Today at the Nutanix 2016 .NEXT conference, we made several exciting announcements. You can read the announcements in this press release. This blog will provide you with more information on the announcements, with details on the new product capabilities and when they will be broadly available.
A Single Platform for All Workloads — Eliminating More Datacenter Silos
Nutanix Enterprise Cloud is the ideal platform for all virtualized workloads in the datacenter. Thousands of customers already run a wide range of applications in the Nutanix Enterprise Cloud, including Oracle, Microsoft SQL, Microsoft Exchange, and Citrix VDI.
We also launched Acropolis File Services, enabling customers to knock down another storage silo—stand-alone file servers and NAS appliances. We are making a couple of additional announcements today that will enable IT departments to eliminate all storage silos within their datacenters.
Acropolis Block Services (ABS) enables applications not running on Nutanix to use Acropolis Distributed Storage Fabric for back-end block storage. Whether applications are virtualized or running on bare metal servers, they can benefit from the simplicity and scalability that Nutanix offers.
Nutanix Acropolis can now expose its distributed storage fabric as block storage over the iSCSI protocol. All of the storage functionality that the platform offers—deduplication, compression, tiering, snapshots, clones, erasure coding, etc. – are available for practically all applications, including non-virtualized workloads.
Even in today’s world where the majority of applications are being virtualized, certain applications will continue to run in physical server environments. We have heard reasons ranging from the need to depreciate existing servers, to application licensing and support, to even organizational dynamics. Nutanix ABS is a simple and easy way for IT to extend the scalability and performance benefits of virtualization to these environments.
With this, high-performance virtualized workloads, virtual desktops, file servers, and data from non-virtualized applications can reside on the same Nutanix cluster and be managed centrally using Prism.
This capability is planned to be available in the 4.7 release.
Acropolis Container Services: While modern stateless applications can benefit from the flexibility and isolation that containers provide, a lot of enterprise applications that are stateful—where the application is tightly integrated with the data—are left behind. Acropolis Container Services (ACS) attempts to bring the simplicity of container technologies from Docker to enterprises at large.
The Acropolis Distributed Storage Fabric has been extended to provide persistent storage support for containers. The volume driver is written in Go, it runs as a Docker volume extension, and behaves as an intermediate container. Thus, the benefits of hyperconverged infrastructure is extended to containers by exposing storage directly, completely bypassing the hypervisor.
Included in the 4.7 release will be the following:
- DSF and Docker Machine that uses the native Docker API and tooling
- Easy Install and Support: The DSF Docker Volume Plug-in and the Docker Machine Driver work right out of the box and are fully supported by Nutanix.
ACS is scheduled to be available with the 4.7 release, and natively managing containers using Nutanix Prism is expected to be available in a subsequent release.
All Flash on All Platforms: Nutanix currently ships an all-flash platform (NX-9000) that is used by customers for high-performance database workloads. As part of a phased roll out starting in July 2016, flash-only configurations are expected to be available across the entire product portfolio of Nutanix at a modest price premium over existing hybrid platforms.
Hyperconvergence, as a form factor, is the ideal architecture for all-flash solutions. Existing all-flash arrays are connected to the servers through Fibre Channel switches, and storage is commonly exposed through two storage controllers. This architecture “just about works” today with all-flash arrays. With technologies, such as NVMe and 3D XPoint, the network and storage controllers will soon become bottlenecks restricting applications from fully utilizing the performance and low-latency benefits these technologies offer.
On hyperconverged infrastructure, flash lives on the server right next to compute. Additionally, with data locality, VM data is always kept on the same node that the VM lives on. This ensures that data is only a PCIe bus hop away from the application, as opposed to having to go through the network every time. For customers looking at an all-flash datacenter strategy, Nutanix now offers a comprehensive suite of solutions for all use cases.
Bridging the Operator—User Gap
The second set of announcements focuses on what we are working on to further simplify datacenter management. The fundamental goal is to provide IT with more tools to improve operational efficiency, and at the same time, provide end users with the independence that public cloud solutions provide.
Nutanix Self-Service: One of the core capabilities that developers and business users love in the public cloud is the ability to provision applications and virtual machines without the intervention of IT. Nutanix Self-Service is designed to bring this simplicity to enterprise clouds, so that users can deploy applications at any point in time based on policies set by IT administrators.
Admins can create a catalog of projects and assign users (through AD/LDAP integration) and resources (storage, compute, and network) to these projects. Line-of-business users/developers can then login with their credentials, and based on what they have access to, will see these projects and resources. This will radically simplify application development and delivery, as well as bring a lot of automation to the process. The screenshot below from the user view of the portal shows the VM images that the admin has given access to for a specific user. The user can pick one of these images and can deploy it anytime.
