MARSYS announces the release of Colada 4.1, a software replication framework
that provides a non-invasive, platform independent solution for clients. Colada
maintains real-time replication and synchronization of data across multiple LAN
and WAN frameworks. In addition, it is a disaster recovery enabler with
the ability to leverage disaster recovery hardware, automatic database
failover, the ability to configure delayed replication and instant backup of
data across multiple servers.
The product is made up of various different components, including the Alarm
Server, the Configuration Server or Configuration Console, the Database Synch
Tool, Client Object, Sequencer and Replicator.
Configuration Console - The Configuration Console simplifies the job of a
DBA. One of the biggest problems of implementing replication today is
that it is not a very simple proposition whether it is in SQL Server or any
other database. Colada's Configuration Console provides a wizard-like
feature that allows you to string databases together into a replication
framework and manage those databases from a central console. The
Configuration Server stores change in the configuration, broadcasting them to
the components.
Replicator - One of the main components within Colada is the Replicator
itself. The Replicator accepts transactions, which are subsequently sent
in a sequential manner to each of the databases to be executed. MARSYS guarantees
that everything is executed in the same order that it was received via this Replicator.
Client Object - One of the initial objectives behind Colada was that it be easy
to implement and easy to access. MARSYS knew the best way to access
things was via drivers, so the client object was created. The Client
Object is a flexible application standard driver (OLEDB, ODBC, JDBC, etc.) used
to access the data Replicator.
Database Synch Tool - The Database Synch Tool examines the transaction logs
generated by the Replicator and Sequencer, and allows you to replace
transactions on any database, should that be necessary.
Sequencer - The Sequencer is used to maintain Replicators in a WAN situation.
In a situation where you are outside of a standard LAN, the Replicators talk to
the Sequencer, which maintains the sequence of all of the Replicators so that
they execute and continue to guarantee results. For example, an individual
taking an online class might start the class in the morning, and complete it in
the afternoon after stepping off a plane in a different location. It is
essential that the data from the data center move with him as he accesses a different
data center later in the day. Colada assists in that process, ensuring
that data is replicated to both data centers and is available to the client as
he moves around the United States.
Standby Server - If corruption occurs, the Alarm Server sends SNMP traps, email
or even to pagers if necessary. When you are working with live replication, you
are susceptible to corruption. All transactions are going through very
quickly; if one that can corrupt the database goes through to all of your
databases, it can cause an enormous problem. To offset this problem, MARSYS
created the notion of a standby server, which allows you to configure delayed
replication to individual servers.
There are different types of replication available, with both pros and cons to
their structures. A negative aspect of transactional replication is if your
master dies, your replication pretty much goes down the tubes at the same
time. Snapshot replication is more of a point-in-time capture used for
reporting systems. Merge replication is one of the most difficult to maintain.
Colada, on the other hand, offers application level
replication, or replication at the transactional/data level, replacing the
reliance on database specific replication. The DBA can manage everything
from a single console, which means he only needs to understand a single
technology rather than numerous ones.
Colada sports a simple to use application interface; the configuration console
is all point and click, which makes setting up a replication scenario
effortless. Many of the replication scenarios on the market are complicated,
especially publisher/subscriber situations that require figuring out how the
publishers and subscribers are talking to each other. In these situations, you
would normally have to set it up on two sets of machines, making sure the
subscriber is listening and the publisher is sending. In addition, the
replication console provides the ability to add and delete databases as well as
the ability to start and stop databases.
For more information about Colada see the MARSYS Web site at http://www.marsys.com/solutions/colada.asp
or contact MARSYS at ColadaArticle@marsys.com.
» 