Cloud+ Topic - Synchronous/Asynchronous Remote Replication
Video "Analyzing Storage Provisioning In The Cloud", Scenario 4:
Wes's answer to the hypothetical is "D" - Asynchronous Remote Replication.
I am totally confused by the answer and the reasoning given.
My understanding from reading "Comptia Cloud+ Study Guide" by Ben Piper is this:
Synchronous replication means that when data is saved, it's written to both the main and backup storage at the same time. The system only tells you everything went well after both places have finished saving the information. This way, the two copies are always kept in sync.
With Asynchronous replication, there's a bit of a wait when data gets copied over to the backup spot. It's all about getting things eventually lined up because it sends the data along in a kind of pass-it-forward system. This method works well for databases that aren't constantly being updated with new info.
So, in a disaster recovery situation, if the disaster happens during the delays related to ARR copying over to the backup spot, there's a greater possibility for data loss, as opposed to Synchronous replication that that syncs both areas simultaneously. It is not "one way"
Please confirm Option 4 is indeed the correct response to Ronnie's scenario.
This is a REALLY tough question because both asynchronous and synchronous replication can be used in conjunction with DR in a scenario like this. In the actual exam, I think we would get more information to make a decision between the two options of asynchronous and synchronous.
The big decision point, as you alluded to, is the level of assurance for the local and the remote data sets to be in sync with each other. Synchronous remote replication is more "expensive", but does offer a higher level of data retention.
We will certainly make this more clear in the next update to Cloud+. Thank you so much for the post.
at 11:00 a.m, @wes-bryan explains why he chose it for the remote replication--one is a "one-way" and other other is a "two-way" synchronization. I mention later that we have to careful because the terminology is tricky and the context are key in this example.
Yes, that's true, But the DR context itself does not invalidate SRR as a viable, or even the preferred and correct choice. I'm sorry, does it?
Sorry for my confusion here!
Thanks, Ronnie! :)
In all instance, you must pick the option that is the "better" or "best" option for the scenario given. The idea of the scenario was never about pick all the valid answers but the best answer for the context. But when it comes facing an exam, Wes and I presented a very valid scenario that doesn't give you right vs wrong answer but good vs better vs best for the scenario.
Thanks, Ronnie. I've been a member of ITPro TV for 4 years now and I've learned a lot in that time. Particularly, from you and Wes.
I guess the main point here is the definition of Synchronous RR stated in the last scenario is not entirely correct.
SRR replicates both ways simultaneously.
The criticality of the data being stored would be a factor. If it was critical data, SRR would be the choice, if not, ARR would be acceptable.
Also, cost effectiveness would influence a decision between the two. One is more expensive than the other.
Sorry, I don't mean to be a hard head about this. I normally include supplemental reading material to enhance and deepen my learning experience. I came across the definitions in my reading and happened to remember the 4 scenario in this video mentioned above.
I understand everything you wrote in your last reply.
Thanks very much! I appreciate all the replies! I don't use the message boards often, but I thought this topic was worth questioning.
Keep up the good work!