Nutanix Plexxi Hyperconverged Webinar OnDemand

Nutanix Plexxi Hyperconverged Webinar

CLICK IMAGE BELOW TO VIEW WEBINAR

nutanix-plexxi-hyperconverged-webinar

TRANSCRIPT:

Rachel:  Good afternoon everyone. Welcome to today’s webinar a Hyper Converged Data Center spotlighting Nutanix and Plexxi. My name is Rachel Cuomo and I am a marketing communications supervisor here at Continental Resources. Just a few housekeeping items before we get going. If everybody could put themselves on mute that would be great. We have speakers that are in different locations, so we can’t mute everybody. So if the participants could put themselves on mute that would be greatly appreciated. If you have any questions throughout the webinar please enter them into the chat window. We will address questions at the end of the webinar today. Now, I would like to introduce Rick Nagengast, Hyper Converge Practice Manager at Continental Resources. Rick, take it away.

Rick:  Thank you Rachel. My name is Rick Nagengast and I work here at Continental Resources responsible for our hyper converged practice. At Conres we believe that IT data centers can be radically simplified with the adoption of emerging technologies such as hyper convergence, software defined networking, and cloud technologies. We are having some problems here going to the slides. Great. For those of you that don’t know who Continental Resources are we are a 50 year old national solutions provider serving over 1500 customers primarily located from Boston down to DC as well as greater Chicago. At Conres our hyper converge and networking practices have extensive knowledge in all the HCI vendors and software defined networks including ACI and SX and Plexxi and that is why we are so excited to talk to you about these two offerings. I would like to take just a second and just bring you up to speed on what we see as the evolution of the industry. Back 25 years ago in the last 25 years we have enjoyed and you have purchased what we call the best of breed. So on the left side you bought the best of breed from servers, from storage providers, networking, disaster recovery software, back up etc. and around the 2010 timeframe with the advent of VMware and server virtualization we saw that server population collapse from 100 to 1 server or 10 but you still had to deal with all of the storage. The San storage and all the maintenance associated and in your network and all of the software. In 2015 we saw the advent of hyper converge where they collapsed the server into storage into one converged environment like Nutanix but you still had to deal with all the complexities of the network the core, the spine, and the leaf and all of the software associated with that. We are excited about what we are starting to see now. As we see the next leg of the stool come to be in and that is taking the hyper converged network of like Nutanix and combining that with the power of what Plexxi brings together collapsing traditional core spine and leaf into a software defined networking capability. We believe these are the emerging technologies that can radically simplify your data center. So when you combine these two what you do is you get the best of both worlds. Everything that Nutanix is going to kind of enlighten you with and then the speaker with Plexxi is going to do, you will have the combination that gives you one system, one network, one administrator, one interface, one solution. So with that I am excited and delighted to introduce our next speaker which is Caesar who is the solution specialist from Nutanix who will give you an update on their value process.

Caesar: Thanks Rick. Can you see my screen?

Rick: Yes we can.

Caesar: Great.

Rick: If everybody could put their phones on mute that would be appreciated.

