In this report, I shall analyse and research the changes improvements that have been incorporated in the new version of Microsoft Exchange 2013. Organisations and people have changed their means of communication to a collaborative focused method of communication. As a result Microsoft has added a range of new rich features, technologies and services to Exchange 2013 to meet the needs of the organisations and people who use it. There are a large amount of improvements and changes that have been made in this version; however, I shall focus on a few that I consider more important, based on the knowledge I have gained on the HP blocks.
In this report, I shall analyse and research the way Microsoft Exchange's has evolved and the changes that have been brought about.
Email messaging has become a prominent aspect of our lives that it is hard to remember not having it. It is no longer a luxury for businesses, non-profit organisations and even individuals to have their own mail server.
Get your grade
or your money back
using our Essay Writing Service!
In today's business world, email is a mission critical application, which has a major impact on both revenue generation and customer satisfaction.
Over the years, they have evolved from providing basic functionality such as, email to encompassing rich collaboration features.
User's messaging requirements have also changed from basic desktop email access to collaboration available on multiple devices.
Microsoft Exchange 2003 was launched in late 2003, and improved the ability to configure front-end/back-end server architecture, which was originally introduced in Exchange 2000. Configuring the front-end/back-end server architecture allows users to access the Exchange 2003 environment by means of the POP3, IMAP4 and HTTP, clients without having to reference the specific Exchange server where their mailbox is hosted. Microsoft recommends one front-end server for every four back-end servers (suggestion, not a rule).
The front-end Exchange server does not host any mailboxes, but simply acts as a proxy (intermediary) server, which handles requests from the user's client computer to the back-end Exchange server. When a request is sent from the client computer, the front-end server queries Active Directory (AD) using the Lightweight Directory Access Protocol (LDAP) to determine which back-end server hosts the user's mailbox. These details are then forwarded to the back-end server, which processes the request and returns it back to the client computer via the front-end server.
A front-end server should have fast CPUs and a considerable amount of memory. They do not require fast or large disks, unless SMTP has been enabled, as SMTP commits queued mail on the local disk of a front-end server. However, no user data is stored on the disk when using POP3, IMAP4 or HTTP protocols.
The bridgehead server accepts mail from other server then distributes the mail to the next server in the route. It forms as a connection point between routing groups, external systems and remote system. For example, if a US based user sends a message with a 5MB attachment to 10 different UK based users, approximately 50MB of mail will be routed from the US site to the UK site. However, if each site had a bridgehead server, then only 5MB would be routed between the US and UK site. Therefore, the function of the bridgehead server is to consolidate the bandwidth between sites.
The figure below showcases the recommended Exchange 2003 front-end/back-end server architecture.
As demonstrated in the figure above, all POP3, IMAP4 and HTTP connections are routed via the front-end Exchange server; however, Outlook MAPI connections are sent directly to the Microsoft Exchange Server Information Store (MSExchangeIS) service, which is located on the back-end server.
The back-end Exchange server hosts mailbox stores (public folders and mailboxes) and routing configuration. The MSExchangeIS (Store.exe) is the heart of an Exchange 2003 server, and it comprises of a number of components.
The figure below is a graphical layout of the components that is found in a typical back-end Exchange 2003 server, where the Store.exe service runs.
Depending on the version of Exchange 2003, there are some differences in terms of the number of Store Groups, Stores and Store size limits. The table below provides more details.
Always on Time
Marked to Standard
Each Store Group will contain one or more Public Folders and Mailbox stores. The system mailboxes and users' mailboxes are hosted within the Mailbox store, and public folders and their contents are stored in the Public Folder store. Each Mailbox and Public Folder store comprises of two files: A Priv1.edb file (Mailbox) and Pub1.edb file (Public Folder), which are rich-text files that contain the text attachments, email messages and headers for user's emails/file stored in the Public Folder tree. In addition to these files, there are the streaming files that contain the multimedia data, which is formatted as MIME data. These files are Priv1.stm and Pub1.stm files for both the Mailbox and Public Folder stores.
Unlike figure 2, depending on the needs of an organisation and the version of Exchange, a single Store Group, which contains one Public Folder store and one Mailbox store, is considered to be sufficient; however, as the database grows in size, it may become more difficult to manage and backup. As a result, it is recommended to split the database into multiple smaller databases, which may require server hardware or additional licenses.
Overall, at the time this version of Exchange was released, it was considered to be the best in its class. There are a number of improvements that Exchange 2003 brought about such as, the front-end/back-end and bridgehead server architecture, compatibility modes that enabled administrators to migrate user's mailboxes to the new system slowly with the least amount of downtime and the enhanced disaster recovery, which allowed administrators to bring the server back online more quickly.
However, as any software company, in order to stay the best in its class and keep abreast with the evolution of technology they released Microsoft Exchange 2007 four years later.
Microsoft Exchange 2007 was launched in September 2007. This version of Exchange introduced several architectural changes and other features, which enhanced the overall performance and its scalability compared to Exchange 2003. The table below summarises the key changes and features that have been implemented, which include the memory usage, storage groups and the server roles.
