How does Replication work in ECS

Question

I am new to ECS.  We are currently using it with Cloud Array. I am also familiar with Avamar and DataDomain as we use those for our nightly backup and our plan is to use ECS and CloudArray for data archive. We have a second site that is approximately 4 hours from our main site and we replicate down to it. In Avamar there is the Activity tab that I may use to watch the replication.  I can see it's progress and how much is left.  I do not see anything like that in ECS.  There is the Geo Replication in ECS, but that more or less just reports what it has done, not what it is doing and how much it has done and how much is left to go.  This is great in Avamar, and I may be 'spoiled' by that in Avamar, but does anyone know if that level of reporting/monitoring is available in ECS, or is there a tool that may be used to see this? Does anyone know if there is a scheduled time that ECS replicates?  So in Avamar one can set different replication schedules, but I am not seeing that in ECS.  I would like to learn how it works in ECS and when it runs and does it uncompress and then replicate and then re-compress?  Does it encrypt during replication? Thank you for any help in learning how replication works in ECS! Mike

JasonCwik · Answer

Hi Mike,

Replication in ECS is handled on a continuous basis, not on a schedule. On the ECS dashboard, you can see three items:

RPO
Data Pending Geo-Replication
Replication Rate

RPO is your Recovery Point Objective. This is basically how far behind the remote site is in case of a disaster in the primary site, e.g. 30 seconds. The next two settings show how much data needs to be replicated and how fast replication is running. If you click the "Geo Monitoring" header, you can drill-down to see details about replication activities.

All data written to disk in ECS is compressed before writing to chunks, so it will be transmitted over the wire in compressed format. All replication traffic is also encrypted over the wire using AES256. Upon arrival, we write the chunk data in the same format, so there's no decompression or unpacking required.

For more information, I'd suggest the ECS Overview and Architecture White Paper, available on the documentation page: Elastic Cloud Storage 2.2.x Product Documentation

samuel_cox · Answer

What could possibly be the issue if the replication data is super high, like 1.3 TB high, what would be something you think that could be causing that?

ECS

Was this post helpful?