4. INTRODUCTION (CONTINUE)4
TeraGrid is an open scientific discovery infrastructure combining leadership class
resources at eleven partner sites to create an integrated, persistent computational
resource
5. DEFINITION OF CLOUD COMPUTING
There is little consensus on how to define the Cloud
A large-scale distributed computing paradigm that is driven by economies
of scale, in which a pool of abstracted, virtualized, dynamically-scalable,
managed computing power, storage, platforms, and services are delivered
on demand to external customers over the Internet.
Ian Foster
Cloud computing is using the internet to access someone else's software
running on someone else's hardware in someone else's data center.
Lewis Cunningham
5
6. DEFINITION OF CLOUD COMPUTING
There is little consensus on how to define the Cloud
A large-scale distributed computing paradigm that is driven by economies
of scale, in which a pool of abstracted, virtualized, dynamically-scalable,
managed computing power, storage, platforms, and services are delivered
on demand to external customers over the Internet.
Ian Foster
Cloud computing is using the internet to access someone else's software
running on someone else's hardware in someone else's data center.
Lewis Cunningham
6
7. Clouds: key points of the definition
Differences related to traditional distributed paradigms:
Massively scalable
Can be encapsulated as an abstract entity that delivers different levels
of service
Driven by economies of scale
Services can be dynamically configured (via virtualization or other
approaches) and delivered on demand
7
8. Clouds: reasons for interest
Rapid decrease in hw cost, increase in computing power and
storage capacity (multi-cores etc)
Exponentially growing data size
Widespread adoption of Services Computing and Web 2.0
apps
8
9. Clouds: relation with other paradigms9
A Web 2.0 site may allow users to interact and collaborate with each other in a social media dialogue as
creators of user-generated content in a virtual community, in contrast to Web sites where people are
limited to the passive viewing of content. Examples of Web 2.0 include social networking sites, blogs,
wikis, folksonomies, video sharing sites, hosted services, Web applications
10. GRID COMPUTING
Grid Computing enables resource sharing and coordinated
problem solving in virtual organizations(VO) where each VO
can consist of either physically distributed institutions or
logically related projects/groups.
Builds a uniform computing environment from diverse
resources by defining standard network protocols and
providing middleware to mediate access to a wide range of
heterogeneous resources (egGlobusToolkit).
10
12. How technologists perceive the Cloud
“The interesting thing about Cloud Computing is that we’ve redefined Cloud
Computing to include everything that we already do. . . . I don’t understand what
we would do differently in the light of Cloud Computing other than change the
wording of some of our ads.”
12
Larry Ellison (Oracle CEO)
Wall Street Journal, September 26, 2008
13. How technologists perceive the Cloud
“A lot of people are jumping on the [cloud] bandwagon, but I have not heard two
people say the same thing about it. There are multiple definitions out there of “the
cloud.”
13
Andy Isherwood (HP VP of sales)
ZDnetNews, December 11, 2008
14. How technologists perceive the Cloud
“It’s stupidity. It’s worse than stupidity: it’s a marketing hype campaign. Somebody
is saying this is inevitable —and whenever you hear somebody saying that, it’s very
likely to be a set of businesses campaigning to make it true.”
14
Richard Stallman (Advocator of Free Software)
The Guardian, September 29, 2008
15. Is Cloud a new name for Grids?
YES: the vision is the same
reduce the cost of computing
increase reliability
increase flexibility (transitioning from self-operation to third
party)
15
The answer is complicated…
IT reinvents itself every five years
16. Is Cloud a new name for Grids?
NO: things are different than 10 years ago
New needs to analyze massive data, increased demand for
computing
Billions of dollars being spent by Amazon, Google,Microsoft to
create real commercial large-scale systems with hundreds of
thousands of computers – www.top500.org shows computers
with 100,000+ cores
Only need a credit card to get on-demand access to infinite
computers
16
17. Is Cloud a new name for Grids?
Nevertheless YES: same problems but different details
Problems are the same in clouds and grids
How to manage large facilities
How to discover, request, and use resources
How to implement and execute parallel Computations
17
18. Clouds: side-by-side comparison with
Grids
Business model
Architecture
Resource Management
Programming model
Application model
Security model
18
19. Cloud vs Grids - Business model
Traditional: one-time payment for unlimited use of software
Clouds: pay the provider on a consumption basis, computing
and storage (like electricity, gas etc)
Grids: assigned a number of service units
19
20. Cloud vs Grids - Architecture20
communication
and authentication
protocols
discovery, negotiation,
monitoring, accounting and
payment of sharing operations
on individual resources
interactions across
collections of
resources,
directory services
21. Cloud vs Grids - Architecture21
resources that
have been
abstracted/encapsulated
collection of specialized
tools, middleware and
services on top of the unified
resources to provide a
development and/or
deployment platform
22. Cloud vs Grids - Architecture
SPI Model
Cloud Software as a Service (SaaS)
Cloud Platform as a Service (PaaS)
Cloud Infrastructure as a Service (IaaS)
22
23. Cloud vs Grids - Architecture
SPI MODEL (IaaS)
The capability provided to the consumer is to provision
processing, storage, networks, and other fundamental
computing resources.
