Cloud Computing: Trends & Opportunities Dr. B.B Prahlada Rao - - PowerPoint PPT Presentation

• Dr. B.B Prahlada Rao

prahladab@cdac.in 19 Oct 2013 Centre for Development of Advanced Computing C-DAC Bangalore, India

2nd NKN Workshop @ IISc

Cloud Computing: Trends & Opportunities

19 Oct 2013 Trends Cloud Comp- NKN WS 2013 1

Presentation Plan

• Introduction to Cloud computing
• Evolution of Technologiees: Computing, Networking
• Definitions Cloud Computing
• Characteristics of Cloud Computing
• Scientific Clouds
• C-DAC Cloud Initiatives
• Government Clouds
• Clouds -IOT
• Sensor Clouds
• Conclusions

19 Oct 2013 Trends Cloud Comp- NKN WS 2013 2

2013 Gartner Hype Cycle for Emerging Technologies

Trends Cloud Comp- NKN WS 2013 19 Oct 2013

3

Definitions- Cloud Computing

• Cloud Comp is a style of computing, where massively scalable IT

related capabilities are provided ‘as a service’ across the Internet to multiple external customers. ….. Gartner – It means Cloud should deliver a service on a subscription basis, allow access from anywhere and provide economies of scale. – Cloud will go cross vendor and feature shared resources.

Q. Is Cloud a metaphor for the Internet?

• Cloud computing is an evolution of the Internet computing or usage of the

Internet as a computing infrastructure & resource. CC is used to provide services such as Data Storage, Software Applications, and email and file exchanges

• CC is Ubiquitous network access
• CC promised infinite elastic computing resources.

19 Oct 2013 Trends Cloud Comp- NKN WS 2013

4

Cloud Computing ?

19 Oct 2013 Trends Cloud Comp- NKN WS 2013 5

Ref No: 11

Cloud Computing…………

Trends Cloud Comp- NKN WS 2013

Ref No: 11

Cloud Computing…………

Trends Cloud Comp- NKN WS 2013

The Next Revolution in IT?

Classical Computing

– Buy & Own

Hardware, SystemSW, Applications: to meet peak needs.

– Install, Configure, Test, Verify, Evaluate – Manage – .. – Finally, use it – $$$$....$(High CapEx)

Cloud Computing

– Subscribe – Use It – $ - Pay for what you use,

based on QoS

– No CapEx ,Only OpEx

Every 18 months?

19 Oct 2013 Trends Cloud Comp- NKN WS 2013

8

New IT Trends

• Ownership -> Leasing Model

– Infrastructure, platform, applications

• PC -> Internet Data Center

– Building applications for a single user  large number of concurrent consumers

• Web 2.0 & Virtualisation

– Building 3rd party IT services by composing multiple independent services

19 Oct 2013 Trends Cloud Comp- NKN WS 2013

9

Grid Computing Utility Computing SaaS Computing Cloud Computing

• Solving large

problems with Parallel Comp.

• Made

Main Stream by Global Alliance

• Offer Comput.

resources as Metered Service

• Introduced in

late 1990s

• Network-based

subscriptions to applications

• Gained

momentum in 2001

• Next-Generation

Internet computing

• Next-Generation

Data Centers

Evolution of Cloud Computing

Ref No: 6

Trends Cloud Comp- NKN WS 2013

Trends Cloud Comp- NKN WS 2013 19 Oct 2013

Promised Benefits of Public Clouds

• No upfront infrastructure investment

– No procuring hardware, setup, hosting, power, etc

• On demand access

– Lease what you need and when you need

• Efficient Resource Allocation

– Globally shared infrastructure, can always be kept busy by serving users from different time zones/regions.

• Nice Pricing

– Based on Usage, QoS, Supply and Demand, Loyalty

• Application Acceleration

– Parallelism for large-scale data analysis, what-if scenarios studies

• Highly Availability, Scalable, and Energy Efficient
• Supports Creation of 3rd Party Services & Seamless offering

– Builds on infrastructure and follows similar Business model as Cloud

11

Inside Cloud

Storage Provisioning VM Provisioning Network Provisioning Pricing Billing Pricing Accounting Monitoring

Cloud Middleware

Interface

Virtual Resources Physical Resources User Corporate

19 Oct 2013 Trends Cloud Comp- NKN WS 2013

12

Cloud Architecture

Virtual Machine (VM), VM Management and Deployment

QoS Negotiation, Admission Control, Pricing, SLA Management, Monitoring, Execution Management, Metering, Accounting, Billing

Cloud programming: environments and tools

Web 2.0 Interfaces, Mashups, Concurrent and Distributed Programming, Workflows, Libraries, Scripting

Cloud Applications: Social computing, Enterprise, ISV, Scientific, CDNs, ... Adaptive Management

Core Middleware User-Level Middleware

System level User level

Autonomic / Cloud Economy

Apps Hosting Platforms

Cloud resources

13 19 Oct 2013 Trends Cloud Comp- NKN WS 2013

Trends Cloud Comp- NKN WS 2013

Cloud Anatomy

Application Services (Services on Demand)

Gmail, GoogleCalender Payroll, HR, CRM etc Sugarm CRM, IBM Lotus Live

Platform Services (Resources on Demand)

Middleware, Integration, Messaging, Information, connectivity etc AWS, IBM Virtual images, Boomi, CastIron, Google App engine

Infrastructure as services (Physical Assets)

IBM BlueHouse, VMWare, Amazon EC2, Microsoft Azure, Sun Parascale …

Three models based on the type of service offered to the user

19 Oct 2013

14

Cloud Deployment Models

Private/Enterprise Clouds

Cloud computing model run within a company’s

• wn Data Center /

infrastructure for internal and/or partners use.

Public/Internet Clouds

3rd party, multi-tenant Cloud infrastructure & services: * available on subscription basis (pay as you go)

Hybrid/Mixed Clouds

Mixed usage of private and public Clouds: Leasing public cloud services when private cloud capacity is insufficient

Trends Cloud Comp- NKN WS 2013

Technology Components for Cloud Infrastructure

Technology Components :

• Virtualization
• Provisioning Automation and Scheduling
• SLA Management– Computing, Storage and Network latencies /bandwidths
• Billing and Charging
• Data-Center Power/ Cooling/ Space

Trends Cloud Comp- NKN WS 2013 19 Oct 2013

16

Traditional Arch. Vs Virtualization Arch.

