1. Cloud computing and data center networking
Zilong Ye, Ph.D.

2. What is Cloud Computing?
Cloud Computing is a general term used to describe a new class of network based computing that takes place over the Internet,
basically a step on from Grid Computing
a collection/group of integrated and networked hardware, software and Internet infrastructure (called a platform).
Using the Internet for communication and transport provides hardware, software and networking services to clients
These platforms hide the complexity and details of the underlying infrastructure from users and applications by providing very simple graphical interface or API (Applications Programming Interface).

3. What is Cloud Computing?
In addition, the platform provides on demand services, that are always on, anywhere, anytime and any place.
Pay for use and as needed, elastic
scale up and down in capacity and functionalities
The hardware and software services are available to
general public, enterprises, corporations and businesses markets

4. Cloud Summary
Cloud computing is an umbrella term used to refer to Internet based development and services
A number of characteristics define cloud data, applications services and infrastructure:
Remotely hosted: Services or data are hosted on remote infrastructure.
Ubiquitous: Services or data are available from anywhere.
Commodified: The result is a utility computing model similar to traditional that of traditional utilities, like gas and electricity - you pay for what you would want!

5. Cloud Architecture

6. Cloud Service Models Google App Engine Software as a Service (SaaS)
Platform as a Service (PaaS)
Infrastructure as a Service (IaaS)
SalesForce CRM
LotusLive
Google App Engine
IaaSdelivers computer infrastructure, typically a platform virtualization environment, as a service. Rather than purchasing servers, software, data center space or network equipment, clients instead buy those resources as a fully outsourced service.
PaaSdeliver a computing platform where the developers can develop their own applications.
SaaSis a model of software deployment where the software applications are provided to the customers as a service.
Adopted from: Effectively and Securely Using the Cloud Computing Paradigm by peter Mell, Tim Grance

7. Basic Cloud Characteristics
The “no-need-to-know” in terms of the underlying details of infrastructure, applications interface with the infrastructure via the APIs.
The “flexibility and elasticity” allows these systems to scale up and down at will
utilizing the resources of all kinds
CPU, storage, server capacity, load balancing, and databases
The “pay as much as used and needed” type of utility computing and the “always on!, anywhere and any place” type of network-based computing.

8. Basic Cloud Characteristics
Cloud are transparent to users and applications, they can be built in multiple ways
branded products, proprietary open source, hardware or software, or just off-the-shelf PCs.
In general, they are built on clusters of PC servers and off-the-shelf components plus Open Source software combined with in-house applications and/or system software.

9. Cloud Computing Characteristics
Common Characteristics:
Massive Scale
Resilient Computing
Homogeneity
Geographic Distribution
Virtualization
Service Orientation
Low Cost Software
Advanced Security
ScalabilityInfrastructure capacity allows for traffic spikes and minimizes delays.
ResiliencyCloud providers have mirrored solutions to minimize downtime in the event of a disaster. This type of resiliency can give businesses the sustainability they need during unanticipated events.
Homogeneity: No matter which cloud provider and architecture an organization uses, an open cloud will make it easy for them to work with other groups, even if those other groups choose different providers and architectures.
On-demand self-service. A consumer can unilaterally provision computing capabilities, such as server time and network storage, as needed automatically without requiring human interaction with each service’s provider.
Broad network access. Capabilities are available over the network and accessed through standard mechanisms that promote use by heterogeneous thin or thick client platforms (e.g., mobile phones, laptops, and PDAs).
Resource pooling. Multi-tenant model.. There is a sense of location independence in that the customer generally has no control or knowledge over the exact location of the provided resources but may be able to specify location at a higher level of abstraction (e.g., country, state, or datacenter). Examples of resources include storage, processing, memory, network bandwidth, and virtual machines.
Rapid elasticity. Capabilities can be rapidly and elastically provisioned, in some cases automatically, to quickly scale out and rapidly released to quickly scale in. To the consumer, the capabilities available for provisioning often appear to be unlimited and can be purchased in any quantity at any time.
Measured Service. Cloud systems automatically control and optimize resource use by leveraging a metering capability at some level of abstraction appropriate to the type of service (e.g., storage, processing, bandwidth, and active user accounts).
Essential Characteristics:
On Demand Self-Service
Broad Network Access
Rapid Elasticity
Resource Pooling
Measured Service
Adopted from: Effectively and Securely Using the Cloud Computing Paradigm by peter Mell, Tim Grance

10. Virtualization Virtual workspaces:
An abstraction of an execution environment that can be made dynamically available to authorized clients by using well-defined protocols,
Resource sharing (e.g. CPU, memory share),
Software configuration (e.g. O/S, provided services).
Implement on Virtual Machines (VMs):
Abstraction of a physical host machine,
Hypervisor intercepts and emulates instructions from VMs, and allows management of VMs,
VMWare, Xen, etc.
Provide infrastructure API:
Plug-ins to hardware/support structures
Hardware OS
App
Hypervisor
Virtualized Stack

11. Virtual Machine Monitor (VMM) / Hypervisor
Virtual Machines
App
App
App
App
App VM VM VM
Xen
Guest OS
(Linux)
Guest OS
(NetBSD)
Guest OS
(Windows)
VMWare
UML
Virtual Machine Monitor (VMM) / Hypervisor
Denali
Hardware
etc.