Caesar: Thank you. My name is Caesar Munoz. I am a local account manager at Nutanix. Today we will be talking about the cloud era, hyper converged infrastructure, and the enterprise cloud as well. So the cloud era is well underway.  Businesses want IT capabilities that will allow them to test new business models and roll out new services quickly. Cloud services both public and private offer enterprises basically a new consumption model for IT resources. Business are able to buy and use, adjust to resources that they need and only when needed. They are able to spin up infrastructure and platform resources on demand with a single click to build, test, deploy all their apps. With the cloud your admins don’t have to spend time deploying, managing, and refreshing infrastructure hardware and software. That offers new capabilities and features that are pretty regular cadence. Now while public cloud is great for applications and workloads that have unpredictable variable demands. For example, if a business, if you launch a new online mobile app, let’s say like a new online shopping experience or a check scanning app for a bank the demand for the app or service is not well understood initially and in this case public cloud allows businesses to ramp up resources for the app depending on demand however as the demand for the app becomes more predictable it is more economical to deploy on prem. By not tying up vital infrastructure when needed business can save cause but for more predictable or stable workloads which is the majority of enterprise applications public cloud services can be considerably more expensive then on demand infrastructure. You know businesses want to balance owning and renting, choosing between private and public infrastructure depending on application workload characteristics. There are situations where it makes sense to rent and others where owning is more economical. So businesses want the public cloud for fractional consumption and pay as you grow economics, infrastructure and platform resources on demand, low operational overhead, continuous innovation of the platform and even in terms of security companies are so into trust public cloud service providers with managing that infrastructure security. They are able to focus on innovation and services without committing resources to maintain security with on prem infrastructure. But they also want a balance between owning and renting to lower total cost especially as workloads change right. They become more predictable. Strict data governance. To know where your data is and quick access to their data when needed. They also need tailored SOA’s for mission critical applications and flexibility in choice of platforms. Being able to pick the public or private infrastructure that bests meets their needs. In short they want the friction less agility, simplicity, and the fractional consumption of public cloud services while still having control over the performance, the location of the data and services, choice of platforms. This is where the enterprise cloud comes in. Now, the problem traditionally is that infrastructure was not designed for cloud environments. The enterprise data center needs to be replatformed to meet the needs of the enterprise cloud. So let’s take a look at how Nutanix solves this infrastructure challenge. Nutanix pioneered the concept of hyper converged infrastructure. Hyper convergence natively converges, computes, and storage into an X86 server with attached flash and hard disc drives that are deployed as a cluster. Customers can start with a small footprint, 3 nodes, and as little as 2 U of rack space and can add, compute, and storage capacity when needed. Hyper convergence radically simplifies how infrastructure is deployed and managed and reduces a footprint of IT infrastructure by up to 90%. But Nutanix hasn’t stopped there. The Nutanix solution natively integrates enterprise class virtualization right into the product. The hypervisor comes preinstalled on all Nutanix appliances and cluster and VM management is built into the integrated management platform but is also part of the standard product. So there is no additional components to install or manage. You get powerful virtualization features such as VM high availability, live migration, and one click high provisor upgrades come standard and can be fully managed using the same management interface. Admins can configure their Nutanix clusters in minutes and start deploying applications right away. So let’s take a look at what infrastructure management looks like today. With Legacy Infrastructure you have multiple management pains. The burden of making them all work together to provide end to end visibility and control it basically shifts to the admin. Nutanix prism gives administrators a simple and elegant way to manage virtual environments. Prism uses powerful data analytics and heuristics to simplify and streamline common workflows within a data center and it eliminates the need to have multiple piece meal management solutions. Prism gives your administrators a simple and elegant way to manage virtual environments. With Prism, admins can centrally manage your entire infrastructure stack all the way from physical clusters in storage up to virtual machines. Prism uses data analytics and machine learning heuristics to simplify and streamline common workflows within a data center eliminating the need to have desperate management solutions. So you are going to get rich insights about day to day operations in the data center. For example, customizable operations dashboards are going to give you an at a glance summary and application infrastructure state. Administrators can also make use of predictive analysis of capacity usage and trends based on application behavior. So Nutanix in a nutshell in this stance automation shouldn’t stop just providing scriptable interfaces for infrastructure. In an enterprise cloud the infrastructure has automation built in so you don’t need to script anything to get to the benefits. Nutanix has approached infrastructure offers real tangible material benefits to customers in terms of greater agility, faster time to value, and lower total cost of ownership. A study conducted by DC to quantify the benefits of Nutanix infrastructure found that customers get over 500% return on investment over a 5 year period and up to a 60% lower total cost of ownership with a 7.5 month payback period. Now the biggest benefits are in terms of lower operational expenses. We give you your weekends back. Nutanix systems reduced management time by over 70% and reduce unplanned downtime occurrences by 98%. So in summary the vision for the future is to have hybrid environments where the boundary between private and public disappears. You may have an application that has some parts components in the cloud. For example, deep storage and others on prem. You could also have situations where the cloud is used as a backup DR target while the production environment is on prem and finally you might have an application that can be on the public cloud early on in its life when demand is unpredictable but as demand becomes more stable it can be migrated back to the on premises environment. To meet all of these requirements you need a paradigm for the enterprise data center that can deliver on both sets of requirements and provide a seamless experience between on premises infrastructure and public cloud services and that is where Nutanix comes in. Now I would like to Erick Crowell from Plexxi.