Hardware Operating System App App App

Traditional Stack

Hardware App OS App App Hypervisor

Virtualized Stack

OS OS Virtual Architecture

• Many OS to one server
• Separation between SW and HW
• Many applications on one server
• Typical load on server is 80 –95%
• Dynamically optimized resources

Traditional Architecture

• One OS on one Server
• Tightly coupled HW and SW
• One Application on One Server
• Typical load on server is 5 –15 %

19 Oct 2013 Trends Cloud Comp- NKN WS 2013

17

• Most Important Component.
• Thin Layer Software
• Allow Multiple OS’s to run Concurrently.
• Dynamically Partitioning and Sharing the physical resources such as CPU,

storage, memory and I/O devices.

• Abstracts the Hardware from the VM’s.

Hypervisors

Scheduler Hypercall CPU Virt MMU Virt

I/O Virt

Control Interface

Network Virt

Hypervisor Components Interrupts

19 Oct 2013 Trends Cloud Comp- NKN WS 2013

18

Cloud Middleware

Cloud Middleware A software used to integrate

services, applications and content available on the cloud.

Key Features:

Management and monitoring Resource Scheduler Data management Identity / security management Service hosting, mediation and management User interfaces and portals Billing and metering

19 19 Oct 2013 Trends Cloud Comp- NKN WS 2013

Cloud Applications

• Scientific/Tech Applications
• Business Applications
• Consumer/Social Applications

Science and Technical Applications Business Applications Consumer/Social Applications

19 Oct 2013 Trends Cloud Comp- NKN WS 2013 20

Scientific Clouds

19 Oct 2013 21 Trends Cloud Comp- NKN WS 2013

Penguin Computing (POD)

http://www.penguincomputing.com/services/hpc-cloud/pod

• Provides HPC on Demand
• Powered by Scyld Cloud Management Platform.
• Storage System is based on the distributed open-source storage

system Ceph which supports file-based, block-based and object- based storage.

• Management of Virtual Servers leverages OpenStack (Open-Source

solution) for creating and managing virtual servers in a cloud environment Benefits:

• Ease of use: Compute resources are available immediately
• True HPC technology Including GPUs, low-latency interconnects,

and a non-virtualized compute stack

• Multiple PODs: Different environments for different industry needs

19 Oct 2013 22 Trends Cloud Comp- NKN WS 2013

OSDC (Open Science Data Cloud)

OSDC Provides :

• Petabyte-scale Data Cloud infrastructure
• Comput’l resources & services for scientists to analyse , manage, and share data
• Platform for data intensive science so that new types of data intensive algorithms can be

Developed Tested and used over large amounts of heterogeneous scientific data . OSDC Projects :

• Bionimbus - collaboration between the Institute for Genomics and Systems Biology

(IGSB) and the OSDC to develop open source technology for managing, analysing, transporting, and sharing large genomics datasets in a secure and compliant fashion.

• Conte Cloud - is cloud computing infrastructure designed to store genomics, electronic

medical records, and other sensitive data in a secure and compliant environment.

– The mission is to apply integrated informatics and mathematical modeling to predict genetic and environmental factors underlying mental health and illness .

• Bookworm - Offers a way to interact with digitized book content and full text search .

Bookworm uses ngrams extracted from books, in the public domain and integrates library metadata, including genre, author information, publication place and date.

• Matsu: open source technology for cloud-based processing of satellite imagery to

support the earth sciences.

19 Oct 2013 23 Trends Cloud Comp- NKN WS 2013

AWS -Amazon Web services

• Amazon Elastic Compute Cloud (EC2): offers two cluster

instance types:

– cluster compute instances and – cluster graphics processing unit (GPU) instances. – Provides scalable virtual private servers using xen hypervisor.

• Amazon Simple Storage Service (S3):

– Provides Web Service based storage.

• Amazon Elastic MapReduce (EMR): Allows processing of vast

amounts of data using a hosted Hadoop framework running

• n the web-scale infrastructure of EC2 and Amazon S3

19 Oct 2013 24 Trends Cloud Comp- NKN WS 2013

Windows Azure

• Cloud computing platform and infrastructure by Microsoft for building,

deploying and managing applications and services . SERVICES :

• Virtual machines(IAAS) - Virtual machines let developers migrate

applications and infrastructure without changing existing code, and can run both Windows Server and Linux virtual machines .

• PAAS : Customers can create web sites in PHP, .NET and Node.js, or

select from several open source applications from a gallery to deploy.

• Windows Azure can be used to build a web application that runs and

stores its data in Microsoft datacenters.

• Hadoop services through HDInsight: Deploys and provisions Apache

Hadoop clusters in the cloud, providing a software framework designed to manage, analyze and report on big data.

• Various database services for data management .

19 Oct 2013 25 Trends Cloud Comp- NKN WS 2013

Google Cloud Platform

• Google Cloud Storage : online file storage web service for storing and accessing your data
• n Google's infrastructure. The service combines the performance and scalability of Google's

cloud with advanced security and sharing capabilities. It is an Infrastructure as a Service comparable to Amazon S3 online storage service .

• BigQuery : Enables interactive analysis of massively large datasets working in conjunction

with Google Storage . It is an Infrastructure as a Service that may be used complementarily with MapReduce. To use the data in BigQuery, it first must be uploaded to Google Storage .

• Google Compute Engine(IAAS) : Google Compute Engine uses KVM as the hypervisor, and
• nly supports guest images running Linux . Every Google Compute Engine instance starts with a

disk resource. Depending on the selected machine type, the instance may start with scratch disk space, persistent disk space, or both .

• Google Cloud SQL : Mysql on the cloud

Google Cloud SQL is a web service that allows you to create, configure, and use relational databases that live in Google's cloud . It is Fully managed and No worrying required about tasks such as replication, patch management, or backups. These are all taken care of.

• http://en.wikipedia.org/wiki/BigQuery
• http://en.wikipedia.org/wiki/Google_Storage

19 Oct 2013 26 Trends Cloud Comp- NKN WS 2013

Future Grid

• FutureGrid Project supports several clouds, distributed among five sites, in

aggregate providing the capacity of over a thousand cores.

