10-20-2008, 10:31 PM
History
RepServer was developed by Sybase in the beginning of the 90's on request of, and initially funded by a Sybase customer (actually a US-based bank), that wanted a product for data replication. Sybase itself has continued development of the product since then.
RepServer basics
RepServer is a "data movement" product; basically, it just copies data modifications from one location to another. The most important characteristics are:
The unit of replication is a transaction, not just an individual DML statement.
RepServer guarantees that replicated transactions are applied in the same order as they occurred on the primary side.
In principle, when individual components (dataservers, RepServer, network connections) of a RepServer-based system are temporarily unavailable, this should be transparent to the overall replication system. When a component becomes available again, replication should continue with minimal or no further manual DBA action. Most of the time, this works quite well.
How do you get familiar with RepServer ?
I'd say there is no easy, painless way to learn how to use RepServer. As a start, make sure you know everything a normal DBA should know about the RDBMS platforms involved in your replication system. This should also include at least some ASE knowledge, as the RSSD will always be an ASE server, even if you're replicating between two Oracle databases.
When you're replicating between ASE databases, you must understand how the ASE transaction log works, how the log is dumped and truncated, how to enlarge databases and logs, how your choice of disk usage affects recovery issues... in short: the works.
Next, I recommend you take a RepServer training from Sybase ("Fast Track to Replication Server"). After that course, you'll probably be back in your office wondering how you should ever manage to understand this thing -- don't worry, it is possible !
In my experience, the only way to learn RepServer is to create -- and solve -- many problems, because only then will you get a feel for the product. The best way to gain this experience is to set up your own private little replication system from scratch: try to create a setup where you replicate just a simple 1-column table from one database to another. By the time you have succeeded in getting this to work, you will have learned a lot -- even though you already did this on the Sybase training. Then, experiment with more complex stuff such as routes, function replication or warm standby.
Also, make sure you have the RepServer documentation handy: go here for details where to get this from.
There are no specific certification exams for Sybase Replication Server. For replication in an ASE environment, it seems reasonable to expect a RepServer DBA to have an ASE DBA certification.
Why is RepServer so difficult ?
I've heard many different opinions on RepServer, ranging from "wonderful" to, as I once heard someone say, "the worst thing since Pompeii". While the latter may be overstating it a bit, RepServer is certainly not an easy product to set up, administrate or troubleshoot. One reason for this is that RepServer doesn't seem to have been developed with user-friendliness as a prime objective; it's a typical example of a very powerful, flexible, yet difficult-to-use piece of software, which is best appreciated by hard-core technology diehards.
Another factor is that the scope and complexity of your DBA work expands significantly when replication becomes involved: those previously independent dataservers are now suddenly closely related because data is replicated between them. This effectively means that they have become one big system from a DBA point of view, thus making a DBA's job significantly more difficult.
Cost of deploying RepServer
It tends to be pretty expensive to use RepServer. On the one hand, the license fees can be stiff, as they tend to be based on the number of users in your dataservers. More importantly though, you'll need plenty of experienced and skilled DBA staff, both for your dataservers and for RepServer, which tend to be expensive.
What kind of organisations use RepServer ?
Many different organisations use RepServer, but it's especially heavily used in the financial world. Many banks have implemented complex bi-directional replication systems, for example to link their security trading operations at stock exchanges around the world. This is the reason why many RepServer DBAs can be found in places like New York, London and Tokyo.
What about replication for ASA ?
Replication for Adaptive Server Anywhere (formerly known as SQL Anywhere) is done through SQL Remote. This is a different product which is not related to, and works completely different than, RepServer.
Thanks,
ZA
Courtesy :Rob Verschoor
RepServer was developed by Sybase in the beginning of the 90's on request of, and initially funded by a Sybase customer (actually a US-based bank), that wanted a product for data replication. Sybase itself has continued development of the product since then.
RepServer basics
RepServer is a "data movement" product; basically, it just copies data modifications from one location to another. The most important characteristics are:
The unit of replication is a transaction, not just an individual DML statement.
RepServer guarantees that replicated transactions are applied in the same order as they occurred on the primary side.
In principle, when individual components (dataservers, RepServer, network connections) of a RepServer-based system are temporarily unavailable, this should be transparent to the overall replication system. When a component becomes available again, replication should continue with minimal or no further manual DBA action. Most of the time, this works quite well.
How do you get familiar with RepServer ?
I'd say there is no easy, painless way to learn how to use RepServer. As a start, make sure you know everything a normal DBA should know about the RDBMS platforms involved in your replication system. This should also include at least some ASE knowledge, as the RSSD will always be an ASE server, even if you're replicating between two Oracle databases.
When you're replicating between ASE databases, you must understand how the ASE transaction log works, how the log is dumped and truncated, how to enlarge databases and logs, how your choice of disk usage affects recovery issues... in short: the works.
Next, I recommend you take a RepServer training from Sybase ("Fast Track to Replication Server"). After that course, you'll probably be back in your office wondering how you should ever manage to understand this thing -- don't worry, it is possible !
In my experience, the only way to learn RepServer is to create -- and solve -- many problems, because only then will you get a feel for the product. The best way to gain this experience is to set up your own private little replication system from scratch: try to create a setup where you replicate just a simple 1-column table from one database to another. By the time you have succeeded in getting this to work, you will have learned a lot -- even though you already did this on the Sybase training. Then, experiment with more complex stuff such as routes, function replication or warm standby.
Also, make sure you have the RepServer documentation handy: go here for details where to get this from.
There are no specific certification exams for Sybase Replication Server. For replication in an ASE environment, it seems reasonable to expect a RepServer DBA to have an ASE DBA certification.
Why is RepServer so difficult ?
I've heard many different opinions on RepServer, ranging from "wonderful" to, as I once heard someone say, "the worst thing since Pompeii". While the latter may be overstating it a bit, RepServer is certainly not an easy product to set up, administrate or troubleshoot. One reason for this is that RepServer doesn't seem to have been developed with user-friendliness as a prime objective; it's a typical example of a very powerful, flexible, yet difficult-to-use piece of software, which is best appreciated by hard-core technology diehards.
Another factor is that the scope and complexity of your DBA work expands significantly when replication becomes involved: those previously independent dataservers are now suddenly closely related because data is replicated between them. This effectively means that they have become one big system from a DBA point of view, thus making a DBA's job significantly more difficult.
Cost of deploying RepServer
It tends to be pretty expensive to use RepServer. On the one hand, the license fees can be stiff, as they tend to be based on the number of users in your dataservers. More importantly though, you'll need plenty of experienced and skilled DBA staff, both for your dataservers and for RepServer, which tend to be expensive.
What kind of organisations use RepServer ?
Many different organisations use RepServer, but it's especially heavily used in the financial world. Many banks have implemented complex bi-directional replication systems, for example to link their security trading operations at stock exchanges around the world. This is the reason why many RepServer DBAs can be found in places like New York, London and Tokyo.
What about replication for ASA ?
Replication for Adaptive Server Anywhere (formerly known as SQL Anywhere) is done through SQL Remote. This is a different product which is not related to, and works completely different than, RepServer.
Thanks,
ZA
Courtesy :Rob Verschoor