Powers said, a new server can
be up and running in three minutes (it used
to take Eli Lilly seven and a half weeks to
deploy a server internally) and a 64-node
Linux cluster can be online in five minutes
(compared with three months internally).
The deployment time is really what
impressed us. It's just shy of instantaneous."
5
Sources: http://www.infosysblogs.com/cloudcomputing/2009/08/the_cloud_computing_quotes.htm and http://www.mytestbox.com
Alright, alright. But
for heaven’s sake,
can someone tell
me what is a cloud?
6
What is a Cloud?
• It’s a cluster! It’s a supercomputer! It’s a
datastore!
• It’s superman!
• None of the above
• All of the above
• Cloud = Lots of storage + compute
cycles nearby
7
What is a Cloud?
• A single-site cloud (aka “Datacenter”) consists of
–
–
–
–
–
–
Compute nodes (split into racks)
Switches, connecting the racks
A network topology, e.g., hierarchical
Storage (backend) nodes connected to the network
Front-end for submitting jobs
Services: physical resource set, software services
• A geographically distributed cloud consists of
– Multiple such sites
– Each site perhaps with a different structure and
services
8
A Sample Cloud Topology
Core Switch
Top of the Rack Switch
If higher bandwidth link,
then a “fat tree” topology
Rack
Servers
9
Scale of Industry Datacenters
• Microsoft [NYTimes, 2008]
–
–
–
–
150,000 machines
Growth rate of 10,000 per month
Largest datacenter: 48,000 machines
80,000 total running Bing
• Yahoo! [Hadoop Summit, 2009]
– 25,000 machines
– Split into datacenters of 4000 machines each
• AWS EC2 (Oct 2009)
– 40,000 machines
– 8 cores/machine
• Google
– (Rumored) several hundreds of thousands of
machines
10
OK, they are massive. But it
is still called a “cluster”! And
that’s not a new concept!
11
“A Cloudy History of Time” © IG 2010
The first datacenters!
1940
1950
Timesharing Companies & Data Processing Industry
1960
Clusters
1970
Grids
1980
1990
PCs
(not distributed!)
2000
Peer to peer
systems
Clouds and datacenters
2010 12
“A Cloudy History of Time” © IG 2010
First large datacenters: ENIAC, ORDVAC, ILLIAC
Many used vacuum tubes and mechanical relays
Berkeley NOW Project
Supercomputers
Server Farms (e.g., Oceano)
P2P Systems (90s-00s)
•Many Millions of users
•Many GB per day
Data Processing Industry
- 1968: $70 M. 1978: $3.15 Billion.
Timesharing Industry (1975):
•Market Share: Honeywell 34%, IBM 15%,
•Xerox 10%, CDC 10%, DEC 10%, UNIVAC 10%
•Honeywell 6000 & 635, IBM 370/168,
Xerox 940 & Sigma 9, DEC PDP-10, UNIVAC 1108
Grids (1980s-2000s):
Clouds
•GriPhyN (1970s-80s)
•Open Science Grid and Lambda Rail (2000s)
•Globus & other standards (1990s-2000s)
13
Why did all of this happen?
14
Trends: Technology
• Doubling Periods – storage: 12 mos,
bandwidth: 9 mos, and (what law is this?) cpu
capacity: 18 mos
• Then and Now
Bandwidth
– 1985: mostly 56Kbps links nationwide
– 2004: 155 Mbps links widespread
Disk capacity
– Today’s PCs have 100GBs, same as a 1990
supercomputer
15
Trends: Users
• Then and Now
Biologists:
– 1990: were running small single-molecule
simulations
– 2004: want to calculate structures of complex
macromolecules, want to screen thousands of
drug candidates, sequence very complex
genomes
Physicists
– 2008 onwards: CERN’s Large Hadron Collider will
produce 700 MB/s or 15 PB/year
• Trends in Technology and User
Requirements: Independent or Symbiotic?
16
Prophecies
In 1965, MIT's Fernando Corbató and the other
designers of the Multics operating system
envisioned a computer facility operating “like a
power company or water company”.
Plug your thin client into the computing Utility
and Play your favorite Intensive Compute &
Communicate Application
– [Have today’s clouds brought us closer to this reality?]
17
So, clouds have been
around for decades! But
aside from massive scale
what’s new about today’s
cloud computing?!
18
What(’s new) in Today’s Clouds?
Three major features:
I.
On-demand access: Pay-as-you-go, no upfront
commitment.
–
II.
