Что такое Grid?

План выступления
Ранние определения GRID
Понятие о Виртуальной Организации
Область зарождения и применения GRID
Перспективы для развития
Минимальная архитектура для участия в GRID
Проект EGEE и участие в нем России

„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.“ Leonard Kleinrock (1969)
Мы, вероятно, увидим распространение «компьютерных услуг», которые, подобно электричеству и телефону, придут в дома и офисы по всей стране.
„A computational grid is a hardware and software infrastructure that
provides dependable, consistent, pervasive, and inexpensive access to
high-end computational capabilities.“ I. Foster, C. Kesselmann (1998), “The Grid: Blueprint for a New Computing Infrastructures”.
Вычислительный Grid - это аппаратно-программная инфраструктура, которая обеспечивает надежный, устойчивый, повсеместный и недорогой доступ к высокопроизводительным вычислительным ресурсам.
Первые определения GRID

Идеалистическое представление о GRID
G
R
I
D
M
I
D
D
L
E
W
A
R
E
Visualizing
Supercomputer, PC-Cluster
Data Storage, Sensors, Experiments
Internet, Networks
Desktop
Mobile Access

Скоординированное разделение ресурсов -
Виртуальная Организация
„The sharing that we are concerned with is not primarily file exchange but rather direct access to computers, software, data, and other resources, as is required by a range of collaborative problem-solving and resource brokering strategies emerging in industry, science, and engineering. The sharing is, necessarily, highly controlled, with resources providers and consumers defining clearly and carefully just what is shared, who is allowed to share, and the conditions under which sharing occurs. A set of individuals and/or institutions defined by such sharing rules what we call a virtual organization.“ I. Foster, C. Kesselmann, S. Tuecke (2000), “The Anatomy of the Grid”.
Разделение (ресурсов), о котором мы говорим, - это не просто обмен файлами, а еще и прямой доступ к компьютерам, программному обеспечению, данным и другим ресурсам, как это требуется для совместного решения задач и стратегий выделения ресурсов. принятых в промышленности, науке и инженерии. Разделение должно тщательно контролироваться поставщиками и потребителями ресурсов, быть ясно и тщательно определено, что распределяется, кому разрешается пользоваться и на каких условиях. Отдельные пользователи и/или институты, подчиняющиеся таким правилам, образуют то, что мы называем Виртуальной Организацией.

The LHC Computing Model
Tier2 Centre ~1 TIPS
Online System
Offline Farm
~20 TIPS
CERN Computer Centre >20 TIPS
RAL Regional Centre
US Regional Centre
French Regional Centre
Italian Regional Centre
Institute
Institute
Institute
Institute ~0.25TIPS
Workstations
~100 MBytes/sec
~100 MBytes/sec
100 - 1000 Mbits/sec
HPSS
HPSS
One bunch crossing per 25 ns
100 triggers per second
Each event is ~1 Mbyte
Physicists work on analysis “channels”
Each institute has ~10 physicists working on one or more channels
Data for these channels should be cached by the institute server
Physics data cache
~PBytes/sec
~ Gbits/sec or Air Freight
HPSS
HPSS
HPSS
Tier2 Centre ~1 TIPS
Tier2 Centre ~1 TIPS
~Gbits/sec
Tier 0
Tier 1
Tier 3
1 TIPS = 25,000 SpecInt95
1 PC (2000) = ~15 SpecInt95
Tier 2
The LHC Computing Model
The LHC Computing Model
The LHC Computing Model
The LHC Computing Model
The LHC Computing Model
The LHC Computing Model
Рисунок Harvey Newman, Caltech

The LHC Computing Model
Tier2 SINP ~1 TIPS
Online System
Offline Farm
~20 TIPS
CERN Computer Centre >20 TIPS
CERN Regional Centre
LAB
LAB
LAB
Physics LAB ~0.25TIPS
Workstations
~100 MBytes/sec
~100 MBytes/sec
100 - 1000 Mbits/sec
HPSS
HPSS
One bunch crossing per 25 ns
100 triggers per second
Each event is ~1 Mbyte
Physicists work on analysis “channels”
Each institute has ~10 physicists working on one or more channels
Data for these channels should be cached by the institute server
Physics data cache
~PBytes/sec
Tier2 ITEP ~1 TIPS
RDIG JINR ~1 TIPS
~Gbits/sec
Tier 0
Tier 3
1 TIPS = 25,000 SpecInt95
1 PC (2000) = ~15 SpecInt95
1 PC (2007) = 250 SpecInt95
Tier 2
The LHC Computing Model
The LHC Computing Model
The LHC Computing Model
The LHC Computing Model
The LHC Computing Model (RDIG)
The LHC Computing Model (RDIG)
Tier 1

Gilder’s Law
(32X in 4 yrs)
Storage Law
(16X in 4yrs)
Moore’s Law
(5X in 4yrs)
Triumph of Light – Scientific American. George Stix, January 2001
Performance per Dollar Spent
Optical Fibre
(bits per second)
Chip capacity
(# transistors)
Data Storage
(bits per sq. inch)
Number of Years
0 1 2 3 4 5
Doubling Time
(months)

Перспективы развития ресурсов для GRID
С 1986 по 2000 годы производительность вычислений при тех же затратах выросла в 500 раз. Ожидается , что до 2010 года производительность вырастет в 30 000,
Сейчас в мире около 500 млн. РС. Из них не более 30% имеют выход в Интернет
Средний коэффициент использования вычислительного ресурса не превышает 5%!