Erick:  Thanks Caesar. Hello folks. My name is Erick Crowell. I am the HCI product champion from Plexxi’s proud management team. If you give me a second here I will share my slides with you. So as Caesar discussed with the advent of modern distributed scale applications the requirements of the enterprise data center and cloud have changed considerably. The older traditional methods of infrastructure deployment are very complex to configure and manage and frankly too rigid and inflexible to meet organizations goals for a cost effective growth and change. This picture shows a traditional data center architecture in which compute and storage are pulled separately and connected by a dedicated storage area network. Typically using IP or more traditional fiber channel. Separately a multi-tier enterprise network was also required to facilitate end user client connectivity. This tiered network topology worked for the client’s server error of computing but it doesn’t provide the ease of use and agility required by the modern applications and data centers required to support them. Nutanix has vastly simplified the computed storage infrastructure by converging these tiers into a single clustered system that can scale out those resources as customers’ needs grow. This represents a major shift in how organizations are deploying their data center infrastructure significantly reducing complexity and cost. Unfortunately the network has still remained a major challenge until now. Plexxi is the hyper converged network that completes division for the hyper converged data center and enterprise cloud. As you can see in this picture Plexxi introduces a flat single tiered networked apology that is very simple and inexpensive to deploy and scale. Gone is the multi-tiered network complexity and multi hop latency inherent in traditional use spine network architecture. Once configured Plexxi switches create a virtual mesh fabric controlled centrally by Plexxi software which allows administrators the agility to quickly deploy new applications with little to no network configuration. All along ensuring the infrastructure needs, SOA’s established for specific workloads. Plexxi’s tight integration with Nutanix’s prism also breaks the stove pipes between the data networking team and the rest of the IT. With Plexxi the network is part of the hyper converged solution. Here at Plexxi we know that choosing the right network to support your HCI investment can help administrators meet the operational goals of their organizations while avoiding the painful and expensive problems that arise when you deploy HCI against Legacy network architectures. We call these networks hyper converged networks and they have some pretty important attributes that help them stand apart from the phantom pipes with lots of buffers many of us are used to. First you want to choose an HCI solution that scales out just like Nutanix. You want to avoid cumbersome and expensive scale up architectures like 00:15:57 that can become complex and expensive to grow. Key here is the ability to grow your HCN fabric incrementally and nondisruptive as more resources are needed administrators can simply extend the fabric to add incremental port and fabric capacity. That is at the core of what we do here at Plexxi. Second, you have a solution that is API programmable and event driven rather than spending valuable time programming COI commands and orchestrating complex network protocols administrators need systems that respond automatically to the lifecycle events provisioning and allocating fabric resources automatically with new hypervisors, VM’s or workloads are added or removed. Gone are the days of cumbersome COI’s and slow to converge network protocols and the Plexxi fabric changes are detected and resources allocated in a fast predictable manner lowering the skills required and cost to operate the network. Next