Today I read a blog from Sean McCown that encourages developers to be professionals when it comes to interacting with a database. In a nut shell, (Go read his blog for more) he says that coders should not write code a specific way to make the DBA happy. They should write the code that way because it is right way to access a database.
I appreciate Sean’s sentiment and would like to take that a bit further. This paradigm should be applied to all disciplines surrounding an application developer. Whether it be the DBA, the QA analyst, the Build Engineer or the Automation Engineer. They all have the “right” way to access or to test or to automate. These are not because they want to make things hard for the app dev, no. It is because each discipline has specific knowledge about how their systems work the best.
Just as a DBA can say this way to code will retrieve the data you want faster and more reliably than the other way, an Automation Engineer can and should say, coding this way will make the automation work more efficiently and have a higher quality return rate.
A coder will have a greater understanding of where efficiencies can be gained in his application if he can reach out to these disciplines and understand the “whys” instead of just making someone happy.
Till next time…
Tuesday, January 12, 2010
Monday, January 11, 2010
Fan of Deduplication!
It has been about 4 months now since we have implemented our new backup solution. I must say that I am a fan! Deduplication is eliminating the redundant data in a given set of data being backed up. For instance, we have several VMs with Windows on them. These VMs take up very little space because the Windows bits are only stored once.
We purchased from EMC their Avamar backup solution. We have two replicating nodes, one here in Chapel Hill and one at our data center at Peak10 in Charlotte. These nodes are currently backing up more than 2.7 TB of data and is only using about 1.5 TB of space. Now you might say that “that's not that great”, but wait. That includes 4 months of additions, changes and restore points. Our retention policy is fairly conservative and the backup times are amazing.
Avamar works seamlessly with our VMs, Exchange, SQL Server and our users directories. Backing up in both locations, it gives us a very easy restore scenario. Also, one in which we had to act on just recently with our web server having issues.
I use Microsoft’s Windows Home Server at home which has its form of deduplication as well. And if you read this blog from time to time you know what a fan of Windows Home Server I am.
Till next time…
We purchased from EMC their Avamar backup solution. We have two replicating nodes, one here in Chapel Hill and one at our data center at Peak10 in Charlotte. These nodes are currently backing up more than 2.7 TB of data and is only using about 1.5 TB of space. Now you might say that “that's not that great”, but wait. That includes 4 months of additions, changes and restore points. Our retention policy is fairly conservative and the backup times are amazing.
Avamar works seamlessly with our VMs, Exchange, SQL Server and our users directories. Backing up in both locations, it gives us a very easy restore scenario. Also, one in which we had to act on just recently with our web server having issues.
I use Microsoft’s Windows Home Server at home which has its form of deduplication as well. And if you read this blog from time to time you know what a fan of Windows Home Server I am.
Till next time…
Subscribe to:
Posts (Atom)