• FutureGrid clouds are configured with Nimbus, OpenStack and Eucalyptus, which

support interfaces that are roughly compatible with AWS EC2/S3, allowing users to move between clouds relatively easily.

• FutureGrid testbed provides capabilities that allow users to experiment with open-

source cloud middleware and virtualization platforms .

• FutureGrid Project Makes it possible for researchers to tackle complex research

challenges in computer science related to the use and security of grids and clouds. These include topics ranging from authentication, authorization, scheduling, virtualization, middleware design, interface design, and cybersecurity, to

• ptimization of grid-enabled and cloud-enabled computational schemes for

researchers in astronomy, chemistry, biology, engineering, atmospheric science and epidemiology.

• FutureGrid provides Hadoop platform and the iterative map/reduce Twister

plaftorm to allow users to experiment with MapReduce applications and middleware . Ref:

• http://scienceclouds.org/infrastructure-clouds/
• http://scienceclouds.org/infrastructure-clouds/futuregrid/

19 Oct 2013 27 Trends Cloud Comp- NKN WS 2013

RED Cloud

Red Cloud is a subscription-based cloud computing service for research and education. Red Cloud services :

• Red Cloud is an Infrastructure as a Service (IaaS) that runs Eucalyptus . Subscribers have root

access to virtual servers and virtual disks.

• Red Cloud with MATLAB” is a Software as a Service (SaaS) that runs MATLAB Distributed

Computing Server and features NVIDIA GPUs. Subscribers program applications on their desktops using their licensed copy of Parallel Computing Toolbox™ and then scale up to Red Cloud with MATLAB using MATLAB Distributed Computing Server.

• The user selects cac scheduler as the MATLAB parallel scheduler. This will cause

parallel/distributed jobs to be submitted to Red Cloud with MATLAB for execution

• Both “Red Cloud” and “Red Cloud with MATLAB” are available to researchers at other

academicinstitutions.

• Industry access is limited to “Red Cloud” as an IaaS .

Ref:

• http://www.cac.cornell.edu/redcloud/
• http://www.cac.cornell.edu/wiki/index.php?title=Red_Cloud_with_MATLAB
• http://www.cac.cornell.edu/RedCloud/FAQ.aspx

19 Oct 2013 28 Trends Cloud Comp- NKN WS 2013

Stratus

http://www.cloudbus.org/ccgrid2012/posters/abstract_stratus_pelle_jakovits.pdf

• Stratus is a new framework based on the BSP model

(Bulk Synchronous Parallel)

• Developed at University of Tartu as part of the Scientific

Computing in the Cloud (SciCloud) project.

• Bulk Synchronous Parallel (BSP) is a distributed computing

model for iterative algorithms that are very often used in scientific simulations.

• Stratus framework: is to provide a Distributed computing

platform and tools for performing large scale Scientific computing simulations and experiments.

19 Oct 2013 29 Trends Cloud Comp- NKN WS 2013

StratusLab

• StratusLab Toolkit: open source cloud distribution Customized

Environments Current version StratusLab is 1.4

• StratusLab project is to create a private cloud distribution that

permits resource center administrators to deploy grid services over the cloud’s virtualized resources

• StratusLab is aimed at service provisioning, networking and research
• f cloud and virtualization technologies to simplify and optimize the

existing distributed computing infrastructures like the European Grid Infrastructure (EGI). Grid Services on the Cloud:Gri

rid d serv rvices ces Depl ploy

• ying over

er virt rtualiz ized ed resour

• urces

ces would ld allow llow easy sy (re) e)depl deploy

• ymen

ent even en if some e node de crashes shes.

Customized Environments : Individual ividual use sers s are also so likely ely to requir ire e cust stom

• miz

ized ed envir ironm

• nments

ents containing, taining, eg, g, their r own proprietary ietary so softw tware.t e.the e ability ility to creat eate e virtua ual l machines achines Provision of Common Appliances: Prov

• vidin

ding simple ple, , sto tock ck images es of f common

• n opera

perating syste stems s lowers ers the e barr rrier er to crea eatin ing customi stomized zed images s as well ll as impr prov

• ves

es the e utili ility ty of the e clo loud d infr frastr astruct cture

19 Oct 2013 Trends Cloud Comp- NKN WS 2013 30

Scientific Cloud: HPC as a Service

• Cluster as an Infrastructure Service (IaaS)
• Provisioning of on demand HPC clusters
• Provisioning of Map reduce clusters to process large data sets
• Support for MPI as a tightly-coupled applications category.
• Design methodology for providing/supporting installation of MPI

libraries in multiple machines simultaneously

• Storage as a Service (StaaS): CloudVault
• SaaS:

– Job Submission Portal (C_JSP) , Cloud Access Portals – Monitoring tools

• PaaS: Program development and analysis Tools
• Applications as Service (AaaS) PSEs

– PSE for Climate Modelling – PSE’s for Bio Informatics

• Types of applications supported

– Sequential & Concurrent Applications – Data Parallel Applications - Map Reduce – Parallel Programs: OpenMP ,Pthreads, MPI

19 Oct 2013 Trends Cloud Comp- NKN WS 2013

31

Nimbus: set of open source tools

Mission: To evolve the infrastructure with emphasis on the needs of science

• Provides Infrastructure-as-a-Service (IaaS) Cloud Comp solution.

Features :

– Open Source IaaS – Storage Cloud Service – Remote Deployment & lifecycle management of VMs – Compatibility with Amazon EC2 & S3 – Support for X509 credentials – Easy to use cloud client – Fast Propagation (LAN torrent) – Per user quota allocation – Easy user management – One click clusters (contextualization) – VM network configuration – Local resource management plugin

32 19 Oct 2013 Trends Cloud Comp- NKN WS 2013

• GridGain is a software middleware that enables development of

HPC& data intensive distributed applications for real-time Big Data processing

• Java based
• Open source

GridGain

33 19 Oct 2013 Trends Cloud Comp- NKN WS 2013

Openstack Features

• NASA & Rackspace (Open source-2010),

3386+ people & 183+ Companies all over the world

• Leverage Resource pools of on-demand
• Focuses HPC deployments
• Dashboard to allows cloud

administrators to create & manage projects

• Simultaneously launch and manage

multiple resource instances

• Users can launch and customize instances

in the OpenStack Dashboard within the limits their administrators have set.

