AnsweredAssumed Answered

MapR as a VM Datastore

Question asked by dzndrx on Jun 19, 2017
Latest reply on Jun 26, 2017 by dzndrx

Where are currently pushing this architecture for our Network Operation Center.

 

This will run on a 3 node cluster with a mirror cluster to other data center for HA and DR.

The machine specs are not yet clear but generally as my assumption it will be like this.

CPU: 16~32 between these values

Memory:128 GB

Disk: 10kRPM disk plus SSD's (around 22TB total cluster capacity)

 

Number of virtual machines to run (50 VMS) all apps, database and other.

 

For these use case (which has a demo on MapR blog) here's the link How to Use MapR-FS as a Random Read/Write Datastore on VMware ESXi | MapR 

 

I just need an answer for these questions ( I really need assistance on this)

 

1. Total Throughput of a the 3 node cluster (or how to compute these providing the formula)

2. Total IOPS of the 3 node cluster (again how to compute this one)
3. Where can I see a Reference Architecture for these setup? Any deployments from MapR side together with VMWARE.

4. MapR support for Network Jumbo Frames protocol.

5. Where should us get a support from? MapR or VMWare in case something goes wrong ( hopefully not).

Outcomes