Acropolis Hypervisor (AHV) I/O Failover & Load Balancing

Many customers and partners have expressed interest in Acropolis since it was officially launched at .NEXT in June earlier this year, and since then lots of questions have been asked around resiliency/availability etc.

In this post I will cover how I/O failover occurs and how AHV load balances in the event of I/O failover to ensure optimal performance.

Let’s start with an Acropolis node under normal circumstances. The iSCSI initiator for QEMU connects to the iSCSI redirector which directs all I/O to the local stargate instance which runs within the Nutanix Controller VM (CVM) as shown below.

AHVMPdefault

I/O will always be serviced by the local stargate unless a CVM upgrade, shutdown or failure occurs. In the event one of the above occurs QEMU will loose connection to the local stargate as shown below.AHVMPfailedlocal

When this loss of connectivity to stargare occurs, QEMU reconnects to the iSCSI redirector and establishes a connection to a remote stargate as shown below.AHVMPremote

The process of re-establishing an iSCSI connection is near instant and you will likely not even notice this has occurred.

Once the local stargate is back online (and stable for 300 seconds) I/O will be redirected back locally to ensure optimal performance.

AHVMPfailback

In the unlikely event that the remote stargate goes down before the local stargate is back online then the iSCSI redirector will redirect traffic to another remote stargate.

Next lets talk about Load Balancing.

Unlike traditional 3-tier infrastructure (i.e.: SAN/NAS) Nutanix solutions do not require multi-pathing as all I/O is serviced by the local controller. As a result, there is no multi-pathing policy to choose which removes another layer of complexity and potential point of failure.

However in the event of the local CVM being unavailable for any reason we need to service I/O for all the VMs on the node in the most efficient manner. Acropolis does this by redirecting I/O on a per vDisk level to a random remote stargate instance as shown below.

pervmpathfailover

Acropolis can do this because every vdisk is presented via iSCSI and is its own target/LUN which means it has its own TCP connection. What this means is a business critical application such as MS SQL / Exchange or Oracle with multiple vDisks will be serviced by multiple controllers concurrently.

As a result all VM I/O is load balanced across the entire Acropolis cluster which ensures no single CVM becomes a bottleneck and VMs enjoy excellent performance even in a failure or maintenance scenario.

As i’m sure you can now see, Acropolis provides excellent resiliency and performance even during maintenance or failure scenarios.

Related Posts:

1. Scaling Hyper-converged solutions – Compute only.

2. Advanced Storage Performance Monitoring with Nutanix

3. Nutanix – Improving Resiliency of Large Clusters with Erasure Coding (EC-X)

4. Nutanix – Erasure Coding (EC-X) Deep Dive

5. Acropolis: VM High Availability (HA)

6. Acropolis: Scalability

7. NOS & Hypervisor Upgrade Resiliency in PRISM

Data Centre Migration Strategies – Part 1 – Overview

After a recent twitter discussion, I felt a Data Centre migration strategies would be a good blog series to help people understand what the options are, along with the Pros and Cons of each strategy.

This guide is not intended to be a step by step on how to set-up each of these solutions, but a guide to assist you making the best decision for your environment when considering a data centre migration.

So what’s are some of the options when migrating virtual machines from one data centre to another?

1. Lift and Shift

Summary: Shut-down your environment and Physically relocate all the required equipment to the new location.

2. VMware Site Recovery Manager (SRM)

Summary: Using SRM with either Storage Replication Adapters (SRAs) or vSphere Replication (VR) to perform both test and planned migration/s between the data centres.

3. vSphere Metro Storage Cluster (vMSC)

Summary: Using an existing vMSC or by setting up a new vMSC for the migration, vMotion virtual machines between the sites.

4. Stretched vSphere Cluster / Storage vMotion

Summary: Present your storage at one or both sites to ESXi hosts at one or both sites and use vMotion and Storage vMotion to move workloads between sites.

5. Backup & Restore

Summary: Take a full backup of your virtual machines, transport the backup data to a new data centre (physically or by data replication) and restore the backup onto the new environment.

6. Vendor Specific Solutions

Summary: There are countless vendor specific solutions which range from Storage layer, to Application layer and everything in between.

7. Data Replication and re-register VMs into vCenter (or ESXi) inventory

Summary: The poor man’s SRM solution. Setup data replication at the storage layer and manually or via scripts re-register VMs into the inventory of vCenter or ESXi for sites with no vCenter.

Each of the above topics will be discussed in detail over the coming weeks so stay tuned, and if you work for a vendor with a specific solution you would like featured please leave a comment and I will get back to you.

Example Architectural Decision – Network Failover Detection Policy

Problem Statement

What is the most suitable network failover detection policy to be used on the vSwitch or dvSwitch NIC team/s in an environment which uses IP storage and has only 2 physical NICs per vSwitch or dvSwitch?

Assumptions

1. vSphere 5.0 or greater
2. Storage is presented to the ESXi hosts is NFS via Multi Switch Link Aggregation
3. A maximum of 2 physical NICs exist per dvSwitch
4. Physical Switches support “Link state tracking”

Motivation

1. Ensure a reliable network failover detection solution
2. Ensure Multi switch link aggregation can be used for IP storage

Architectural Decision

Enable “Link state tracking” on the physical switches and Use “Link Status”

Justification

1. To work properly, Beacon Probing requires at least 3 NICs for “triangulation”  otherwise a failed link cannot be determined.
2.“Link state tracking” can be enabled on the physical switch to report upstream network failures where an “edge” & “core” network topology is used, therefore preventing the link status from being OK when traffic cannot reach the destination due to an upstream failure
3. Beacon Probing and the “route based on IP hash” network load balancing option is not compatible which prevents a single VMKernel being able to use multiple interfaces for IP storage traffic

Implications

1. Link state tracking needs to be supported and enabled on the physical switches

Alternatives

1. Use “Beacon Probing”