34 19 Oct 2013 Trends Cloud Comp- NKN WS 2013

• Open source Tech to Build Ubiquitous public & private cloud computing

platform

• Open source, Open design ,Open development, Open community

Open Cirrus

• Open Cirrus is an open cloud-computing research testbed.
• It is shared : Researches , applications , infrastructure, data sets.
• Global services: sign on, monitoring, store. Open source stack

(prs, tashi, hadoop).

• Sponsored by HP, Intel, and Yahoo! (additional support from NSF).
• 15 sites currently, target of around 20 in the next two years.

35 19 Oct 2013 Trends Cloud Comp- NKN WS 2013

Services Provided By Open Cirrus Sites

• PRS Service : The lowest level service is based on the notion of a physical

resource set (PRS).

• Cluster Management Services : It runs several different cluster management

services like CaaS , Tashi.

• Application Framework Service: Open Cirrus sites also provide higher level

services, such as Hadoop, Pig, and MPI, that support user-level applications and services.

C-DAC Cloud Initiatives

• Pan C-DAC Cloud Computing-Initiative:

– Meghdooth: Open Source Cloud Stack – Scientific Cloud Computing: Scientific Cloud Stack, IaaS, StaaS, PaaS – E-Sikshak: C-DAC’s E-Learning on Cloud

– MANTRA: MAchiNe Assisted TRAnslation System over CLOUD

– Disaster Recovery: (DR) solution on Cloud – eSanjeevani : Web based Telemedicine Appn, CDAC Mohali. – Megh Shushrut: C-DAC Hospital Mgmt Information System

• Pan C-DAC Cloud Infrastructure
• Nano Computing Clouds

19 Oct 2013 Trends Cloud Comp- NKN WS 2013 36

Meghdoot Stack

19 Oct 2013 Trends Cloud Comp- NKN WS 2013 37

• One stop solution for

implementing Cloud environment.

• One single middleware

bundle that incorporates all pre-requisites to establish a complete functional Cloud

• Provision to offer IaaS,

PaaS, SaaS

• Free and Open source
• Based on Open standards
• Compliance to cloud

recommendations by International bodies.

C-DAC Scientific Cloud

Scientific Cloud is a model for on-demand access to a shared pool of HPC resources (e.g.,networks, servers, storage, applications) that can be easily provisioned as and when needed by the researchers and scientists. Benefits of Scientific Cloud

• Easy access to large infrastructures
• On demand access to HPC resources
• Virtual ownership of cloud resources
• Ease of deployment
• Scientific Cloud provide an alternative for HPC
• utreach

19 Oct 2013 Trends Cloud Comp- NKN WS 2013 38

Scientific Cloud: HPC as a Service

• Cluster as an Infrastructure Service (IaaS)
• Provisioning of on demand HPC clusters
• Provisioning of Map reduce clusters to process large data sets
• Support for MPI as a tightly-coupled applications category.
• Design methodology for providing/supporting installation of MPI

libraries in multiple machines simultaneously

• Storage as a Service (StaaS): CloudVault
• SaaS:

– Job Submission Portal (C_JSP) , Cloud Access Portals – Monitoring tools

• PaaS: Program development and analysis Tools
• Applications as Service (AaaS) PSEs

– PSE for Climate Modelling – PSE’s for Bio Informatics

• Types of applications supported

– Sequential & Concurrent Applications – Data Parallel Applications - Map Reduce – Parallel Programs: OpenMP ,Pthreads, MPI

19 Oct 2013 Trends Cloud Comp- NKN WS 2013

39

SuMegha Cloud Stack

19 Oct 2013 40 Trends Cloud Comp- NKN WS 2013

SciCloud Service Offerings

19 Oct 2013 Trends Cloud Comp- NKN WS 2013 GUI

IaaS

PaaS SaaS

Cloud Portal

( Selects the Resource)

Virtual Machine

(OS, Time, Type)

Virtual Cluster

(No. of nodes, type, Time)

Hadoop Cluster

(No. of nodes, Time)

MPI Cluster

(No. of nodes, Time)

Cloud Vault (StaaS) Web interface

Smal l Medium Large

PSEs for

Bioinformatics Climate Modelling

Job Submission Portal

41

Cloud Vault

End-User

Desktop Application Web Application Mobile Application

Swift

Storage and Data Center

Physical

Storage OpenStack Object Storage

Cloud Vault is your personal hard drive in the

• cloud. Store any kind of file like Documents,

Music, Videos, Photos, and Virtual Machine Images on CDAC Cloud Vault secure servers.

19 Oct 2013 Trends Cloud Comp- NKN WS 2013

42

CloudVault: C-DAC Cloud StaaS- Features

• C-DAC Cloud Vault: Storage As a Service.
• Organizations can use CloudVault to store Large data efficiently,

safely, and cheaply.

• Capable of storing billions of Storage objects across Distributed

Storage Servers.

• Extremely Scalable in terms of Storage Size (Several Petabytes)

and Storage Capacity (Number of objects).

• User can Upload, Download, file or Directory irrespective of

Size/Type.

• Users can access CDAC CloudVault via GUI or Command Line or

through APIs.

• Similar to Amazon Web Services (S3).

19 Oct 2013 43 Trends Cloud Comp- NKN WS 2013

Dissemination: Workshop(s) / Seminar(s) Training

• Prahlada Rao B.B., Payal Saluja, Vineeth Simon Arackle, Half-Day Tutorial “Cloud Computing for HPC” C-

DAC’s National Conference on Parallel Computing Technologies (ParCompTech2013), NIAS, Bangalore, 21-23rd Feb 2013.

• Prahlada Rao B.B, Payal Saluja, Neetu Sharma, Ankit Mittal, Shivay Veer Sharma, “Cloud Computing for IOT

& Sensing Based Applicationss”, in IEEE International Conference on Sensing Technologies, ICST 2012, Kollkata, India, Dec 2012.