The new server roles concept has replaced the front-end, back-end and bridgehead server roles, which were in Exchange 2003. Messaging administrators can now strategically split functions of an Exchange server and place each role on a separate server, or a combination of roles on a single/different server within the organisation. This can be done for many reasons i.e. performance, management or any other reason deemed necessary.
As illustrated in the above table, the five new server roles are:
Each of these roles played a unique part within the Exchange organisation, and some of them are mandatory and others cannot be consolidated within a single server.
The Mailbox Server role is similar to the back-end server role in Exchange 2003, which hosts all mailboxes, public folders and other messaging data, such as meeting items, address lists and resource schedules. In this version of Exchange, it also accepts MAPI connections from Outlook clients on the LAN (illustrated in figure 5), as it did in Exchange 2003. When deploying the Mailbox role as high availability using SCC or CCR replication, based on Microsoft Cluster Services, no other roles can be consolidated with the Mailbox role.
The Client Access Server role or CAS is similar to the front-end server role in Exchange 2003. It provided client access for OWA, ActiveSync, POP3, IMAP4 and external Outlook client access by means of Outlook Anywhere (RPC/HTTP), without a Virtual Private Network (VPN) connection (illustrated in figure 5).
The Hub Transport Server role is similar to the bridgehead server role in Exchange 2003, and is a mandatory role in Exchange 2007, and it is responsible for functions such as, server-side rules and journaling. It also handles user message delivery and internal message flow across Exchange components (cross-site). Lastly, each AD site that contains a Mailbox server role must have a Hub Transport server deployed within it.
The Edge Transport Server role is deployed in the organisation's perimeter network and is designed to establish a boundary between the Internet and the organisation's network, reducing the attack surface. This server does not have access to AD; however, all recipient and configuration is stored in an Active Directory Application Mode (ADAM) directory service on the server. EdgeSync is used to establish one-way replication between the Hub Transport server and the ADAM instance on an Edge Transport server. Therefore, this server cannot be consolidated into a single-server system with other servers because of its location (perimeter network). It also provides rule-based protection, filtering and scans all mail for potential viruses or spam.
This Essay is
a Student's Work
This essay has been submitted by a student. This is not an example of the work written by our professional essay writers.Examples of our work
Unified messaging is new to Exchange 2007, it integrates faxes, voice mail and email messages into user's mailboxes. In addition, Outlook Voice Access allows users to access their email, voice mail, calendars, contacts, faxes and directory from a telephone.
These five new server roles add new features and refine the traditional roles, which are available in Exchange 2003. The figure below illustrates the typical placement of the roles and mail flow within an Exchange organisation.
Where memory is concerned, Exchange 2003 is a 32-bit application, and therefore is limited to 4GB of addressable memory. In contrast, Exchange 2007 memory limitations have been overcome by providing 64-bit application support when running on Microsoft Windows 2003 x64 platforms.
In addition, Microsoft has increased the number of supported Storage Groups to 50 and the maximum database size to 16TB, which is a significant increase from Exchange 2003 that only supported a maximum of 4 Storage Groups and a maximum database size of 16GB for the Standard edition. This enables messaging administrators to split mailboxes across multiple Storage Groups, allow databases to grow larger and simplify their backup and recovery operations.
Traditionally, fax data, email and voice mail have been maintained in a separate system, but with the introduction of the Unified Messaging role in Exchange 2007, these three types of data can now be routed into user's Outlook Inboxes, which allows the user's inbox to act as a consolidated repository. This data can be accessed from mobile devices, an Outlook client, or via a web browser using Outlook Web Access (OWA). Lastly, a unique flag is appended to the emails that contain the voice mail or incoming fax.
There are four types of Exchange clustering:
Local Continuous Replication (LCR) is single server solution, which uses asynchronous log shipping and log replay technology that creates/maintains a copy of the Storage Group on a second physical disk of the same production server.
Single Copy Cluster (SCC) is a clustered Mailbox server, which uses a shared storage location, configured in a failover cluster allowing many servers to manage a single Storage Group copy.
Cluster Continuous Replication (CCR) is a combination of asynchronous log shipping and replay technology, with the management and failover provided by the Microsoft Cluster service.
Standby Continuous Replication (SCR) is designed to be used as a standby server. It extends the continuous replication feature by means of log replay technology used by the LCR and CCR to provide additional deployment configurations and options.
Overall, this version of Exchange introduced a number of cutting-edge features and improvements, which added to the resilience and scalability of Exchange 2007. In addition to the new "server roles" concept, Microsoft also improved the OWA client, allowing user to open email attachments such as, office documents (Word, Excel, PowerPoint and PDF) via their web browser even though Office is not installed on the client computer.
Although at the time these features might have seemed sufficient in 2007. In order for Microsoft to keep up with technology and their competitors, they made further improvements and introduced Exchange 2010 approximately three years later.
Microsoft initially released Exchange 2010 during November 2009. There are a number of new and rich features, services and technologies, which have been introduced in this version of Exchange such as, high availability for the Mailbox server and Client Access server to provide a higher level of messaging continuity. These new feature and improvements enable customer of all segments, of any size to deploy a more efficient and resilient messaging service. A few of the key feature and improvements are highlighted in the table below.
