Wednesday, June 4, 2014

Sharepoint Architecture Template Period 2


SOFTWARE ARCHITECTURE DOCUMENTATION FOR
PORTAL
APPLY FARM IN SHAREPOINT

Revision: 0.1



DOCUMENT CHANGE LOG
Revision
Description
Author
Date
Approved
Date












































































































 

Table of contents




List of figures

1      Introduction

·      PORTAL = Automotive Supply Chain Portal, a private community site that internal and external parties need to have an account to login and contribute information.
o     Internal parties refer to MSTECHSHARING user, such as: developer, support rep, sales rep, and marketing.
o     External parties refer to MSTECHSHARING customers that can be 1st tier-supplier and 2nd tier-supplier.

1.1   Purpose

This document defines the MSTECHSHARING Architectural Design based on the MSTECHSHARING Functional Design Specifications. This Architectural Design document shall not repeat those design specifications. It details the high level key system components and also the deployment architecture to meet the functional design specifications. Key technologies are identified for each MSTECHSHARING service component.
The MSTECHSHARING architectural design strives to be both scalable and flexible. The system design is component based and provides great flexibility in deployment. The technical overview defines major system components and related technologies. It also provides deployment architecture.
The use cases mentioned in this document capture the specific functional and business rules applied on each independent business process activity and should enable the system users to validate the functionalities supported implicitly and explicitly by the system.  The use cases also identify the end users of the system.

1.2   Target audience

This document is developed by MSTECHSHARING Portal development team and will be used by project stake holders for development and maintenance.

1.3   Definitions

TBD

1.4   Scope

This document is applied to PORTAL project

1.5   Acronyms, and Abbreviations

Abbreviations and definitions used throughout the document are presented here to aid the reader.
Table 1: Acronyms and Abbreviations
Term
Definition
SRS
Software Requirement Specification
API
Application Programming Interface






1.6   References

1.7   Overview

This document has following contents:
o     Section 1: Show document purposes and overview.
o     Section 2: Describe system overview and features.

2      Overall Description

2.1   System overview

2.1.1     MSTECHSHARING Technology Overview

MSTECHSHARING’s architecture design recommendations are the Functional Requirements document which supported for MSTECHSHARING’s customer and MSTECHSHARING’s staff, so we choose “SharePoint Technology” to manage information of each customer

2.1.2     Architectural Strategy Statement

To meet the MSTECHSHARING’s core requirements of a robust, scalable architecture, a combination of technologies has been recommended. We recommend this flexible suite of solutions to take advantage of the strengths of SharePoint
Figure 1: System overview

2.1.3     Inputs

Overview (dated October 18, 2012).
MSTECHSHARING SRS (dated August 23, 2012).

2.2   Software’s limit and storage data

2.2.1     Software limit

As we know, any software has limit, so I give some limit which it affect to our project directly:
-         300 Content database per Web application
-         200 GB per Content database (SP1 4TB)
-         100 GB per site collection (recommend)
Figure 2: Web Application Structure

2.2.2     Storage data

We create many content databases, each content database we have some site collections, so we can backup - restore database easily.
Figure 3: Store content data in special server

3      Recommended Architectural Components

3.1   Standard Farm model

3.1.1     Overview

There are two key components to those Mapping Services. One is the enterprise strength and spatially enabled relationship base to store, version, and archive both spatial and attribute database, and the other is an expandable server application framework.

3.1.2     Technology Considerations

Based on the MSTECHSHARING Functional Design Specifications document, MSTECHSHARING shall build and publish the standard farm model.

3.1.3     Software and Hardware Requirements

The following requirements shall form the starting point for building a basic Farm architecture. At the minimum One Single Server Farm can include all system components.  
Table 2: Software and Hardware Requirements
Software and hardware
Description
Note
Window server 2008 R2 64bit
Follow to Microsoft’s requirement

Microsoft SQL 2008 R2 64bit
Integrate reporting services and some features

SharePoint server 2010 SP1 64bit
Can store 4TB per content database

Microsoft IIS 6 or above
Follow to Microsoft’s requirement

CPU: Intel(R)Xeon(R)CPU E5640 @2.67GHz
Pending

32 GB RAM
Pending

60TB for many database storage space
Store in many content database example 120 content database, each contain database content 50 company, each company contain 10GB
Just store 500GB per content database, we can backup restore easily



4      SharePoint 2010 farm topology

4.1   Define web server, application server, Database server, query server, crawl server

4.1.1     Web server

This server (also known as a front-end Web server) hosts all Web pages, Web Parts, and Web services used when your server farm receives a request for processing.
Note: Web server or WFE contain “Microsoft SharePoint Foundation Web Application” service. Click to view detail, view more detail, and more.

4.1.2     Application server

This server hosts the service applications running in the farm, such as Visio Services. It includes central admin and many other services.

4.1.3     Database server

This server stores most of the data associated with a SharePoint 2010 implementation — including configuration settings, administration information, data associated with the service applications, and user content.

4.1.4     Query server (Query component)

This server is responsible for querying the index, finding the matching content, and then sending the content back to the Web servers for presentation to users.

4.1.5     Crawl server (Query component)

This server crawl (accesses and catalogs) content sources and then propagates the results to the query servers. The crawl server uses a crawl database to store the URLs of all sources crawled.