• Prahlada Rao B.B., Payal Saluja, Vineeth Simon Arackle, Half-Day Tutorial “Open Source Platforms for

Scientific Clouds” CSI’s Annual Conference on Advances in Cloud Computing (ACC2012), 26th July 2012, Bangalore.(60+ participants from Industry, Academia)

• Payal Saluja, Prahlada Rao, Ankit Mittal, Rameez Ahmad, “CDAC Scientific Cloud: On Demand Provisioning of

Resources for Scientific Applications”, in Proceedings of 18th International Conference PDPTA 2012, Las vegas, Nevada, USA, 16-19 July 2012.

• Prahlada Rao B.B., Given a Tutorial On “Cloud Computing”, International Conference on Trends

in Knowledge and Information Dynamics, ICTK 2012, organized by ISI, NIMHANS Conv Centre, Bangalore, India, 10 July 2012.(150 participants from Industry, Academia)

• Dr. Prahlada Rao BB, Cloud Computing for Telemedicine, National Workshop on Telemedicine Today &

Tomorrow, C-DAC Mohali, An Invited Talk, 6th July 2012.

• Prahlada Rao B. B. , National Workshop on Cloud Computing, conducted at CDAC Knowledge Park,

Bangalore, 21st May 2011 .(80+ participants from Industry, Academia)

• Vineeth Simon Arackal, Arunachalam B, Payal Saluja, Prahlada Rao B B, Co-existence of Cloud and Grid: A

Case Study in Service Oriented Grid GARUDA. The International Conference on Information Technology- New Generations (ITNG 2011), Las Vegas, USA. 11-13April 2011.

• Vineeth Simon Arackel, “ Coexistence of Grid and Cloud” at "Grid To Cloud (G2C)- Challenges and

Applications" Workshop at VIT, Vellore, 2nd April 2011. 19 Oct 2013 Trends Cloud Comp- NKN WS 2013 44

Cloud Vault: Features

• Reliable Storage 24x7
• Storage Access through Internet :
• Requested MB/GB Storage
• Required duration
• Interfaces

 GUI - Access cloud vault files

• File /Directory Operations :

– Create Directory – Files /Directories Listing – File/ Directories Upload/ Download – File Editing – Files Synchronization

 Java APIs

• Authentication/Login
• File Operations

– Listing – File upload – File Download

Functionality:

• Virtual image storage (used for provisioning
• f IaaS)
• Storage of any type of file including mpeg,

jpeg, videos

• Unstructured Data Storage (NO-SQL

database)

• Backup, Archive/ Retention
• Scientific Databases in the cloud

Technologies & Softwares:

– Openstack Swift Storage – GlusterFS – Cyberduck – Rackspace Storage APIs

19 Oct 2013 Trends Cloud Comp- NKN WS 2013

45

SuMegha Lab Kit

For Colleges & Research Organizations

19 Oct 2013 46 Trends Cloud Comp- NKN WS 2013

SuMeghaLab: Cloud Lab Kit

Components :

• Hypervisor

 Xen

• Cloud Middleware

 Nimbus, Openstack Swift

• Storage

 Glusterfs

• Portal

 SuMegha portal  Job Submission Portal  Cloud Vault Portal

• PSE

 PSE for SFM  PSE for NGS Pipeline

Supported Images:

• Virtual Machines

 CentOS (small, large, medium)  Job Submission Portal

• Virtual Clusters

 MPICH enabled with job submission portal  HADOOP enabled

• PSE enabled images

 SFM  NGS Pipeline

19 Oct 2013 47 Trends Cloud Comp- NKN WS 2013

• Packaged Software to Create automated deployment of Private Clouds.
• Suitable for Academic institutes & Research organizations to setup cloud

labs, by users having no expertise of system administration.

SuMegha Lab: Cloud Lab Kit

1. SuMegha Lab- Basic – 1 node+ Storage – 2 node+ Storage – 3 node + Storage

• 2. SuMegha Lab- Advance

– VC with 1-16 node + Cloud Storage – MPI Applications Framework – Hadoop Applications Framework

• 3. Cloud Storage
• 4. SuMeghaLab- PSEs

– Cloud based PSE for SFM – Cloud based PSE for NGS

• 5. C-DAC Cloud Servises:

– Free Softwares – Installations & Trainings – Support & Advise

19 Oct 2013 48 Trends Cloud Comp- NKN WS 2013

MEGHA: A Cloud interface for GARUDA

• Provides virtual Infrastructure on Garuda resource

– Supports IaaS and SaaS – Supports various OS flavors RHEL, Ubuntu and Fedora.

• Provides Virtual Clusters Inbuilt with

– clustering tools – Parallel paradigm like Mpi & OpenMP. – Torque as scheduler

• Provides on demand Scientific Application like

– SFM : Seasonal Forecasting Model, – FastA : Protein Sequence Analysis – Scilab : Open Source Equivalent of Matlab – VRGeo : Collaborative mapping platform for Geospatial information

• Supports OpenId for user login
• Present focus includes

– Providing Scientific tools in the area of CFD, Bioinformatics, Molecular Biology, Statistical Computing etc – VO enabled On demand Grid

19 Oct 2013 Trends Cloud Comp- NKN WS 2013 49

eSanjeevani 3.1: Telemedicine App. on Cloud

eSanjeevani is a web based Telemedicine Application By CDAC Mohali.

Hosted on: www. esanjeevani.in.

The key features include:

• MIS based application
• Doctor to Doctor consultation
• Patient to Doctor consultation
• Comprehensive EMR
• Reports and Analytical tools
• Appointment Scheduler
• Medical Equipment Interfacing

Trends Cloud Comp- NKN WS 2013

19 Oct 2013 50

Megh Shushrut: e-Shushrut On Cloud

• e-Sushrut: C-DAC’s Hospital Management Information System. e-

Shushrut hosted on cloud is named as Megh-Shushrut.

• The Megh Shushrut cater the needs of the state for State of Rajasthan,

Government of National Capital Territory Delhi and North Eastern States, in the very near future

• Megh-Shushrut (MS) will be launched using Software as a Service (SaaS)

model where different modules like Patient Registration, Out Patient Management, In Patient Management, Patient Billing, Laboratory Information System and EMR are deployed as service

• Megh-Shushrut will be deployed at Centralized Data Center Provided by

the State

• MS has been configured such that, each hospital can have access to their
• data. In that way patient EHR is being maintained at the centralized data
• center. The implementation in these cases is supposed to state wide.
• The realization is will be very costly to build separate hardware and

software infrastructure in all the hospitals across the state

• Cloud based solutions offer a Cost Effective solution

19 Oct 2013

Trends Cloud Comp- NKN WS 2013

51

MEGHA: Cloud interface for GARUDA

• Provides virtual Infrastructure on Garuda resource

– Supports IaaS and SaaS – Supports various OS flavors RHEL, Ubuntu and Fedora.