it is important to lower the number of devices required to transport the various kinds of traffic present in these HCI environments. Administrators needing HCI solutions that can securely converge all the network traffic into a single fabric where through software control administrators can segregate workload traffic, restrict interference, and insure optimize workload placement and performance all under a simple management umbrella. Finally, you need an HCN solution that provides value as you grow your HCN investment from a single rack deployment to multirack across the enterprise and into the cloud. A solution that offers ease of use and administration no matter what the size. So let’s take a look at an example that illustrates why the Plexxi HCN should be your network of choice for Nutanix. Here we see a traditional leaf spine network deployment supporting 2 racks of network, Nutanix equipment. The workload is storage traffic to pass from rack 1 on the left to rack 2 on the right and must first move through the spine ware at the top that exists to provide connectivity to each of these racks. Because the traffic needs to pass through this additional switching layer the traditional leaf spine architecture introduces additional latency into the traffic between these racks. These additional switches also require more ports, cables, and optic to construct. In addition configuring and operating this infrastructure requires and advanced understanding of network principles and protocols as well as a working knowledge of several proprietary command line interface commands. These usually perform by someone with a fair amount of training, a certification on networking, and often require working with each device separately to configure the end to end management infrastructure. The Plexxi hyper converge network offers a much simpler design that reduces both miss cost and complexity and this design the Plexxi scale of fabric provides a multiple direct paths between racks and Nutanix equipment. Reducing the work latency and boosting application performance and because there is no spine layer there is less equipment to buy. In addition through Plexxi’s centralized control design and tight integration with Nutanix prism and VMware sphere resources like hypervisors, VM’s, and storage workloads are automatically detected, provisioned, and isolated to ensure maximum performance. No advanced networking group knowledge is required and administrators can rely on powerful visualizations and provisioning tools to understand how the system is performing and quickly diagnose problems. Finally when the time comes to grow to a third rack or beyond the Plexxi HCN network will continue to offer a simple nondisruptive building block scalability verses the expense and down time associated with expanding the spine network when ports are exhausted. So Plexxi completes the hyper converge picture by bringing the industries only hyper converged network to the market. Plexxi is a network that provides the same level of management simplicity and building block scalability that Nutanix users expect from today’s modern infrastructures and through Plexxi’s tight integration with Nutanix and VMware, many of the everyday networking tasks are automated and administrators work with the unified solution that is managed through a single pane of glass V center or prism or both. Together Nutanix and Plexxi provide customers with a complete solution that includes converged, compute storage and networking. A solution that automates application awareness and securely isolates important workloads. A solution that allows enterprises to quickly assemble and deploy infrastructure on demand regardless of the scale. So at this point I would like to hand it over to Yen Wang a network security specialist from Nutanix. Yen is going to take us to a live demo of the Plexxi Nutanix solution in action.

