[MarkLogic Dev General] Data sync up within ML clusters in different zones
prakashkumar.patel at tcs.com
Thu Jan 21 00:13:44 PST 2016
We are supporting a content publishing system where in there are writers who inputs a story which passes through a document work flow where editors updates the story and creates a version of story that get published at the end in the newsletter. These newsletters are published on a daily basis. The writers and editors are spread across different geographical regions and it is the latency of the application that we need to have three active data centers.
Please find my comments to yours questions below.
From: general-bounces at developer.marklogic.com [mailto:general-bounces at developer.marklogic.com] On Behalf Of David Gorbet
Sent: Wednesday, January 20, 2016 10:51 PM
To: MarkLogic Developer Discussion
Subject: Re: [MarkLogic Dev General] Data sync up within ML clusters in different zones
* Is this database insert-only or does it also allow updates?
? It allows updates as well
* If it allows updates, are those updates mastered in one of the three locations for a given document, or is it possible for any location to update any document at any time?
? It is possible for any location to update any document any time.
* If it's possible for multiple locations to update the same document at the same time, what are the business rules for resolving conflicts?
? Currently we have only one data center in USA which takes care of all the traffic with lots of Singapore users complains regarding the slowness.
* When you query, do you need to ensure a consistent view across data in all three locations (even if that means it's only up to date with respect to the slowest location), or do you want an inconsistent view, but one that has the most recent data from each location?
? We want a consistent view across data in all the three locations.
In general, anything you can tell us about the application could shed light on the query and update patterns would help, including why you want three active data centers in the first place (is it latency of the application, or something else?
From: general-bounces at developer.marklogic.com<mailto:general-bounces at developer.marklogic.com> [mailto:general-bounces at developer.marklogic.com] On Behalf Of Prakashkumar Patel
Sent: Wednesday, January 20, 2016 6:03 AM
To: MarkLogic Developer Discussion <general at developer.marklogic.com<mailto:general at developer.marklogic.com>>
Subject: [MarkLogic Dev General] Data sync up within ML clusters in different zones
We are using Marklogic 7 and we are having three different ML clusters which are all primary in USA, UK and Singapore. They are all supporting the same application that inserts data into those clusters. Data in all the three clusters has to be same. For example if there is a document insert in USA cluster, it has to be inserted in UK and Singapore clusters as well. Same applies for any document delete. And all the three cluster has to have the most latest version of a document if that is getting updated frequently.
Is there any feature in Marklogic that enables us to do this? If not, is there any best practice that you will suggest us to follow?
Notice: The information contained in this e-mail
message and/or attachments to it may contain
confidential or privileged information. If you are
not the intended recipient, any dissemination, use,
review, distribution, printing or copying of the
information contained in this e-mail message
and/or attachments to it are strictly prohibited. If
you have received this communication in error,
please notify us by reply e-mail or telephone and
immediately and permanently delete the message
and any attachments. Thank you
-------------- next part --------------
An HTML attachment was scrubbed...
More information about the General