• Provides Virtual Clusters Inbuilt with

– clustering tools – Parallel paradigm like Mpi & OpenMP. – Torque as scheduler

• Provides on Demand Scientific Applications:

– SFM : Seasonal Forecasting Model, – FastA: Protein Sequence Analysis – Scilab : Open Source Equivalent of Matlab – VRGeo : Collaborative mapping platform for Geospatial information

• Supports OpenId for user login
• Present focus includes

– Providing Scientific tools in the area of CFD, Bioinformatics, Molecular Biology, Statistical Computing etc – VO enabled On demand Grid

19 Oct 2013 52 Trends Cloud Comp- NKN WS 2013

Governments from around the world are starting to take notice of the flexibility,

• perational benefits and substantial cost

savings that cloud computing can provide.

Trends Cloud Comp- ACC 2013

Governament Clouds

Indian Government: Heading To Cloud

• DietY declared E-governance on national cloud with all data centers

connected.

• In National Network Cloud all states will have their private cloud
• National Cloud Benefits

– With Reduced IT management load agencies can concentrate on core activities – Elasticity – Agility

19 Sept 2013 Trends Cloud Comp- ACC 2013 54

• Govt. Applications which can benefit on CLOUD

http://www.cisco.com/web/strategy/docs/c11-687784_cloud_omputing_wp.pdf

• Pension
• Taxes
• Education Portal with e-learning virtual class room
• Health domain – connect remote medical institutes, real time patients data

analysis and treatment, medicine –at-a-distance

• Databases on Cloud- Can Host the Citizens details Will ease the process of sharing

data between different public sector organisations e.g. pension, elections , Driver’s License

• etc .

http://www.computerweekly.com/blogs/public-sector/2011/02/g-cloud-introducing-the-neo-

da.html

19 Sept 2013 Trends Cloud Comp- ACC 2013 55

Government Cloud Services

Government Cloud Computing Services Can Be Categorized as:

 Strategic Services  Which form only 10%  Financial Initiatives (Can be planned Over Private Cloud

Networks)

 Public Services  Income Tax Returns Filing  National Level Examination Conduction  Post Office Services  Health Care Services  IT services- S/W testing Utilities (Sporadic Nature of Usage)

UIDAI INITIATIVE: Aadhaar is a 12-digit

unique number which the Unique Identification Authority of India (UIDAI) will issue for all residents. The number will be stored in a centralized database and linked to the basic demographics and biometric information – photograph, ten fingerprints and iris – of each individual. NKN: National Knowledge Network: NKN will interconnect all the research, higher education and scientific institutions in the country, over a period

• f three years.

Cloud Computing and NKN: We are learning creation, operation and negotiations of Virtual organizations (VO) in NKN

• P.S.Dhekne,Raja Ramanna Fellow, BARC
• Ref. No 4

Trends Cloud Comp- ACC 2013

Cloud Trends

19 Oct 2013 57 Trends Cloud Comp- NKN WS 2013

• In internet Computers are attached
• Internet of things, things around us

like car, refrigerator etc will be attached.

• IoT Consists any device with the

ability, to gather and process information, and communicate it across the network.

58 19 Oct 2013 Trends Cloud Comp- NKN WS 2013

How Does This Work?

• Different types of Devices

comprise a network (LAN, ZigBee as shown in Left)

• Applications may have

different interfaces: PC, cellphone, tablets etc

• Devise networks

communicate with the Applications via Internet

59 19 Oct 2013 Trends Cloud Comp- NKN WS 2013

Problem With this Approach?

 To connect to Internet and Communicate Information to Applications,

we need back-haul network (Circle in figure).

 The many types and individual components of infrastructure can create

massive complexities for solution providers. Challenges:

 Scalability, less usability of resources, Security, Interoperability

Solution ?

 An Infrastructure that can address above challenges ?

19 Oct 2013 60 Trends Cloud Comp- NKN WS 2013

On Demand Cloud

1. IaaS (Infrastructure as a Service)

• 2. User worry only about the Application &

not about complex infrastructure On Demand Cloud for Internet of Things

61 19 Oct 2013 Trends Cloud Comp- NKN WS 2013

Internet of Things on Cloud

 The individual devices will gather the information and send

it to the cloud, the application will read the information from the cloud and show it in different interfaces.

 The features of Internet of things on Cloud are:

• 1. Scalability,
• 2. Flexibility
• 3. Better Resource Utilization,etc

62 19 Oct 2013 Trends Cloud Comp- NKN WS 2013

Applications of Internet Of Things

Smart Parking:

 Monitoring of parking spaces availability in the city

Traffic Congestion:

 Monitoring of vehicles and pedestrian levels to optimize

driving and walking routes. Radiation Levels

 Distributed measurement of radiation levels in nuclear

power stations surroundings to generate leakage alerts.

Fall Detection(e-Health)

 Assistance for elderly or disabled people living

independent.

63 19 Oct 2013 Trends Cloud Comp- NKN WS 2013

Sensor Clouds

19 Oct 2013 64 Trends Cloud Comp- NKN WS 2013

Sensor Clouds

• Sensor Clouds: A new paradigmfor cloud computing, using sensors to collect

information of physical things and putting all sensor data into cloud computing platforms

• Efficiently handle sensor data for monitoring and mining in clouds

– Efficient storage and retrievals – Partition schemes for MapReduce – Deal with real-time sensor data

65

Sensor networks http://research.microsoft.com/en-us/events/cloudfutures2010/peng.pdf

19 Oct 2013 Trends Cloud Comp- NKN WS 2013

Sensor Cloud Infrastructure

http://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=5635688&tag=1

• Sensor-Cloud infrastructure (SCI):

manages physical sensors on IT infrastructure.

• Sensor-Cloud Infrastructure

virtualizes a physical sensor as a virtual sensor on the cloud computing.

• Dynamic grouped virtual sensors on

cloud computing can be automatic provisioned when the users need them.