Network
Ресурсы необходимые для участия в виртуальной организации

Упрощенное взаимодействие сервисов/ Серверов )
Web
Browser
Web
Server
Web Сервис
Минимальная конфигурация
для участия в GRID
Initial screen
Proxy information

The LHC Computing Model
Tier2 SINP ~1 TIPS
Online System
Offline Farm
~20 TIPS
CERN Computer Centre >20 TIPS
CERN Regional Centre
LAB
LAB
LAB
Physics LAB ~0.25TIPS
Workstations
~100 MBytes/sec
~100 MBytes/sec
100 - 1000 Mbits/sec
HPSS
HPSS
One bunch crossing per 25 ns
100 triggers per second
Each event is ~1 Mbyte
Physicists work on analysis “channels”
Each institute has ~10 physicists working on one or more channels
Data for these channels should be cached by the institute server
Physics data cache
~PBytes/sec
Tier2 ITEP ~1 TIPS
RDIG JINR ~1 TIPS
~Gbits/sec
Tier 0
Tier 3
1 TIPS = 25,000 SpecInt95
1 PC (2000) = ~15 SpecInt95
1 PC (2007) = 250 SpecInt95
Tier 2
The LHC Computing Model
The LHC Computing Model
The LHC Computing Model
The LHC Computing Model
The LHC Computing Model (RDIG)
The LHC Computing Model (RDIG)
Tier 1

EGEE: Partners
Leverage national resources in a more effective way for broader European benefit
70 leading institutions in 27 countries organised into regional federations

Состояние ресурсов участников Проекта EGEE

EGEE Activities
JRA1: Middleware Engineering and Integration
JRA2: Quality Assurance
JRA3: Security
JRA4: Network Services Development
SA1: Grid Operations, Support and Management
SA2: Network Resource Provision
NA1: Management
NA2: Dissemination and Outreach
NA3: User Training and Education
NA4: Application Identification and Support
NA5: Policy and International Cooperation
24% Joint Research
28% Networking
48% Services
Emphasis in EGEE is on operating a production
grid and supporting the end-users
Starts 1st April 2004 for 2 years (1st phase) with EU funding of ~32M€

EGEE activities’ relative sizes
“Networking” (NA1-5): 30%
Emphasis in EGEE is on operating a production
grid and supporting the end-users
Mware/security/QA
(JRA1-4): 22%
Grid operations (SA1,2): 48%

Grid Operations Management Structure

Активности России в EGEE (Общая координация - Ильин В.А. НИИЯФ МГУ) NA2 - Dissemination and Outreach (Руководит Т. Стриж, ОИЯИ)
Operate the dissemination Web site, support mail lists and Web base collaborative tools, organize presentations, visits, 2 project conferences and showcases per year, publication of dissemination material.
NA3 – User Training and Induction (Руководит Е. Слабоспицкая, ИФВЭ)
Produce training and course material. Deliver on-site courses and support distributed courses in the project.
NA4 – Application Identification and support (Рук. Е. Тихоненко, ОИЯИ)
Support HEP and Bio pilot applications. Identify early users. Introduction of new user communities. Definition of common application interfaces and tools. Creation of an Industry Forum.
SA1 – European Grid Support, Operation and Management (Руководят А. Крюков, НИИЯФ МГУ и Ю. Лазин, ИФВЭ)
Core infrastructure services. Grid monitoring and control. Middleware deployment and resource induction. Resource and user support. Grid management.
SA2 – Network Resource Provision (Рук. В.Добрецов, КИ)
Definition in collaboration with RN Geant,...

Развитие Проекта EGEE

Developing Grid Standards
Increased functionality,
standardization
Custom
solutions
1990
1995
2000
2005
2010
Slide courtesy of Ian Foster, Argonne National Labs

Полезные ссылки:
European DataGrid (EDG) www.edg.org
LHC Computing GRID (LCG) cern.ch/lcg
GridLab www.gridlab.org
EGEE public.eu-egee.org
RDIG www.egee-rdig.ru
GRIDCLUB www.gridclub.ru
GRID в ИФВЭ www.ihep.su/egee