Consumer is able to deploy and run arbitrary software, which
can include operating systems and applications.
The consumer does not manage or control the underlying
cloud infrastructure but has control over operating systems,
storage, deployed applications, and possibly limited control of
select networking components (e.g., host firewalls).
Typical examples are Amazon EC2 Service and S3
23
25. Cloud vs Grids - Architecture
SPI MODEL (PaaS)
The capability provided to the consumer is to deploy onto the
cloud infrastructure consumer created or acquired
applications created using programming languages and tools
supported by the provider.
The consumer does not manage or control the underlying
cloud infrastructure including network, servers, operating
systems, or storage, but has control over the deployed
applications and possibly application hosting environment
configurations.
25
26. Cloud vs Grids - Architecture
SPI MODEL (SaaS)
The capability provided to the consumer is to use the
provider’s applications running on a cloud infrastructure.
delivers special purpose software that is remotely accessible.
E.g,: Google Maps, Live Mesh from Microsoft etc
The consumer does not manage or control the underlying
cloud infrastructure including network, servers, operating
systems, storage, or even individual application capabilities,
with the possible exception of limited userspecific application
configuration settings.
26
27. Cloud vs Grids – Resource Management
Compute model
Data model
Virtualization
Monitoring
27
28. Cloud vs Grids – Resource Management
COMPUTE MODEL
Grids: batch-scheduled (queueing systems)
Clouds: resources shared by all users at the same time in
contrast to dedicated resources in queueing systems
Maybe one of the major challenges in clouds: QoS!
28
29. Cloud vs Grids – Resource Management
DATA MODEL
Data locality: to achieve good scalability data must be
distributed over many computers
Clouds: use map-reduce mechanism like in Google to
maintain data locality
In Grids, data storage usually relies on a shared file systems
(e.g. NFS, GPFS, PVFS, Luster), where data locality cannot be
easily applied
29
30. Cloud vs Grids – Resource Management
Virtualization
Abstraction and encapsulation
Clouds: rely heavily on virtualization
Supports cost-effective use of cloud’s physical resources
Grids: do not rely on virtualization as much as clouds.
due to policy and having each individual organization
maintain full control of their resources
However, there are efforts in Grids to use virtualization as well,
such as Nimbus
30
31. Cloud vs Grids – Resource Management
Monitoring
Virtualization poses challenges to fine-grained control over
monitoring
Service-oriented view means resources below service API are
not visible
Monitoring may not be as important because of abstractions
Grid trust model allows users via their identity delegation to
access and browse resources at different sites
Resources not highly abstracted & virtualized
31
32. Cloud vs Grids – Programming Model
CLOUD
most use the map-reduce programming model.
Implementation: Hadoop that uses Pig as a declarative
programming language
GRID
Complicated by issues like multiple administrative domains,
resource heterogeneity, etc
heavy use of workflow tools to be able to manage large sets
of tasks and data. others: MPICH-G2, WSRF, GridRPC…
32
33. Cloud vs Grids – Application Model
CLOUD
Traditionally can support same apps as grid except HPC (due
to low latency needs) but this is changing
Interactive, loosely-coupled, transaction-oriented apps
GRID
Batch-oriented apps
Support High-Performance Computing (HPC) through High
Throughput Computing (HTC)
Support workflows of loosely-coupled applications
33
34. Cloud vs Grids – Security Model
Clouds currently more homogeneous and single provider so
security simpler
Still an important concern for cloud users
Email address & credit card gets you an account
Grids Built on assumptions of heterogeneous and dynamic resources
and multiple admin domains
Stricter procedure to acquire an account
34
Xsede کلکسیونی از منابع وسرویسهای دیجیتال تجمیع است و یک سیستم مجازی واحد را برای استفاده برای اشتراک منابع محاسباتی و داده ها ارائه میدهد
Cloud یک پارادایم محاسباتی در مقیاس بزرگ است
که شامل استخری از منابع محاسباتی و ذخیره سازی – سرویس ها – و پلتفرم هایی است که virt و abs و با مقیاس متغیرند برای استفاده بر حسب تقاضای کاربر روی اینترنت
دولتمردان – موسسات تحقیقاتی و مدیران صنعتی به شدت دنبال به کارگیری cloud اند تا مشکلات محاسباتی و ذخیره سازی خود را حل کنند.