• The approach enables the sensor

management capability on cloud computing.

• Since the resource and capability of

physical sensor devices is limited,

• the cloud computing on the IT

infrastructure can be behalf of the sensor management such as availability and performance of physical sensors. Fig.Sensor Cloud Infrastructure http://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=5635688&tag=1

19 Oct 2013 66 Trends Cloud Comp- NKN WS 2013

Sensor Clouds

• The number of sensors is increasing for various purposes.
• Sensors sensing the phenomenon are used by a lot of services such as

environment, healthcare and government services.

• Because each sensor is closely linked to each service directly, other services

cannot use sensors easily.

• Sharing sensors by other unlimited services will accelerate service innovation.
• SCI is the extended cloud computing to manage sensors.
• SCI provides sensor devices as a part of IT resources (CPU, memory and disk)

for end users.

• SCI enables to provision service instances automatically, to monitor sensors

and to control sensors.

• These functions can be used via the user interface via web browser.
• IoT envision that sensors to be attached everywhere. In such environment
• wner will be able to generate data.
• Sensing as Service Model Evolving: can be done at personal , private and

public organization level.

• Owner of sensors will be able to generate data and get a return on

investment.

67 19 Oct 2013 Trends Cloud Comp- NKN WS 2013

Sensing As A Service Model

68

• Service instances can be

created automatically when they are requested.

• Automation is an important

characteristic of CC service CCS.

• The service delivery time and

the service cost are improved by automation because of no human operations.

• Standardization is important

characteristic of CC service.

• Service providers provide

services with different configurations by each service requester’s requirements before CCS

19 Oct 2013 Trends Cloud Comp- NKN WS 2013

Sensing As A Service Model

69

• Service instances can be

created automatically when they are requested.

• Automation is an important

characteristic of CC service CCS.

• The service delivery time and

the service cost are improved by automation because of no human operations.

• Standardization is important

characteristic of CC service.

• Service providers provide

services with different configurations by each service requester’s requirements before CCS

19 Oct 2013 Trends Cloud Comp- NKN WS 2013

Fujitsu Laboratories: Sensor Cloud

70

• Its R&D Provide users with

information, in most appropriate /desirable forms in accordance with time/place/occasion, based

• n:
• analysis of cloud-

accumulated sensing data

• accumulated via networks,
• by sensing individuals'

activities

• and user environments
• through sensors in mobile

devices and placed at various sites

• Human-friendly services are in demand for fields such as healthcare, transport,

agriculture, and electrical utilities, in which ICT systems had not been fully leveraged.

• Fujitsu Laboratories employs sensing technologies for R&D to enable

comprehension and awareness of activities, behavior, and intent of individuals active in the real world.

M2M Sensors into Cloud

M2M Smart Services Developer Kit fully supports cloud-based Cumulocity M2M Application Platform, which delivers server-side database platform, network agents, and sensor libraries to bring M2M connected sensors and devices into cloud. Server-side database platform collects and analyzes data from Developer Kit connected sensors and devices and controls devices in secure and scalable manner so that service kit is tailored for vertical M2M industries such as fleet management and smart energy grids.

19 Oct 2013 Trends Cloud Comp- NKN WS 2013

Other Sensor Services in Cloud

71

IBM Management System Infrastructure as a Service for IT & Sensors

• The purpose is to enable the enhanced CC system that manages sensor devices

as a part of IT resource in the same way as CPU, storages, and networks to provide the sensor devices as an infrastructure for users.

• This Infrastructure software includes three key features as cloud computing;

1. Virtualization: Multiple users can share the sensor devices. 2. Automation: The system configures the sensor devices and software instead of the users. 3. Standardization: VM template has not only software and also the sensor

• devices. Users can utilize the sensor devices easily and rapidly in this

cloud computing environment as well as IT resources M2M Sensors into cloud: M2M Smart Services Developer Kit fully supports

• Cloud-based Cumulocity M2M Application Platform.
• Delivers server-side database platform, network agents, and sensor libraries to

bring M2M connected sensors and devices into cloud.

• Server-side database platform collects and analyzes data from Developer Kit

connected sensors and devices and controls devices in secure and scalable manner so that service kit is tailored for vertical M2M industries such as fleet management and smart energy grids.

19 Oct 2013 Trends Cloud Comp- NKN WS 2013

Sensing As A Service & BigData

• Modern world is full of devices comprising sensors, data

processor.

• Such resources enable sensing, capturing, collection, and

processing of real time data from billions of connected devices. Fact:

• 2010-Total data on earth exceeded one ZB (zettabyte).
• End of 2011- Total data number grew up to 1.8 ZB.
• 2020- Expected that the number will reach 35 ZB in 2020

19 Oct 2013 Trends Cloud Comp- NKN WS 2013 72

Sky Computing: Distributed Cloud Computing

HTC Scientific Computng in a Distributed Cloud Environment General supporting wide range of applications

19 Oct 2013 73 Trends Cloud Comp- NKN WS 2013

Components of Big Science

19 Oct 2013 74 Trends Cloud Comp- NKN WS 2013

• Success
• Future Systems

Big Science- Evolving

19 Oct 2013 75 Trends Cloud Comp- NKN WS 2013

Conclusion

• Next Generation Complex Systems are becoming a Reality

– Peta/ExaScale Computing Systems – Combines Big Compute & Big Data to deliver

– Exascale Performance in Supercomputers

• National Missions are Emerging

– National Grid Computing:GARUDA – National Knowledge NetWork:NKN – National Supercompution Mission – National BioDiversit Organization – National Cloud Innovation Council – National BigData Mission ?

• HPC Grid and Cloud infrastructures are converging:

– Compute Resources on demand to Grids – Cloud Bursting Research: Grids Vs Clouds – On Demand Cloud Services for IoT , Cloud Services for BigData

• Newer Services Models “Sensing as a Service” are Evolving
• New Science “Big Science” Evolving
• C-DAC is Engaged in: HPC, Grid, Cloud, BigData, and IoT

19 Oct 2013 Trends Cloud Comp- NKN WS 2013 76

C-DAC Dissemination Activities

29 April 2013 NCIC CDAC Cloud Initiatives 77

Think Parallel Parallel Programming for Engineers and Scientists

• Two Week training for Engineering College faculty, Scientific Community

from R&D Labs.