Yen:  Ok. Thank you Erick. This is Yen. My name is Yen Wang. I work for Nutanix. I am the network and security specialist. So let me share my screen real quick. Ok, can you guys see my screen ok?

Rick:  Yes we can.

Yen:  Ok. So what I like to go through is really do a demonstration of Nutanix integration with Plexxi. What I am looking to do is give you a brief description of our integration, showed integration between Nutanix running ESX and VCenter, right, and the integration between VCenter and Plexxi and Nutanix. Several step we are going to do is we are going to go through the process of creating a VM, migrating VM and, deleting the VM and show you from that standpoint how does the integration work and what does each side see. Alright. So with that said, let’s take a look. This is a small cluster that we have in our hosted POC lab. This is running an enhanced cluster with 3 nodes of ESX. There are multiple VM’s running on it. This is a pretty typical environment right from any customer perspective running ESX. So this is what it looks like from VCenter perspective. From Nutanix perspective this is what our screen looks like. From our platform we support multiple hypervisor. ESX, hyper V, you know and our own hypervisor called Acropolis hypervisor. So you can see from here, you can see this is the ESX running 6.0 and it has 6 VM on it. Currently there is about 8 hosts on this environment. From Plexxi’s dashboard this is what Plexxi’s look like, we have 4 switches in the environment and the ESX nodes are actually connected to this specific one. So from here is you look at the V center screen you can see there are 3 hosts and they are named 10.21, 1.25, 227. So we go into the Plexxi interface, I can select these hosts and now we can go into details and we can actually see that these hosts are here and you can actually drill into the specific because of the kind of of integration Plexxi has with Nutanix and VMware you actually drill into the specific host and see what virtual machines are running on each of the hosts. Alright. So from here what we are going to do is we are going to show the activity monitor for Plexxi and what we are going to do is we are going to go into Vcenter and we are going to take one of these VM’s and we are going to clone it. This is a pretty common task in most of my customer’s environment. So let’s call this ESX task04 and we will say next, select the cluster, next and let’s put this on host number .26. Hit next and hit next. Alright so this machine is being cloned right now. Let’s give it a second for it to finish up. So this path is complete and you see the VM show up right here. Now if you notice on the background on the Plexxi interface on the activity menu all of a sudden there is an event that just happened. It says, create a virtual machine named ESX04. So instantaneously new VM coming up out into your environment is instantly recognized what’s in the network itself. So now we are going to take this VM and we are going to power it down. Alright. So once this VM powers out we will be able to search for it on the network side. From here you can see this ESX test 04 which tells you what host it is on and you can actually drill directly onto host and find out a little bit more information about this VM. Now this VM is running on host 1.26 but what if there is a Vmotion event on a network. What if we want to migrate this into another host? Let’s change this host from .26 to .25 and next, high priority, finish. So now it is migrating this VM, it says it is complete. So let’s give the Plexxi environment a second to recognize that. So now another event has just been triggered. Moving virtual machine named ESX04 after Vmotion. So it sees that. Now if we go search for this VM again and ESX04 is now on .25 verses .26 just a minute ago. Now what if you say we are trying to delete some VM’s on the environment. I know that doesn’t happen very often but we want to make sure that it is recognized also on the network. It is network aware of it. So let’s say we delete from the disc. Before we do that lets make sure that it shows that this VM actually existed before. So if we do ESX test, see ESX1 exists here and it shows that it exists on 1.25. So now we go back to our V center and we delete it. So now it is deleted from the ESX V center host. Now another event just got triggered instantaneously. The Plexxi fabric. It says deleting virtual machine in ESXtest01 if we do another search come up virtual machines. So as you can see ESXtest01 no longer exists on the network from Plexxi’s perspective. These type of tools are very very powerful in terms of what you can do in terms of integration. Things we can do such as we can extract the meta data from the hypervisor to the network knowing where VM’s are coming up, where does it go across the fabric and also what type of 00:27:24 they need to be associated to, such as Vlance or security policy and so on and so forth. The integration Nutanix has with Plexxi allows us to do things make things much easier to operationalize not only from a virtualization perspective but also from a network and security perspective. That VLan needs to only be configured on one port and it will actually go that VM no matter where it goes across the fabric. Thus eliminating a lot of needs for network team to get involved on every step of the way when we are doing some sort of Vmotion process or making sure those VM’s propagate across the entire data center. That concludes the demonstration for the Plexxi and Nutanix integration.

Rachel: Thank you Yen. Now we are going to do questions and answers and we have a few questions. So the first question we have is, what hypervisors does Nutanix support?

Yen: I can take that, this is Yen. So in terms of hypervisor support, Nutanix believe in offering our customers choices. So from a hypervisor support perspective we have absolute support for ESX all versions of it up to the latest 6.5. We also support Hyper-V Microsoft Hyper-V. We have our own hypervisor which is called acropolis hypervisor which is KDM based and recently we also added support for Zen Server.

Rachel: Which hypervisors does Plexxi support?

Rick: Thanks. I will take that one. Plexxi supports both VMware, ESX, and also the Nutanix AHD hypervisor which means that we directly plug into the VM ware V center UI and expose a number of networking functions there as well as respond as Yen showed you through the demo but also do the same for our Nutanix Prism interface as well on the AAHV hypervisor. We detect and isolate the storage traffic presented by the control VM’s in the Nutanix environment helping to boost the application performance.

