At VMworld Barcelona 2014 next week, we are announcing Metro Availability, a technology that enables continuous availability across different sites. This, in my opinion, is one of the biggest and the most significant features we have added since our inception.
In my previous blog, I spoke about some of the Data protection capabilities Nutanix offers. For obvious reasons, I didnt cover Metro Availability there. In this blog, I will give you a quick overview of what Metro Availability is, some of the use cases and basic requirements for customers to get going with this feature. We will follow-up with more blogs that goes into the setup and configuration, best practices, design considerations etc.
Customers deal with demanding SLAs and need to build resiliency into all aspects of their datacenter. While traditional solutions have provided redundancy at individual hardware layer, customers have challenges dealing with downtime due to infrastructure glitches, site maintenance or natural disasters. A recent research from Ponemon institute predicts that customers, on an average, incur a cost of $690,000 per data center outage incident.
Most solutions that exist today are expensive, complex and built with archaic principles and infrastructure. As a result, majority of the workloads are not protected or are under-protected and vulnerable to outage. With NOS 4.1 release coming out later this year, we are introducing a new capability called Metro Availability using which data is synchronously written across different sites enabling near 100% uptime during an entire site failure due to an unforeseen disaster or a planned event.
With Metro availability, we now have an entire spectrum of solutions protecting your data and applications depending on your SLAs and RPO/RTO needs, all built into the platform. Setup and management is simple, intuitive and done from within our Prism UI. This feature will be a true game changer in how customers view DR solutions. The simplicity and ease of management is unparalleled and for the first time enterprises will have a modern consumer-grade experience when it comes to Disaster Recovery and High Availability.
Here is a quick Q&A about the product:
1. How is it priced?
Metro Availability will be part of the Ultimate software edition. The primary site and the secondary site will need to have Ultimate license.
2. Can customers buy metro availability for existing clusters?
Yes. If the customer already has Ultimate edition software, then they can upgrade to 4.1 to get the feature.
3. Can I have different platforms on the two sites?
Yes, as long as both sides are Nutanix clusters you can have different HW platforms on both sides.
4. What are the network requirements between the two sites for Metro Availability?
Nutanix Metro Availability is flexible in terms of network connectivity between the two sites. IP networks can be used. There are three requirements for the network:
- Bandwidth: There is no hard-set bandwidth requirement. Bandwidth required depends on the workload profile. For write-heavy workloads, more bandwidth is recommended.
- Latency: The maximum latency between the two sites should be less than 5ms round trip time
- Redundancy: Recommend having redundant physical network to ensure connectivity
- Host connectivity: If customers want to use Metro Availability in conjunction with VMware HA (recommended), then ESXi hosts on both sites need to be configured appropriately
5. How is Nutanix Metro Availability differentiated?
Following are a few areas where MA differentiates in comparison to other solutions:
- No additional hardware components
- Flexible network rules (as long as RTT latency within 5ms with ample bandwidth)
- Can be enabled on existing Nutanix clusters (with an upgrade to NOS 4.1)
- Can mix Nutanix NX models within clusters
- Managed through Prism – our single pane UI
- Uncompromising simple management: 5-click setup, 1-click failover, 2-click failback
6. What are some use cases that we can sell Metro Availability into?
Metro Availability enables data loss prevention and 100% uptime. This functionality can be used in several scenarios:
1. Disaster Recovery: MA enables zero-RPO and near-zero RTO disaster recovery in the event of an unplanned site outage. If a site fails, a standby site takes over data serving responsibilities for both sites.
2. Planned Events: In the case of planned events such as site maintenance, software/hardware maintenance and data migrations/disaster avoidance, metro availability offers continuous data availability during the event with seamless transition.
I hope this gives you a quick overview of Metro Availability. We will get into more details (including demo videos) very soon. Stay tuned.