Asst.Prof.Dr.Surasak Mungsing

What is demand computing? Cluster, grid, and cloud computing

Examples of demand computing Demand computing handling techniques How cloud handle demand computing? Limitation of demand computing?

On-demand computing (ODC) is a business computing model that allows companies to provide access to computing resources as they become necessary, rather than full time

a computing and communications infrastructure that facilitates flexible business service delivery and provides the basis for: ◦ Autonomic computing◦ Fast response to external business-affecting changes.◦ Adaptive business processes to protect revenues

and contain costs◦ Complex interactions inside and outside of

organizational boundaries◦ Resilience against external threats such as viruses,

intrusions, and power outages.

IBM has identified five levels of autonomic computing: Basic: Highly skilled staff required; very hands-on; business is

often hostage to IT outages. Managed: Skilled staff required to interpret management data;

small amount of automation; business may still be hostage to IT outages.

Predictive: Not-so-skilled staff interpret and approve actions; more automation; business may still be hostage to IT outages, though less frequently than before.

Adaptive: Staff monitor automated changes/corrections and ensure good system performance; business may still be hostage to IT outages, but these can be contracted using service-level agreements.

Autonomic: Fully automated; staff focuses on facilitating the business in line with stated policies.

a group of linked computers, working together closely thus in many respects forming a single computer

Cluster computing category◦ High-availability (HA) clusters◦ Load-balancing clusters◦ Compute clusters

Source: http://www.rdi.ku.ac.th/Techno_ku60/res-96/index96.html

High-availability (HA): clusters are linked together to share computational workload or function as a single virtual computer

Load-balancing clusters: multiple computers are linked together to share computational workload or function as a single virtual computer

Compute clusters: Often clusters are used primarily for computational purposes, rather than handling IO-oriented operations such as web service or databases

Closed system cluster computing◦ Internal Connected ◦ high security◦ only 1 IP address, each node cannot manage

external data

Open system cluster computing◦ Direct connect to external network◦ each node has its own IP address◦ suitable for being web servers or ftp servers

a special type of parallel computing that relies on complete connected to a network by a conventional network interface, such as Ethernet.

combination of computer resources from multiple administrative domains to reach a common goal

Grid size can vary by a considerable amount

Source : http://webboard.crsc.kmitl.ac.th/forum/index.php?topic=27236.0

กริ�ด คอมพิ�วเตอริ ถู�กออกแบบมาเพิ��อการิแบ�งก�นใช้�ทริ�พิยากริท�าให้�ทริ�พิยากริทางไอท!ถู�ก ใช้�อย�างม!ปริะสิ�ทธิ�ภาพิ จึ(งเก�ดการิแสิวงห้าความริ�วมม�อริะห้ว�างกลุ่*�มว�จึ�ยต�างๆ ท��วโลุ่ก โดยห้น(�งในน�-นก.จึะม!กลุ่*�มในปริะเทศไทย ช้��อว�า กลุ่*�มไทยกริ�ดได�ก�อต�-งขึ้(-นต�-งแต�  พิ.ศ . 2544  โดย

ขึ้ณะน!-ม!สิถูาบ�นห้ลุ่�ก  16  สิถูาบ�น  ค�อ  มห้าว�ทยาลุ่�ยเกษตริศาสิตริ , สิถูาบ�นเทคโนโลุ่ย!พิริะจึอมเกลุ่�า พิริะนคริเห้น�อ, สิถูาบ�นเทคโนโลุ่ย!พิริะจึอมเกลุ่�าเจึ�าค*ณทห้าริลุ่าดกริะบ�ง, สิถูาบ�นเทคโนโลุ่ย!พิริะจึอมเกลุ่�าธินบ*ริ!, มห้าว�ทยาลุ่�ยเทคโนโลุ่ย!สิ*รินาริ!, มห้าว�ทยาลุ่�ยขึ้อนแก�น, สิถูาบ�น