Rachel: Ok. The next question is, if I build out my Nutanix environment using Plexxi how do I connect the Plexxi fabric to my existing enterprise network?

Rick: Thanks, great question. One way to think about your Nutanix deployment is that it is similar to how customers have implemented sans. The Sans is a special purpose infrastructure network that handles service storage workload traffic and is typically separate from the enterprise network for several reasons including consistent performance, workload isolation, and most importantly control of the environment. Plexxi enables you to very easily and quickly deploy your Nutanix solution using the HCN fabric and reserve a couple of 40 gig or 100 gig backbone links to connect back to the enterprise fabric. This approach allows you to deploy quickly and segregate your server storage traffic from your enterprise client server traffic continuing to not disruptively scale your Nutanix network using the same building blocks without having to involve the network team in every expansion.

Rachel: Ok. Another question we have is what type of switches does Plexxi sell?

Rick: I am sorry can you repeat the question one more time? My apologies.

Rachel: Yeah no problem. What type of switches does Plexxi sell?

Rick: Sure yeah. Thank you. Very easy one. Yeah this is a good question. We get this a lot because of our unique fabric design folks are often confused about what is at the core. Plexxi traditionally deploys a wide array of Ethernet switching platforms that deliver multiple ports of 10 or 40 gig Ethernet. More recently Plexxi introduced a new switch 3 series that supports 25, 50, and 100 gig Ethernet connectivity’s. Plexxi we are literally wired differently. That is what makes us different and how our workload of our software can leverage unique Ethernet based fabrics to deliver higher performance.

Rachel: Ok. Thank you. We had another question come through the chat window. How are the Plexxi switches patched together so that they result in the same if not lower latency then our traditional leaf spine architecture?

Rick: Sure. We have multiple paths between each of our switches that are connected directly together into a fabric mesh. There is no spine layers. So you never need to leave one rack go to a spine switch and pass through in some cases up to 3 different Ethernet switching chips and then come out to the second rack. You literally leave switch one and head directly to switch two which cuts at minimum one and up to three different switch hubs out of the equation which you know significantly lowers latency. Very simple.

Rachel: Ok we have a follow-up for that. Does the Plexxi fabric extend across multiple data centers and how is that done?

Rick: Definitely. Sure yes. It depends on the technology you want to use and what is between the data centers. Technically speaking Plexxi supports the world’s largest global SDN network directly connecting under ceilings switch to switch across continents but in a lot of enterprise data centers there is an existing network there that we need to travel through. So you can use either a technique of L2 tunneling, L3 tunneling or direct connections between Plexxi’s switches to achieve and unify different sites together.

Rachel: Ok great. Thank you. Our last question is, what is the minimal number of Nutanix needed to start off with in a cluster?

Caesar: I will take that. So you start with a minimum of 3 nodes in a cluster and we have different types of nodes for Nutanix. We have some storage heavy, some compute heavy, and you can mix and match within a cluster as well.

Rachel: Great. Thank you very much.

Rick: Back to Rick here and I just want to summarize by saying hey these are exciting times and if you are looking at hyper converge we would absolutely encourage you to look at Plexxi and get the benefit of these 2 compelling emerging technologies. So if I guess our call out is if you are evaluating hyper converge and you are assessing your network by all means call us. Conres specializes in this. We have been helping customers for 50 years kind of sort this all out. If you need us we were there to help you with the evaluation, the proof of concept, the procurement, and the deployment. So our only ask is hey if you are looking at it give us a call. Thank you so much.

Rachel: Thank you. That concludes today’s webinar.

 


Print Friendly and PDF
  • Share
  • FacebookTwitterGoogle+LinkedInPinterest

Request Demo - IT

Contact ConRes to request a product demonstration. Demos are available in the United States only. Please complete the fields below and a representative or product manager will be in touch within 24 hours.
  • This field is for validation purposes and should be left unchanged.

@ConRes
Follow Us →