Anyone can access it (e.g., Washington Post – Hillary Clinton
example)
Data-intensive Nature: What was MBs has now
become TBs.
–
–
III.
Daily logs, forensics, Web data, etc.
Do you know the size of Wikipedia dump?
New Cloud Programming Paradigms:
MapReduce/Hadoop, Pig Latin, DryadLinq, Swift, and
many others.
–
High in accessibility and ease of programmability
Combination of one or more of these gives rise to novel
and unsolved distributed computing problems in cloud
19
computing.
I. On-demand access: *aaS
Classification
On-demand: renting a cab vs (previously) renting a car, or buying one. E.g.:
– AWS Elastic Compute Cloud (EC2): $0.086-$1.16 per CPU hour
– AWS Simple Storage Service (S3): $0.055-$0.15 per GB-month
•
HaaS: Hardware as a Service
– You get access to barebones hardware machines, do whatever you want with
them
– Ex: Your own cluster, Emulab
•
IaaS: Infrastructure as a Service
– You get access to flexible computing and storage infrastructure. Virtualization is
one way of achieving this. Often said to subsume HaaS.
– Ex: Amazon Web Services (AWS: EC2 and S3), Eucalyptus, Rightscale.
•
PaaS: Platform as a Service
– You get access to flexible computing and storage infrastructure, coupled with a
software platform (often tightly)
– Ex: Google’s AppEngine
•
SaaS: Software as a Service
– You get access to software services, when you need them. Often said to
subsume SOA (Service Oriented Architectures).
– Ex: Microsoft’s LiveMesh, MS Office on demand
20
II. Data-intensive Computing
•
Computation-Intensive Computing
– Example areas: MPI-based, High-performance computing, Grids
– Typically run on supercomputers (e.g., NCSA Blue Waters)
•
Data-Intensive
– Typically store data at datacenters
– Use compute nodes nearby
– Compute nodes run computation services
•
•
In data-intensive computing, the focus shifts from computation to the data:
CPU utilization no longer the most important resource metric
Problem areas include
–
–
–
–
–
–
–
Distributed systems
Middleware
OS
Storage
Networking
Security
Others
21
III. New Cloud Programming Paradigms
Dataflow programming frameworks
• Google: MapReduce and Sawzall
• Yahoo: Hadoop and Pig Latin
• Microsoft: DryadLINQ
• Facebook: Hive
• Amazon: Elastic MapReduce service (pay-as-you-go)
• Google (MapReduce)
– Indexing: a chain of 24 MapReduce jobs
– ~200K jobs processing 50PB/month (in 2006)
• Yahoo! (Hadoop + Pig)
– WebMap: a chain of 100 MapReduce jobs
– 280 TB of data, 2500 nodes, 73 hours
• Facebook (Hadoop + Hive)
– ~300TB total, adding 2TB/day (in 2008)
– 3K jobs processing 55TB/day
• Similar numbers from other companies, e.g., Yieldex, eharmony.com,
etc.
22
This is all confusing. Can
you give me some examples
of clouds?
23
Two Categories of Clouds
• Industrial Clouds
– Can be either a (i) public cloud, or (ii) private cloud
– Private clouds are accessible only to company employees
– Public clouds provide service to any paying customer:
• Amazon S3 (Simple Storage Service): store arbitrary datasets ,pay per GBonth stored
• Amazon EC2 (Elastic Compute Cloud): upload and run arbitrary images, pay
per CPU hour used
• Google AppEngine: develop applications within their appengine framework,
upload data which is then imported into their format, and run
• Academic Clouds
– Allow researchers to innovate, deploy, and experiment
– Google-IBM Cloud (U. Washington): run apps programmed atop
Hadoop
– Cloud Computing Testbed (CCT @ UIUC): first cloud testbed to support
systems research. Runs: (i) apps programmed atop Hadoop and Pig, (ii)
systems-level research on this first generation of cloud computing
models (~HaaS), and (iii) Eucalyptus services (~AWS EC2).
http://cloud.cs.illinois.edu
24
– OpenCirrus: first federated cloud testbed. http://opencirrus.org
Academic Clouds
• CCT = Cloud Computing Testbed
– NSF infrastructure
– Used by 10+ NSF projects, including several nonUIUC projects
– Housed within Siebel Center (4th floor!)
– Accessible to students of CS525!
