Overview
Microsoft
Sync Framework is a comprehensive synchronization platform that enables
collaboration and offline access for applications, services, and devices. It
features technologies and tools that enable roaming, sharing, and taking data
offline. Using Microsoft Sync Framework, developers can build sync ecosystems
that integrate any application, with any data from any store using any protocol
over any network.
Benefits of Synchronization
Synchronization can be
described as the process of bringing together two end points or data stores.
When the contents of the two data stores are the same, they are known to be in sync with each other.
For example, if you want to synchronize two databases manually,
you
do the following.
1 1 Determine changes in the source database.
·
Takes the application,
store, or services offline: The biggest
advantage of synchronization is that it enables you to take your application
offline. If you build synchronization into your application, users of the
application can interact with their local data stores until they need the items
that are not contained in a local repository or until the application is back
online.
·
Builds a faster and
richer user interface: Building synchronization into the
application allows you to build a richer user interface without worrying
about the performance of the application. Because data is usually
fetched from a local store, your application can provide faster responses.
·
Reduces the network
cost: Sync-enabled applications or services
upload and download only incremental changes, thereby reducing the amount of
the data that needs to be sent over the network.
Now
that you understand the need for synchronization, the next section examines some
issues associated with synchronization and how Sync Framework can resolve them.
Life Before Sync Framework
The
importance of Sync Framework can’t be understood without discussing problems
involved
while implementing synchronization, including the following:
· Storage and application errors and failover handling: Imagine that you’re synchronizing two databases, and an
application error such as a connection timeout or a constraint violation occurs
while changes to the destination database are applied. What happens to the
record that needs to be synchronized? It is the responsibility of the Sync
Framework to recover from such application and storage errors.
•
Network failure: Imagine the same example of synchronizing two databases. What
happens
if one of the databases is downloading changes from another and suddenly
the
Internet connection goes down? Sync Framework can recover the network
failures.
•
Conflict detection: A conflict is said to occur if the same item was modified at
both
end
points at the same time. The Sync Framework should be able to detect the
conflicts
and provide a mechanism to resolve or log the conflicts.
Installing Microsoft Sync Framework
At
the time of writing, Microsoft Sync Framework 1.0 was the latest release. You
can get a
copy
of the Microsoft Sync Framework software development kit (SDK) in three ways:
•
Sync Framework ships with Microsoft SQL Server 2008.
•
Sync Framework ships with Microsoft Visual Studio 2008 Service Pack 1.
•
You can download the Sync Framework from the Microsoft download page:
www.microsoft.com/downloads/details.aspx?FamilyId=C88BA2D1-CEF3-4149-
B301-9B056E7FB1E6&displaylang=en.
Core Components
The Sync Framework comes with support for ADO.NET, file systems,
RSS, Atom feeds,
and even custom data stores. You can divide the components of the
Sync Framework into
the following three categories:
• Microsoft Sync Framework runtime: This SDK allows developers to use the built-in
sync providers and create their own sync providers.
• Metadata services: Provide the necessary infrastructure to store the
sync metadata.
Sync metadata is used by the Sync Framework runtime during the
synchronization.
The Sync Framework ships with the built-in Microsoft SQL Server
Compact Edition
(CE) that can be used to store the metadata.
• Built-in providers: The Sync Framework ships with the following three
built-in
providers:
o
Synchronization services for
ADO.NET provide offline and collaboration support
for ADO.NET enabled data
stores.
o
Synchronization services for
SSEs are the built-in providers for synchronizing RSS
and Atom feeds.
o
Synchronization services for
file systems are the built-in providers for synchronizing
files and folders on
Win32-compatible file systems.
Synchronization Flow:
Synchronization
providers enable you to synchronize data between different replicas.
Replicas
are also known as end points or data stores. The actual data is stored in the
replica. You need to have one sync provider for each replica for the replica to
synchronize its data with other replicas. A replica synchronizes its data with
another replica by establishing a sync session.
As
shown in below screen shot, synchronization providers communicate with each
other using a sync session. The two sync providers are attached to the sync
agent, and the sync application initiates the communication between the two
providers using the sync agent. The sync agent is responsible for establishing
and managing the sync session. Sync providers can receive and apply changes to
the replicas. There are two sync providers on the top of the Sync Framework
runtime: the source sync provider and destination sync provider, respectively.
After
being invoked by a sync agent, the destination sync provider sends its
knowledge
to
the source sync provider. The source provider uses this knowledge to determine
the
changes
and sends its knowledge to the destination. The destination provider compares
its
knowledge with the source, resolves the conflicts, and then sends the request
to the
source
provider for changed data. The source provider sends the changes to the
destination
provider,
and the destination provider applies the changes to the destination replica.
Within
a sync session, synchronization flow is always in one direction. What this
means is
that
the source provider and the destination provider cannot work simultaneously. At
any
given
point within a sync session, information flows between the source and
destination
replicas
or between destination and source replicas, but doesn’t flow simultaneously
between
both.
In
its simple form, a sync session contains a sync agent and two sync providers.
One of the providers is a source provider that sends the changes; the other is a
destination provider that receives and applies the changes. Of course, the sync
agent controls this flow.
The
sync providers shown in below screen shot illustrate a scenario in which the
metadata is
stored
in the built-in metadata store provided by the Sync Framework, which is very
easy
to
use and is built on top of the SQL Server.
Click here to create sample application using Microsoft sync framework.
No comments:
Post a Comment