I get a ton of emails asking me exactly how ESRI’s geodatabase replication works. It is pretty slick to be honest so I think everyone who has SDE should probably learn more about it. ESRI has posted a podcast titled “Geodatabase Replication: Working with Replication that explains in more detail what geodatabase replication is.
Geodatabase replication enables the distribution of datasets across several geodatabases, and provides a mechanism for keeping those datasets in synch by sending changes over the network or the Internet. This discussion contains suggestions for planning a replication strategy that will help you implement a comprehensive distributed data workflow. The mechanics of creating and synchronizing replicas are also discussed.
Learning about geodatabase replication is cool!
10 responses so far ↓
1
Lefty
// Nov 20, 2007 at 11:11 am
I love the ESRI Podcast Dude.
Seriously though, geodatabase replication interests me greatly as we have many offices spread around the world working on many of the same projects. I really need to read up more on this subject.
2
Owen Evans
// Nov 20, 2007 at 12:20 pm
James, thanks for posting about my replication podcast, and I hope you find it useful. I’m working on some additional replication topics for future podcasts including geodata services, replication workflows, and tips and tricks.
3
deadplanet
// Nov 20, 2007 at 1:03 pm
Will there be a way to replicate a SQL Server 2008 database in a similar way to ESRI geodatabase replication?
4
Bill
// Nov 20, 2007 at 1:33 pm
@deadplanet
You may want to check out Matt Priour’s comment in James’ post about SQL 2008 Express.
5
J Wallis
// Nov 21, 2007 at 6:58 am
@#$@%! I don’t WANT a podcast. I want a white paper and a how to!
I @#%$@#! hate LAZY documentation teams!
Pull out a keyboard you iPod wearing slacker bums!
I swear, the people who came up with this “great” idea to provide intense technical discussion as a “podcast” are probably the same people who use “video resumes”.
Would someone please fire these people and hire some REAL technical writers?
6
David E. Wright
// Nov 21, 2007 at 3:14 pm
Ok, but how is this replication happening?
Is this the ArcCatalog version where you have to manually do it? Or is there a scheduled process that you can use to kick it off?
I agree with J Wallis we need a White-Paper/HowTo as well as a Forum topic for HA SDE Platform Support!!!
7
James Fee
// Nov 21, 2007 at 3:56 pm
David this will explain all….
http://webhelp.esri.com/arcgisdesktop/9.2/index.cfm?TopicName=Understanding_distributed_data
8
James Fee
// Nov 21, 2007 at 4:26 pm
And there is a whitepaper available for those who don’t want to listen to ESRI podcasts at work
http://support.esri.com/index.cfm?fa=knowledgebase.whitepapers.viewPaper&PID=66&MetaID=1314
9
J Wallis
// Nov 26, 2007 at 1:43 pm
Thanks for the whitepaper link. After reading it I’ve come to the conclusion that replication still sucks.
I get SDE export files on a monthly basis from a vendor. The procedure is to “init” the tables and dump in all the data from these texport files. In this scenario I have to essentially sync the entire database all over again. ESRI needs to have a way to deal with this.
10
Fran O
// Jan 22, 2008 at 12:09 pm
I’m hitting my head hard on the geodatabase replication right now. All I wanted was to have one group of developers maintain their data, but not hose up production. I thought that I could set this up with multiple one-way replicas (one from their sandbox to my dev, then from my dev to my test, and when it had been fully checked out, from my test to dev without having to schedule a server outage. But it’s now looking like I can’t do that without using two-way replication — which then introduces the possibility of a data oopsy back to the source The one way replicas are read-only, and as such can’t refresh other replicas. Not sure why this should be so hard.
Leave a Comment