เทคโนโลุ่ย!แห้�งเอเซี!ย, มห้าว�ทยาลุ่�ยศ�ลุ่ปากริ, จึ*ฬาลุ่งกริณ มห้าว�ทยาลุ่�ย,มห้าว�ทยาลุ่�ยมห้�ดลุ่,กริมอ*ต*น�ยมว�ทยา,มห้าว�ทยาลุ่�ยเช้!ยงให้ม�,มห้าว�ทยาลุ่�ยวลุ่�ยลุ่�กษณ , มห้าว�ทยาลุ่�ยสิงขึ้ลุ่านคริ�นทริ ,

ศ�นย นาโนเทคโนโลุ่ย!แห้�งช้าต�แลุ่ะศ�นย เทคโนโลุ่ย!อ�เลุ่.กทริอน�กสิ แลุ่ะคอมพิ�วเตอริ แห้�งช้าต�  ป6จึจึ*บ�นได�พิ�ฒนาเป8น “ศ�นย กลุ่างการิพิ�ฒนาเทคโนโลุ่ย! กริ�ดแห้�งช้าต� (Thai National Grid

Center)” โดยม!มต� คริม.ให้�จึ�ดขึ้(-นเม��อว�นท!� 28 ธิ�นวาคม 2547 สิ�งก�ดกริมสิ�งเสิริ�มอ*ตสิาห้กริริมซีอฟต แวริ กริะทริวงเทคโนโลุ่ย!แลุ่ะการิสิ��อสิาริ

การพั�ฒนาและประยุ กต์�ใช้�เทคโนโลยุ�กร�ดคอมพั�วเต์อร� ในประเทศไทยุ

Cloud computing means using multiple server computers via a digital network, as though they were one computer.

◦ Often, the services available are considered part of cloud computing.

Cloud computing shares characteristics with:

◦ Autonomic computing — computer systems capable of self-management

◦ Client–server model – client–server computing refers broadly to any distributed application that distinguishes between service providers (servers) and service requesters (clients)

◦ Grid computing — a form of distributed computing and parallel computing, whereby a 'super and virtual computer' is composed of a cluster of networked, loosely coupled computers acting in concert to perform very large tasks

◦ Mainframe computer — powerful computers used mainly by large organizations for critical applications, typically bulk data processing such as census, industry and consumer statistics, enterprise resource planning, and financial transaction processing

◦ Utility computing — the "packaging of computing resources, such as computation and storage, as a metered service similar to a traditional public utility, such as electricity

◦ Peer-to-peer – distributed architecture without the need for central coordination, with participants being at the same time both suppliers and consumers of resources (in contrast to the traditional client–server model)

◦ Service-oriented computing – Cloud computing provides services related to computing while, in a reciprocal manner, service-oriented computing consists of the computing techniques that operate on software-as-a-service.

self-managing computing model named after, and patterned on, the human body's autonomic nervous system

control the functioning of computer applications and systems without input from the user

The goal of autonomic computing is to create systems that run themselves, capable of high-level functioning while keeping the system's complexity invisible to the user

a type of distributed computing in which computer owners donate their computing resources (such as processing power and storage) to one or more "projects“

Costs for volunteer computing participants◦ Increased power consumption - A CPU that is

idle generally has lower power consumption than when it is active

◦ Decreased performance of the PC - If the volunteer computing application attempts to run while the computer is in use, it will impact performance of the PC

Grid Computing is computing technology that includes a combination of computer resources that offers seamless access to computing power and data storage capacity distributed over the globe

Cloud Computing: Developed from grid computing technology, cloud computing technology offers highly flexible on-demand provisioning of its resources

Grid computing is focused towards solving a computational problem whereas Cloud computing is focused towards providing 'On Demand' services.Types of services provided by Cloud Infrastructurea) SaaS  (Software as a Service)b) PaaS ( Platform as a Service)c) IaaS  (Infrastructure as a Service)

Grid Computing is possible only with applications that can be parallelized whereas Cloud computing does not have any such restriction.

In Grid, resource allocation is primarily controlled by the contract between the user (known as Virtual Organization) and allocator (known as Admin Domain). Therefore a user, irrespective of his requirement will always get the resources ONLY as agreed upon in the contract.On the other hand in Cloud there is no such restriction. Cloud supports 'On-Demand' provisioning as it follows the principle of 'Pay as you go'. Based on the user's requirements Cloud will allocate the required resources thus enabling easy scalability in order to achieve the desired SLA.