This capability is expected to be available in a subsequent release.
Just-in-Time Capacity Provisioning through Scenario Based Modeling: Prism-based capacity planning, which was introduced earlier this year, provides capacity runway projection and capacity optimization recommendations. This enables customers to understand when they will run out of capacity and what they can do to optimize their existing capacity.
The next addition to this capability is understanding if the system has enough resources to support new demand (applications or virtual machines). As an example, if the admin gets a new project to deploy 1,000 new VMs, 20,000 Microsoft Exchange mailboxes, or an Oracle application, the admin will be able to use the scenario-based modeling capability to predict if there is enough capacity to support this project.
Since the solution is custom-built for hyperconverged architecture, recommendations to add additional hardware are bite-sized and staggered so that customers can scale-out in smaller increments, as opposed to making large 3-5 year investments.
This capability is expected to be available in a subsequent release.
Network Visualization: In addition to managing virtual machines, compute, and storage, a future version of Prism will also include support for monitoring the network. Admins will be able to see how a virtual machine is connected to the host, top of rack switches, and other VMs that constitute the same application. The connections can be classified by host, by application, by user, by VM, and so on. The capability will visually highlight the health of the connection, the status of NICs/ports, and also VLAN misconfigurations.
LLDP/CDP will be used to discover and validate the network topology and SNMP to get the configuration information from the switches. This means network visualization will work with almost any Top of Rack switch.
Network Visualization is expected to be available in a subsequent release.
We are continuing to add more and more capabilities to Nutanix’s native hypervisor—AHV. With the upcoming releases, the following capabilities are expected to be available:
AHV Dynamic Scheduling: In addition to supporting intelligent initial placement of VMs based on ideal fit, AHV will also include Dynamic Scheduling that will move VMs after they are deployed. In addition to factoring in compute and memory resources that legacy virtualization solutions do, AHV will also factor in storage resources before deciding where to place the virtual machines.
Affinity and Anti-Affinity Rules: Admins will also be able to set policies to decide which hosts are ideal and non-ideal fits for specific VMs. This way, when AHV automatically moves VMs within the cluster, specific hosts can be favored or ignored based on application preferences.
Both of these capabilities are expected to be available in a subsequent release.
Nutanix continues to redefine the definition of multi-hypervisor support. In addition to Nutanix’s native hypervisor, other hypervisors, including VMware ESXi and Microsoft Hyper-V, are also supported. To add further choice and extensibility, we are also announcing the following:
VMware ESXi Management from Nutanix Prism: The one-click simplicity of Prism will be extended to ESXi as well, and customers will be able to perform common VM operations on ESXi VMs from Nutanix Prism. While this doesn’t obviate the need for vCenter, it eliminates the need to go back and forth for frequent VM CRUD operations. This capability is expected to be available in a subsequent release.
Microsoft CPS Standard on Nutanix: Nutanix and Microsoft have jointly engineered a hybrid cloud solution that brings together Microsoft’s datacenter technologies, including Hyper-V, System Center, and Windows Azure Pack, to Nutanix as a single integrated product that is factory configured based on best practices from both Microsoft and Nutanix. You can learn more about this product here. This product is expected to be available by July 2016.
OpenStack Drivers: For customers looking to deploy an OpenStack-based solution and also benefit from the simplified integrated experience that Nutanix provides, the OpenStack package for AHV is now available for download. This includes storage, virtualization, network, and image management drivers. You can read more about it here.
While there are a lot of other improvements to the product in the upcoming releases, this should give a good overview as to how Nutanix is continuing to out-innovate itself and delivering new capabilities at cloud speed. Over the next several weeks, we will be posting detailed blogs that focus on each of these topics.
We will leave you with one final closing image. The stat below shows how fast Nutanix customers have adopted the newer product capabilities within their own datacenters. 43% adoption for an enterprise product upgrade within 100 days is not very common. With non-disruptive one-click upgrades from Nutanix, innovation is only a click away. We are upgrading tens of thousands of Nutanix nodes that are deployed across the world with the latest release of the software. This is exactly the service public clouds provide – continuous innovation without disruption.
Take the Next Step!
Delivering continuous innovation is what the enterprise cloud is all about. From ABS, to containers, to all-flash systems, to self-service portal and operational simplification, to turnkey solutions for deploying private and hybrid cloud solutions, we invite you to start the conversation on how Nutanix enterprise clouds can work for your IT environment. Send us a note at firstname.lastname@example.org or follow us on Twitter @nutanix.