Information class
fislal
is_3003_chapter_5a.pptxIS 3003
Chapter 5A
1
Cloud computing
Cloud computing is a set of services, such as processing of various sorts, storage, applications, DBMS, and other services that are provided in different arrays for a fee by the vendor to the user, or buyer
Virtualization is a form of cloud computing where different software capabilities may be mixed, priced, and offered to the buyer; this mixture of capabilities may represent different O/Ss, and various arrays of computational functionality
The NIST (national institute of standards and technology) defines cloud computing as having the following characteristics:
2
On demand self-service – users may obtain computing capabilities, such as server time, and network storage as needed and on their own
Ubiquitous network access – cloud services may be accessed using standard network and Internet devices, including mobile devices
Location independent resource pooling – computer resources are pooled to serve multiple users
Rapid elasticity – computing resources can be quickly provisioned (initiated), or changed to meet changing demand
Measured service – it means that charges are made based on time and resources used
3
Other definitions
Cloud infrastructure as a service – customers pay for resources (processing, storage, networking, etc.) from cloud service providers; Amazon sells its excess capacity to various users, such as Netflix
Cloud platform as a service – customers use infrastructure and programming tools supported by the cloud provider to develop their own applications
Cloud software as a service – customers use software hosted by a vendor on the vendor’s cloud
4
clouds
Public cloud – owned and maintained by a cloud service provider, and made available to the public, or industry
private cloud – is operated solely for an organization
On-demand computing – purchasers pay for only the services, and computing power they use
Green computing or green I.T. – involves the creation, use, and disposal of computational resources so as to reduce their impact on the environment
5
High performance and power savings processors – this is another way to reduce power requirements and hardware sprawl, namely to use more efficient, power–saving processors, and to plan processor loading so as to minimize hardware numbers
6
The green data center
The typical data center consumes about 100 times more power than the typical office building; power and cooling costs have skyrocketed
Companies are looking to “green” or power efficient solutions to such problems; the standard metric is Power Usage Effectiveness (PUE); it is defined as the ration of total annual power consumed by the data center divided by how much is annually by an equivalent array of I.T. equipment
The lower the ratio, the more desirable the target; 1.0 is the desired target, but 2.0 is the usual value obtained
7
Virtualization, previously discussed, is the tool of choice for green computing, because reduces the number of servers and storage devices by optimizing services
Google’s new data center has a PUE of 1.16; Yahoo’s latest data center has a PUE of 1.08; FedEx’s newest data center in Colorado is located at 6000 ft to take advantage of environmental factors to achieve a low PUE
The Carbon Disclosure Project predicted that by 2020, companies with over $1B in revenues will be able to save $12.3B and reduce annual carbon output by 200M barrels of oil
8
High performance power savings
Power requirements may also be reduced by using multi-core processors; these devices perform various tasks simultaneously on a single chip
A multicore processor is defined as an integrated circuit to which two or more processor cores have been attached to increase performance (speed), reduce power, cause more efficient processing of simultaneous tasks
N-core processors are common today; your laptop probably has a dual-core processing chip
The incentive to save power is not just part of the “green” initiative; there is a big incentive for computer companies to reduce power needs in order to preserve battery power and prolong the use of a single charge to power a mobile device
9
Autonomic computing
Some computer scientists believe that large array processing is not possible any more; so, another approach is being explored and has been for the past 20 years at least
The approach is called autonomic computing; it is roughly defined as being self-configuring systems that optimize themselves, and are self-healing, self-reconfiguring, self-optimizing
Virus, malware, firewall, and other such programs are autonomic of sorts; they automatically configure, reconfigure, defeat virus and other attacks, update themselves, and can under certain circumstances recover from nullification
10
Infrastructure software
O/S is the software that manages and controls the activities of a computer
Other O/S terms of importance are:
GUI (graphical user interface) – provides a picture of the various controls, icons, bars, and boxes that allow the user to direct the processing work of the CPU
Multitouch – allows user to implement swipe, dust, sweep, etc. motions to control various activities
Chrome O/S – a light weight O/S that is accessed through the web browser and provides applications, and data storage facilities
11
Android – open-source O/S for mobile devices; open-source means that the software package is available to all and under certain restrictions is free or very low cost; any changes made by a developer can either be uploaded back to host, or may be kept and used by the developer
UNIX – multi-user, multi-tasking O/S developed by Bell Labs in 1969; it supports a high degree of networking, and communications
Linux – similar to UNIX, but is very robust and low cost (usually free); is low cost alternative to Oracle, and expensive O/Ss; is a favorite for running back office servers and LANs; it is also an open-source product;
Other well-known open-source software is Apache HTTP web server, and Mozilla Fire Fox
12
