Presentation is loading. Please wait.

Presentation is loading. Please wait.

Evolution of Distributed Computing

Similar presentations


Presentation on theme: "Evolution of Distributed Computing"— Presentation transcript:

1 Evolution of Distributed Computing
Scalable computing over the internet

2 Scalable Computing Over Internet Age of Internet Computing
Platform evolution : mainframes like IBM 360 : low cost minis – VAX : PC  VLSI :portable computers and pervasive devices : shared web resources over Internet. Using HPC and HTC in grid/ cloud env is elaborated in figure Degrees of parallelism Bit level : 4 to 8…64 bit CPU; Serial to word parallel Instruction level: simultaneous multiple inst, pipelining- branch prediction, dy sch, seculation Data level: SIMD, vector machines Task level: fine grained, difficult programming; parallel Job level: course grained; distributed

3 Web 2.0, Clouds, and Internet of Things
HPC: High-Performance Computing HTC: High-Throughput Computing P2P: Peer to Peer MPP: Massively Parallel Processors Source: K. Hwang, G. Fox, and J. Dongarra, Distributed Systems and Cloud Computing,

4 Top 10 Technologies for 2010

5 Mainly for file sharing Geographically dispersed peers
P2P Mainly for file sharing Geographically dispersed peers Autonomous nodes Decentralised Clusters Resource sharing Close to each other, Usually homogenous Centralised control, cooperative working Shared Memory Computing Parallel systems, multicore Divide and conquer synchronization Tightly Coupled High Throughput Computing Distributed Computing, loosely coupled Disparate Autonomous heterogenous systems Computation intensive long period,Sharing , single adm High Performance Computing Tightly coupled, fine grain parallelism Homogenous Systems high computing power, short period Low latency communication GRID Heterogeneous systems, HTC VO – trust groups, dynamic, cross organisational Geographically dispersed Resource sharing Scientific, distribution of work among all resources CLOUD Heterogeneous systems , HPC On demand resource provisioning over Internet Data centric with grid backbone, utility value Elastic , Business, full utilization of resources Web Services Application integration Separation of concerns Data integration, interoperability Virtualisation System integration Multi tenancy Sharing a resource among multiple clients Viewing a single system as multiple resources

6 Centralised Computing
all computer resources are centralized in one physical system. Fully shared and tightly coupled within one integrated OS. Many data centers and supercomputers are centralized systems, but they are used in parallel, distributed, and cloud computing applications High Performance Computing (HPC) Parallel computing Super computing Cloud computing High Throughput Computing (HTC) Distributed computing Grid computing Cluster computing Utility computing Many Task Computing (MTC)

7 High Performance Computing
HPC contains High Cost tightly coupled systems that are interconnected using low latency communication channels that executes parallel tasks on a particular domain which needs large amount of computing power over a short period of time Metric : Flops/sec, MB/sec Nature of Job: fine grain parallel Parallel computing, Cloud computing, Super Computing

8 Parallel Computing Parallel computing is a form of computation in which many calculations are carried out simultaneously, operating on the principle that large problems can often be divided into smaller ones, which are then solved concurrently ("in parallel"). Open MP Message passing shared mem

9 High Throughput Computing
HTC consists of loosely coupled independent systems that executes sequential jobs that requires large amount of compute power over a long period of time. Comm is through network High flux computing Metric : Operations /Month throughput Nature of Job: Coarse grain parallel, Embarrassingly parallel. Distributed computing, Grid computing, Cluster Computing, Utility computing

10 Distributed computing
A distributed system consists of multiple autonomous computers that communicate through a computer network. There are several autonomous processes, each of which has its own local memory. The processes communicate with each other by message passing. Issues: No common Clock, Load Balancing, latency Programming Languages: MPI

11 Many Task Computing MTC is similar to HTC, but it differs in the emphasis of using many computing resources over short periods of time to accomplish many computational tasks (i.e. including both dependent and independent tasks), where the primary metrics are measured in seconds (e.g. FLOPS, tasks/s, MB/s I/O rates), as opposed to operations (e.g. jobs) per month. MTC denotes high-performance computations comprising multiple distinct activities, coupled via file system operations. Tasks may be small or large, uniprocessor or multiprocessor, compute-intensive or data-intensive. The set of tasks may be static or dynamic, homogeneous or heterogeneous, loosely coupled or tightly coupled.

12 Trend towards utility computing
When the Internet was introduced in 1969, Leonard Klienrock of UCLA declared: “As of now, computer networks are still in their infancy, but as they grow up and become sophisticated, we will probably see the spread of computer utilities, which like present electric and telephone utilities, will service individual homes and offices across the country.” Utility computing focuses on a business model in which customers receive computing resources from a paid service provider. All grid/cloud platforms are regarded as utility service providers

13 Vision of utility computing

14 Hypecycle of new technologies
The expectations rise sharply from the trigger period to a high peak of inflated expectations. Through a short period of disillusionment, the expectation may drop to a valley and then increase steadily over a long enlightenment period to a plateau of productivity 2010 2 years, to 5 years , 5 to 10 years, more than 10 years, obsolete

15 Applns of HTC and HPC

16 Cyber Physical System A cyber-physical system (CPS) is the result of interaction between computational processes and the physical world. A CPS integrates “cyber” (heterogeneous, asynchronous) with “physical” (concurrent and information-dense) objects. A CPS merges the “3C” technologies of computation, communication, and control into an intelligent closed feedback system between the physical world and the information world The IoT emphasizes various networking connections among physical objects, while the CPS emphasizes exploration of virtual reality (VR) applications in the physical world. transform how we interact with the physical world just like the Internet transformed how we interact with the virtual world


Download ppt "Evolution of Distributed Computing"

Similar presentations


Ads by Google