4.2   How many topologies for SharePoint Server 2010 and how many server on each topology

4.2.1     Limited Deployments

One – server farm: One-server farm with all the tiers installed on one server.
Two – tier farm: Two-server where the web and application tiers are installed one server and the database would be installed on an existing SQL Machine.
The one-server farm is described as an Evaluation environment for under 100 users while the two-server farm would support up to 10,000 users

Figure 5: Limited Deployments

4.2.2     Small Farm Topology Deployments

Small farm architectures serve a larger number of users and scale out based on how heavily services are used. Because of the greater number of services, including client Web applications, more requests per user are expected in the new version compared with the old version.
Figure 6: Small Farm Topology Deployment

4.2.3     Medium Farm Topology Deployments

The medium server farm illustrated is scaled for search to serve approximately 40 million items. Beyond this search scale, the recommendation is to deploy a dedicated search farm. Scale out all other servers based on the utilization of other service applications and services within the farm and the volume of content the farm will host.
Figure 7: Medium Farm Topology Deployments

4.2.4     Large Farm Topology Deployments

The recommendation for scaling out a large farm is to group service applications, services, or databases with similar performance characteristics onto dedicated servers and then scale out the servers as a group. The following topology illustrates a practical example of this concept. The red text lists one possible way to build server groups.
Figure 8: Larger Farm Topology Deployments

4.3   Server role in each topology: front-end, back-end, database

4.3.1     Web server roles

-         Host Web pages, Web services, and Web Parts that are necessary to process requests served by the farm.
-         Direct requests to the appropriate application servers.
-         This role is necessary for farms that include other SharePoint Server 2010 capabilities. In dedicated search service farms, this role is not necessary because Web servers at remote farms contact query servers directly.
-         In small farms, this role can be shared on a server with the query component.

4.3.2     Application server roles

Application server roles are associated with services that can be deployed to a physical computer.
-         Each service represents a separate application service that can potentially reside on a dedicated application server.
-         Services with similar usage and performance characteristics can be grouped on a server and scaled out onto multiple servers together. For example, client-related services can be combined into a service group.
-         After deployment, look for services that consume a disproportionate amount of resources and consider placing these services on dedicated hardware.

4.3.3     Database server roles

In a small farm environment, all databases can be deployed to a single server. In larger environments, group databases by roles and deploy these to multiple database servers.

4.4   How to solve: we use farm model because number of user and Data storage are large

4.4.1     Starting point architecture follows number of items

Refer this document: 
Table 5: Starting point architecture follows number of item
Number of Items
Starting point architecture
Description
0-1 million
Limited deployment
All Search role can coexist on one or two servers, as shown in the Limited Deployment architecture example
1-10 million
Small farm topology
Move the crawl server role to another server, while the query server role remains on the web servers, as shown in the small farm architecture example
10-20 million
Medium shared farm topology
Add a crawl server. Each crawl server has one crawler. Create another index partition with query component and distribute these across query servers, as shown in the medium shared farm architecture example
20-40 million
Medium dedicated farm topology
Add query servers and index partitions, with distributed query components. Add another crawl database, associated with a new crawler on each crawler server. Also, add another property database to the second database server, and mirror both property databases, as show in the medium Dedicated Farm architecture example
40-100 million
Large dedicated farm topology
Isolate each topology layer into “server group” in which each role is deployed to its own servers. Each server group can then be scaled out to meet specific requirement for the components in that role

4.4.2     At current time:

Our current status if we have 6.000 users, need one Front end web server and database server in SharePoint farm
It can solve:
-         Store many company
-         Store <= 10.000 users
-         Store from 0-1 million items
-         Can add on more server when system is huge (data, user)
Figure 10: Two tier farm limited deployment
The database server: contain all content database and other database
The SharePoint server: This server hosts the service applications running in the farm, such as Visio Services, reporting services, user profile services… and hosts all Web pages, Web Parts, and Web services used when your server farm receives a request for processing

4.4.2.1      How to setup two tiers limited deployment:

Please refer to this document: 

4.4.3     At next time:

Our current status if we have 18.000 users, two Front end web servers and database server
It can solve:
-         Store many company
-         Store from 10.000 to 20.000 users
-         Store from 1-10 million items
-         Can use Network Load Balance
-         Can add on more server when system is huge (data, user)
Figure 11: Small server SharePoint farm with two-tiers
The database server: contain all content database and other database
The SharePoint server: This server hosts the service applications running in the farm, such as Visio Services, Search service Application, reporting services, user profile services… and hosts all Web pages, Web Parts, and Web services used when your server farm receives a request for processing

4.4.3.1      How to setup two tiers small farm:

Please refer to this document: 

4.4.4     In the future

Each company will increase user, so we must change Two-Server SharePoint farm to Small server SharePoint farm with three-tier because each WFE include 10.000 user.
It can solve:
-         Store many company
-         Store from 10.000 to 20.000 users
-         Store from 1-10 million items
-         Optimized search by move search service to dedicate server
-         Can use Network Load Balance
-         Can add on more server when system is huge (data, user)
Figure 12: Small server SharePoint farm with three-tiers
Combined Web and query servers
-         Place the Web server role on the same servers with the query components.
-         There is one index partition, which includes the full index. The index partition includes a primary copy of the query component and a mirror copy (m). For redundancy, the mirror copy is placed on the second computer.
Crawl server
-         The Search Administration component is on the crawl server.
-         The crawl server has one crawler, which is associated with the one crawl database.
Database server (clusters or mirrors)
-         Search Administration database on one server
-         Property database
-         Crawl database
-         All other SharePoint databases

4.4.4.1      How to setup three tiers:

Please refer to this document: 

0 comments:

Post a Comment