• Almost half of SP09/SP10 course used CCT for their projects
• OpenCirrus = Federated Cloud Testbed
– Contains CCT and other sites
• If you need a CCT account for your CS525 experiment,
let me know asap! There are a limited number of these
25
available for CS525
Cloud Computing Testbed (CCT)
26
CCT Hardware in more Detail
•128 compute nodes = 64+64
•500 TB & 1000+ shared cores
27
Goal of CCT: Support both
Systems Research and
Applications Research
in Data-intensive Distributed
Computing
28
CCT Software Services
Accessing and Using CCT:
I. Systems Partition (64-8 nodes):
–
–
CentOS machines
Dedicated access to a subset of machines (~
Emulab), with sudo access
–
User accounts
•
•
User requests # machines (<= 64) + storage quota (<= 30
TB)
Machine allocation survives for 4 weeks, storage survives
for 6 months (both extendible)
II. Hadoop/Pig Partition and Service (64 nodes)
III. Eucalyptus Partition (8 nodes)
29
CCT Software Services
Accessing and Using CCT:
I.
Systems Partition (64-8 nodes)
II. Hadoop/Pig Partition and Service (64 nodes):
–
Looks like a regular shared Hadoop cluster service
•
•
•
•
–
User accounts
•
•
III.
Users share 64 nodes. Individual nodes not directly
reachable.
4 slots per machine
Several users are reporting stable operation at 256
instances
During Spring 09/10, 10+ projects running simultaneously
User requests account + storage quota (<= 30 TB)
Storage survives for 6 months (extendible)
Eucalyptus Partition (8 nodes)
30
CCT Software Services
Accessing and Using CCT:
I. Systems Partition (64-8 nodes)
II. Hadoop/Pig Partition and Service (64
nodes):
III. Eucalyptus Partition (8 nodes):
•
•
Based on open-source version of
Eucalyptus from UCSB (Rich Wolski)
Exports same interface as AWS EC2 and
S3.
31
CCT Software Services
•
Some Services running inside CCT
– ZFS: backend file system.
– Zenoss: Systems Monitoring. Shared with
department’s other computing clusters
– Hadoop + HDFS
– Ability to make datasets publicly available
•
How do users request an account: two-stage
process (go to http://cloud.cs.illinois.edu )
1. User account request – require background check
2. Allocation request
32
Open Cirrus Federation
Founding 6 sites
33
Open Cirrus Federation
First open federated cloud testbed
Shared: research, applications, infrastructure (9*1,000 cores), data sets
Global services: sign on, monitoring, store, etc., Federated clouds, meaning each is different
RAS
Intel
HP
KIT (de)
ETRI
Yahoo UIUC CMU
IDA (sg)
MIMOS
34
22 March 2016
Grown to 9 sites, with more to come
34
OK, so that’s what a cloud
looks like today. Now,
suppose I want to start my
own company, Devils Inc.
Should I buy a cloud and own
it, or should I outsource to a
public cloud?
35
We’ll do that next week…
• For now, it’s important to start thinking of
who’s on your project team…
Projects
• Groups of 2 (need not be same as
presentation groups). Could be 3.
• We’ll start detailed discussions “soon” (a
few classes into the student-led
presentations)
Entr. Tidbits: Selecting your Team
• Selecting your partner is important: select
someone with a complementary personality to
yours!
– Apple: Wozniak loved being an engineer and hated
interacting with people, Jobs loved making calls,
doing sales and preferred engineering much less
– Flickr: Stewart was improvisational, Fake was goaldriven
– Levchin loved to program and break things, Thiel
talked to VCs and did sales.
– Hansson says that development of Ruby on Rails
benefited from having a small team and a small
budget that kept them focused – this is why the big
giants could not beat them.
37
Next Week
• We will continue discussion of cloud
computing
– How MapReduce works
– What is PlanetLab and Emulab
– What is Grid computing
• Then we will start to discuss Basics of P2P
systems
• Please read at least one paper from each
38
session
Administrative Announcements
Student-led paper presentations (see instructions on
website)
• Start from February 10th
• Groups of up to 2 students present each class,
responsible for a set of 3 “Main Papers” on a topic
– 45 minute presentations (total) followed by discussion
– Set up appointment with me to show slides by 5 pm day
prior to presentation
– Select your topic by Jan 31st
• List of papers is up on the website
• Each of the other students (non-presenters) expected
to read the papers before class and turn in a one to
two page review of the any two of the main set of
papers (summary, comments, criticisms and possible
future directions)
– Email review and bring in hardcopy before class
– Reviews are not due until student presentations start
– Submit reviews for any 15 sessions (from 2/10 to 4/28)
39