12. Virtualization in General
Advantages of virtual machines:
Run operating systems where the physical hardware is unavailable,
Easier to create new machines, backup machines, etc.,
Software testing using “clean” installs of operating systems and software,
Emulate more machines than are physically available,
Timeshare lightly loaded systems on one host,
Debug problems (suspend and resume the problem machine),
Easy migration of virtual machines (shutdown needed or not).
Run legacy systems!

13. What is Hadoop?
At Google MapReduce operation are run on a special file system called Google File System (GFS) that is highly optimized for this purpose.
GFS is not open source.
Doug Cutting and others at Yahoo! reverse engineered the GFS and called it Hadoop Distributed File System (HDFS).
The software framework that supports HDFS, MapReduce and other related entities is called the project Hadoop or simply Hadoop.
This is open source and distributed by Apache.

14. Fault tolerance Failure is the norm rather than exception
A HDFS instance may consist of thousands of server machines, each storing part of the file system’s data.
Since we have huge number of components and that each component has non-trivial probability of failure means that there is always some component that is non-functional.
Detection of faults and quick, automatic recovery from them is a core architectural goal of HDFS.

15. Hadoop Distributed File System
HDFS Server
Master node
HDFS Client
Application
Local file system
Block size: 2K
Name Nodes
Block size: 128M
Replicated

16. HDFS Architecture Namenode Metadata ops Client Block ops Read
Metadata(Name, replicas..)
(/home/foo/data,6. ..
Metadata ops
Client
Block ops
Read
Datanodes
Datanodes B
replication
Blocks
Rack2
Rack1
Write
Client

17. mapreduce
MapReduce is a programming model Google has used successfully to process its “big-data” sets (~ peta bytes per day)
A map function extracts some intelligence from raw data.
A reduce function aggregates according to some guides the data output by the map.
Users specify the computation in terms of a map and a reduce function,
Underlying runtime system automatically parallelizes the computation across large-scale clusters of machines, and
Underlying system also handles machine failures, efficient communications, and performance issues.

18. Large scale data splits
Map <key, 1>
<key, value>pair
Reducers (say, Count)
Parse-hash
Count
P-0000
, count1
Parse-hash
Count
P-0001
, count2
Parse-hash
Count
P-0002
Parse-hash
,count3

19. Cloud computing research topics
Virtual machine placement
One-dimensional VM placement
Multi-dimensional VM placement
VM placement in single DC
VM placement in multi DC
Virtual machine live migration
Availability-aware virtual machine placement

20. Virtual machine placement
One single physical machine can host multiple VMs as long as the capacity of the physical machine is not exceeded
VM placement: determine how to allocate VMs on physical machines
Objective: minimize the number of physical machines used
Constraint: physical machine capacity, other QoS requirements, SLA requirements
One-dimensional VM placement problem is similar to the bin-packing problem.

21. Bin Packing (1-D) Bin Packing Problem Items to be packed
The bins;
(capacity 1)
Bin Packing Problem 1 …… .5 .7 .5 .2 .4 .2 .5 .1 .6
Items to be packed

22. Bin Packing (1-D) Bin Packing Problem Optimal Packing……
Optimal Packing .5 .1 .2 .5 .6
N0 = 4 .7 .5 .4 .2

23. Next Fit Packing Algorithm
Bin Packing Problem .5 .1 .2 .5 .6
N0 = 4 .7 .5 .4 .2
Next Fit Packing Algorithm .7 .2 .2 .1 .6 .5 .5 .5 .4
N = 6

24. First Fit Packing Algorithm
Next Fit Packing Algorithm .7 .2 .2 .1 .6 .5 .5 .5 .4
First Fit Packing Algorithm .1 .5 .2 .7 .2 .5 .6 .5 .4
N = 5

25. Other approaches Best fit bin packing Load balanced bin packing
First fit with decreasing demand

26. Multi-dimensional vm placement
Not only CPU capacity is considered, but also memory and storage capacity are also considered, so it becomes a multi- dimensional bin packing problem
Possible solutions:
Map VM to a physical machine that can satisfy all the requirements
The physical machine has the highest remaining capacity in terms of the average of all the requirements

27. Vm placement in multi-datacenters
Considering not only the VM placement, but also the network consumption between different VMs.
Similar to the Virtual Infrastructure Mapping problem 35 75 e a 60 1 2
60G b
30G 40
40G
60G
40G
10G
10G d e d 6
140G 3 a
50G c 25 35
100G 30
25G 65 c 50 5 4
120G
40G b 10 60
Virtual request
Physical substrate

28. Vm migration Dynamic VM placement problem Why VM migration?
Failure
Migration to save energy
Cons: additional network consumption, overhead (the new VM should be initiated first, and get all the replicated data from the old VM before the old VM shuts down)
Objective: minimize disruption time and SLA violation
Questions: when to migrate? How to migrate?

29. Other dynamic VM placement problems
VM splitting in order to increase the utilization of the physical machine
Computing overhead, e.g., a VM with 100 computing demands can be split into two VMs, each of which has 55 computing demands
Additional network load between the VMs which are split from the original one
The VMs’ computing demand are changing over time (so best fit may not be the bet)
There may be upgrading VMs joining the application, which may need to be provisioned close to where the original VMs locate

30. Resiliency in vm placement
One working VM should have around 3 or 5 backup VMs
Data are replicated from the working VM to the backup VMs
The dataflow can follow unicast or multicast
How to choose the multicast routing to save bandwidth
How to make the dataflow reliable against failure
Dynamic VM resiliency enhancement, considering the application state, the mean time to repair