نه تنها cloud با گرید overlap دارد بلکه گرید backbone آن هم هست
شیفت از چیزی که زیرساخت منابع ذخیره سازی و محاسباتی ، به جایی که بر پایه اهداف اقتصادی بنا شده و سرویس و منابع انتزاعی بیشتری فراهم میکرد.
در اواسط دهه 90 مفهوم گرید ابداع شد
فاستر و دیگران ادعا کردند که با استانداردسازی پروتکل های مورد استفاده در درخواست توان محاسباتی جرقه computing grid را میتوان زد.
اما هیچ grid computing provider تجاری ای تا امروز به صورت ماندنی ظهور نکرد.
یک محیط محاسباتی واحدی را ارائه میدهد که شما با استفاده از middleware و پروتکل استاندارد شبکه میتوانید به منابع ناهمگون زیادی دسترسی داشته باشید
chief executive officer (CEO)
He was the Chief executive officer of the software company Oracle Corporation between its foundation in 1977 and 2014. In 2014, he was listed by Forbes as the third-wealthiest man in America and as the fifth-wealthiest person in the world, with a fortune of $56.2 billion
vice president and general manager of HP's worldwide Software نایب رییس
در فناوری اطلاعات که به شدت تکنولوژی بزرگ می شود و تقریبا هر 5 سال خودش را بازسازی می کند هیچ جواب رک و راستی برای این سوال وجود ندارد.
قابلیت اطمینان – انتقال computer ها از چیزی که ما خودمان می خریم و روی
با درک مزایای مهاجرت از mainframe ها به کلاسترها ولی این کلاسترها نیز بسیار پر هزینه هستند برای انجام عملیات و ما در مقابل virtualization کم هزینه را داریم
Cloud مثل utility computing است
آمازون یک cloud متمرکز حاوی محاسباتی به نام amazon ec2 و amazon s3
در قبلی شارژ بر حسب ساعت/نوع بود. در این مدل گیگ بر ماه است.داده انتقالی برحسب tb/ماه است. یک credit card
موسسات در مثلن تراگرید وقتی عضو میشدند میپذیرفتند که منابعشان توسط موسسات دیگر به اشتراک گذاشته میشود همچنین انگیزه شان دسترسی به منابع دیگران بود
دسترسی به منابع محاسباتی high perf گران و سخت بود لذا هدفش متمرکز کردن منابع محاسباتی و ... از چندین موسسه با منابع گوناگون بود
برای اینکار نیاز به پشتیبانی از مفهوم virtual Org بود
برای این مفهوم گرید یک سری پروتکل های استاندارد را تعریف کرد
Cloud روی اینترنت برای حل مشکلات محاسباتی
استخری از منابع مختلف که با یک واسط انتزاعی و بوسیله پروتکل های استاندارد مورد دسترسی قرار گیرند
Uni معمولا بوسیله مجازی سازی بنابراین این منابع برای لایه های بالاتر میتوانند به عنوان منابع تجمیع ظاهر شوند
Platform یک سریس زمانبند
در بیشتر مواقع در لایه unifi است ولی ممکن است fabric هم شامل شود
توسعه دهندگان یک سری محدودیت ها را باید برای برنامه های خود بپذیرن( برای مقیاس دهی برنامه های قبلی )
مثال google app engine است که کاربران را قادر میسازد تا برنامه های وبی در همان سیستم مقیاس پذیر که برنامه های گوگل را توانمند میسازد بسازند.
این نوع سرویس مربوط به برنامه های خاص منظوره برای استفاده از راه دور توسط کاربر میباشند
استفاده از چند cloud
استاندارد خاصی برای واسط های این سه تعریف نشده
مشکلات تعمیم cloud ها
گرید از مدل محاسباتی زمانبندی دسته ای استفاده میکند. یک مدیر منبع محلی LRM منابع محاسباتی را مدیریت میکند برای یک سایت گرید. کابران تقاضای منابع برای زمان مشخص را ثبت میکنند
پس نیاز به زمانبندهای data aware دارید برای زمانبندی مناسب task های محاسباتی
جز حتمی همه cloud هاست مثل thread ها که تصور غلط را ایجاد میکردند
Abstra بوسیله اینکه منابع زیرین (شبکه و ) به صورت استخری از منابع به صورت واحد جلوه کنند
گرید مبتنی بر زمانبندی داده هاست
در cloud هیچ library هم وجود ندارد برای نظارت منابع زیرین
مدل برنامه نویسی در گرید خیلی با محیط های توزیع شده فرق ندارد اما ...
تغییر زیاد در منابع ( امدن و رفتن )
زیرساخت ابر بر پایه web form هاست روی ssl تا اطلاعات اکانت کاربر را ساخته و آن را مدیریت کند کاربر را قادر میسازد تا رمز خود را ریست کرده و با ایمیل رمز جدید را تخت ارتباطی غیر امن و غیر کدگذاری شده دریافت کند.