As illustrated in the above table, in Exchange 2007 there are four different types Exchange clustering available for the Mailbox server role. These are LCR, SCC, CCR and SCR. In contrast, the Mailbox server in Exchange 2010 has replaced these four types with one high availability (HA) option, which is called Database Availability Groups (DAG). The DAG is essentially a collection of Mailbox server (up to 16) that provides HA in an Exchange 2010 environment. The underlying continuous replication technology that the DAG uses is a combination of CCR and SCR, which was first introduced in Exchange 2007. In addition to the continuous replication technology, it also makes use of the Windows component, Windows Failover Clustering.
If required, a DAG can be split across AD sites (Geographically Dispersed Clusters), which could provide mailbox HA within a single datacentre or across multiple datacentres (Stretched DAG). The figure below illustrates a typical configuration of a Stretched DAG across two datacentres.
The concept of Storage Groups in Exchange 2007 is no longer available in Exchange 2010. The database, log files and a check point file is now called a "database", which is similar to CCR in Exchange 2007, where a single Storage Group could only contain one database. Therefore, as demonstrated in the figure below, there are three Mailbox servers, each with one active database and two passive database copies. Unlike Exchange clustering that required the entire cluster group to be failover to an alternative node. In Exchange 2010 DAG each mailbox database can be failover or switchover independently when necessary without affecting user connections to other mailbox databases within the DAG.
The five server role concept is still present in Exchange 2010; however, contrary to Exchange 2007, all client access connections to the mailbox database, including client MAPI connections are now sent through the CAS server(see figure 10). It is therefore recommended to implement a CAS array within an AD site (cannot span across multiple AD sites), which provides a single name endpoint i.e. mail.domain.co.uk. Therefore, by implementing a CAS array using Network Load Balancing (NLB) it enables client access to be highly available. This allows for improved redundancy and network load balancing resulting in minimal client impact in the event of a database switchover/failover. In addition, should the need arise, moving a mailbox from one server to another (intentionally or unintentionally) the user will still be able to connect to their mailbox with little or no impact; this was not the case in previous versions of Exchange. The figure below showcases the typical configuration of high availability for an Exchange 2010 CAS array.
Shadow Redundancy is a concept that protects email messages whilst in transit. In the event that the Edge or Hub Transport server fails, after the message has been received for processing, but before it was delivered to the next server for example Hub to Edge or vice versa. The transport server that sent the message is able to detect the failure and redirect it to a different Edge or Hub Transport server depending on the direction of the message.
In the diagram below, one can see the architecture and mail flow for an Exchange 2010 messaging network.
The most visible improvement to end users is OWA, which is now referred to as Outlook Web App (Outlook Web Access in previous versions). The aim of implementing this new version of OWA is to create a seamless cross browse experience for users, whether they are using Safari on an Apple Mac or Internet Explorer on a Windows computer. In addition to that rich user experience, Microsoft has narrowed the gap between the Outlook client and the web based client (OWA). Lastly, there are a lot of new features that this version has introduced such as, Search Folders, integrated Office Communicator (Lync) etc. The image below showcases the improved user experience that Outlook Web App in Exchange 2010 has to offer (right) versus Outlook Web Access in Exchange 2007 (left).
The new technologies, features and improvements that Exchange 2010 brought about has helped a number of organisation archive higher levels of resilience and high availability with more flexibility, but at the same time offering an improved user experience. This version of Exchange is still considered the best messaging application in its class, and will be for years to come. However, whilst most organisations are still using and upgrading to this version of Exchange, Microsoft just recently released Exchange 2013. It may take some time before organisations decide to upgrade to Exchange 2013.
Microsoft Exchange 2013 is the latest version and it was released in October 2012. Microsoft has made a number of major changes to the server role architecture as well as theâ€¦
In Exchange 2007, Microsoft introduced the multi-server architecture, which was refined and improved in Exchange 2010. These roles have now been consolidated into just two roles in Exchange 2013, similar to that of Exchange 2003.
The two Exchange server roles are:
Client Access Server
These two roles can be installed on separated hosts or co-exist on a single host. As in previous versions of Exchange the Client Access Server (CAS)/Front-end server acts as a proxy that forwards client requests to the appropriate Mailbox Server.
Making the CAS server highly available was introduced in Exchange 2010 (CAS array); this feature is also available in Exchange 2013, by means of network load balancing. There are two components that are required to make this possible; namely, the Client Access service and the Front End Transport service.
The Client Access service is responsible for handling connections between the client and the mailbox server.
The Front End Transport service routes emails between the outside world and the Exchange servers, and it performs numerous traffic functions.
Since there are only 2 server roles, the other
One of the most significant changes that Microsoft has made to Exchange 2013 is it architecture. In earlier versions, for example Exchange 2010, there were five server roles. These roles are the Mailbox Server, Client Access Server, Hub Transport Server, Edge Transport Server and Unified Messaging Server.