• 3 Times Conducted, 180+ Teachers & Scientists are Trained on Parallel

Programming Theory & Hands-on:

• Parallel Programming Models
• Parlllel Computer Architectures
• Hybrid Systems
• Cluster, Grid and Cloud Computing
• Recently conducted - June10- 20th 2013

PARCOMPTECH

• Two-days Conference on Parallel Computing Technologies
• Objective – to encourage National level research
• Four tutorials

– Think Parallel, Big Data, TI Multicore, Cloud Computing

78

Think Parallel

Think Parallel: Parallel Programming for Engineers and Scientists (10June 2013 -20 June2013)

• Advanced Computer Architectures
• Introduction to Parallel Programming
• Principles of Parallel Algorithm Design
• Building Compute Clusters
• Numerical Computing
• Programming Paradigms
• OpenMP
• Message Passing Interface (MPI)
• Advanced MPI
• Performance Analysis and
• Debugging Parallel Programs
• Parallel Applications/Case Studies
• GPGPU & CUDA programming
• Grid Computing
• Cloud Computing

29 April 2013 NCIC CDAC Cloud Initiatives

29 April 2013 NCIC CDAC Cloud Initiatives 79

C-DAC, a Scientific Society of Ministry of Communications and Information Technology, an organization known for its expertise in High Performance Computing, is organizing the 2nd National Conference on “Parallel Computing Technologies (PARCOMPTECH-2014) during 27th & 28th Feb 2014 at Pune, India. The conference aims to provide a platform for Indian scientific and engineering computing fraternities from academics, scientific labs, R&D institutions, industry and government to share their research work, knowledge and experience in the field of parallel computing technologies. This conference will act as a ground to create a collaborative environment, ultimately promoting scientific and technological development. Authors are invited to submit original, unpublished research manuscripts demonstrating their current research in parallel computing technological area. The conference topics of interest include but not limited to:

• Parallel and Distributed Architectures
• Multi-core computing
• Giant scale computing
• Interconnection networks
• Parallel I/O and Storages Systems, Databases
• GPU / Accelerators/ Heterogeneous computing
• Parallel software – Algorithms, Compilers, OS, Programming languages and libraries
• Grid and Cloud computing
• Middleware for Cloud and Grid
• Power optimization and Green computing
• Self aware and Self healing systems
• Fault-Tolerant, Scheduling and Load balancing Algorithms
• Performance Analysis, Benchmarking.
• Peta and Exascale Computing
• Scientific/Engineering/Commercial Applications and Workflows

For accepted papers, at least one of the authors need to register and present their work at the conference. Selected papers will be published through IEEE Xplore and PARCOMPTECH Conference Proceedings. Further instructions for PARCOMPTECH2014 paper submission is available at the following web link :http://parcomptech.cdac.in/

Conference Dates: 27th & 28th February 2104 Venue: YASHADA, Baner Road, Pune, India.

• GARUDA Challenge
• GARUDA NKN Meet

– 75 R&D labs and academic institutions – 1000+ users – Application Drivers

• OSDD
• Bioinformatics
• CFD
• Weather
• Material Science
• GARUDA Boot Camps
• CDAC-Industry Collaboration Workshops

– Texas Instruments(Multicore DSP), Intel (Smart Grid), ARM, NVIDIA, etc.

GARUDA Dissemination Works

29 April 2013 NCIC CDAC Cloud Initiatives 80

Dissemination/ Conference/Workshops/Tutorials

• Prahlada Rao B.B., “HPC for BigData” International Workshop on Data Analytics and Applications (IWDAA) 2013,

BITS, Goa Campus, 1st Mar2013.

• Prahlada Rao B.B., Payal Saluja, Vineeth Simon Arackle, Tutorial “Cloud Computing for HPC” C-DAC’s National

Conference on Parallel Computing Technologies (ParCompTech2013), NIAS, Bangalore, 21-23rd Feb 2013.

• Prahlada Rao B.B, Payal Saluja, Neetu Sharma, Ankit Mittal, Shivay Veer Sharma, “Cloud Computing for IOT & Sensing

Based Applicationss”, in IEEE International Conference on Sensing Technologies, ICST 2012, Kollkata, India, Dec 2012.

• Prahlada Rao B.B., Payal Saluja, Vineeth Simon Arackle, Half-Day Tutorial “Open Source Platforms for Scientific

Clouds” CSI’s Annual Conference on Advances in Cloud Computing (ACC2012), 26th July 2012, Bangalore.(60+ participants from Industry, Academia)

• Payal Saluja, Prahlada Rao, Ankit Mittal, Rameez Ahmad, “CDAC Scientific Cloud: On Demand Provisioning of

Resources for Scientific Applications”, in Proceedings of 18th International Conference PDPTA 2012, Las vegas, Nevada, USA, 16-19 July 2012.

• Prahlada Rao B.B., Tutorial On “Cloud Computing”, International Conference on Trends in Knowledge and

Information Dynamics, ICTK 2012, organized by ISI, NIMHANS Conv Centre, Bangalore, India, 10 July 2012.(150 participants from Industry, Academia)

• Dr. Prahlada Rao BB, Cloud Computing for Telemedicine, National Workshop on Telemedicine Today & Tomorrow, C-

DAC Mohali, An Invited Talk, 6th July 2012.

• Prahlada Rao B. B. , National Workshop on Cloud Computing, conducted at CDAC Knowledge Park, Bangalore, 21st

May 2011 .(80+ participants from Industry, Academia)

• Vineeth Simon Arackal, Arunachalam B, Payal Saluja, Prahlada Rao B B, Co-existence of Cloud and Grid: A Case Study

in Service Oriented Grid GARUDA. The International Conference on Information Technology- New Generations (ITNG 2011), Las Vegas, USA. 11-13April 2011.

• Vineeth Simon Arackel, “ Coexistence of Grid and Cloud” at "Grid To Cloud (G2C)- Challenges and Applications"

Workshop at VIT, Vellore, 2nd April 2011. 29 April 2013 NCIC CDAC Cloud Initiatives

81

19 Oct 2013 Trends Cloud Comp- NKN WS 2013

Thank You

82

• Dr. B.B Prahlada Rao

C-DAC Bangalore

email: prahladab@cdac.in