How to deploy Veeam Cloud Connect

okay now that we're at the lightboard let's take a deeper look into exactly what Veen Cloud Connect provides our customers the first thing to remember is the best practice strategy for data protection - three to one rule that means three copies of your data production does count as one of those two different media types with one being off-site now when we look at the offering that Cloud Connect provides not only can we help you to achieve the one in the three to one being off-site but we can arguably also help achieve a different media type as well because many of the service providers offer tape out as a service now what does Cloud Connect actually do so if we look at this side of the lightboard with the customer environment depicted we've got a basic virtualized environment with production storage we've got a host and we also have a Veeam repository so if you have backup and replication deployed if you also have the agents deployed perhaps protecting physical workloads and you've got all your backups located on your on-site beam repository if you were to have another location add er site another data center with another disk based storage target that you could back up to you could simply create a backup copy job with or without way an acceleration to go to the remote location but what if you're looking for a remote location that you don't have to manage and you'd like to be able to scale simply by making a phone call that's where our veem cloud and service provider ecosystem comes in also referred to as V CSP so if you look on this side of the board similar environment but this is depicting the actual service providers data center now if we look at the customer location let's say that you've got a few backups that you want to send to cloud storage inside one of our service providers data centers now historically if you needed to backup to a remote location you'd have to worry about networking MPLS VPN tunnels whatever the case may be there's no networking complexities one of the really nice things around being Cloud Connect is we simply establish a connection with the service providers data center over a single port via SSL now once this connection is established the data flow will leave the on-site repository via this single port established the SSL connection with the vc SP now on the provider side you're going to have a component known as a cloud gateway I'm simply going to abbreviate that cg for cloud gateway and this is going to be a veem component that will accept the inbound traffic from the internet and that's going to be the ingress point into the providers data center now in the context of Cloud Connect backup once that data has arrived the data is simply going to be written to a subfolder on that particular service providers repository now this is going to be disk-based storage initially but then depending on the provider they do have the option to offer tape out as a service now all these backups will be separated like I mentioned in a subfolder so when they have more than one tenant each tenant will have their own subfolder on this repository now additionally like I mentioned they do have the option to do tape as a service that's an additional cost it would be additional option that could be a value to you as a vena customer now you do have the ability to encrypt your backup source side and some providers even have the added option of forcing encryption before they will accept inbound backups just for a security perspective now what if you're using an acceleration this is another feature that beam offers and on the provider side it can be an optional component that you can turn on it would come with an Associated cost now in this case from a data flow standpoint the data will still come in via the cloud gate weight but there'll be a new component inserted here and we'll call it the W a four-way an accelerator and the whole idea of the way an accelerator is to try to locate any duplicate blocks of data that may already live at the provider side in their way an accelerator cash pool and if there's any duplicates such as common OS blocks that will relay the signal back to your location as the customer and you simply won't have to transmit those duplicate blocks of data so when you look at this this can be quite a time-saver depending on what type of backup data you're sending up to the cloud and over to the provider side and if they do enable this on their side you only have to be on enterprise licensing not Enterprise Plus like you would have to be if you're using the Wayne accelerators anywhere else so the VCS pcommunity with Cloud Connect offers an advantage of Wayne acceleration only needing to be on enterprise licensing now when you look at dataflow that's it it's a very simple protocol you're gonna leave the repository here single SSL connection up to the cloud the cloud is going to go over to the cloud gateway so the traffic is flowing from your site to the internet arriving at the cloud gateway and then being channeled to a disk based repository inside that providers data center with or without Wayne accelerators being in place ok same thing over here if you're using Wayne acceleration you will also introduce an additional component we'll just put that here for waa when accelerator on your side that will do the very same thing so it's looking at digests on the source side target size looking at actual blocks of data to locate duplicates and that's the way the when accelerator concept works within veem we're trying to eliminate transferring duplicate data blocks over the WAM now the whole argument behind what this provides Veen customers i would like to call this backup storage as a service now many people may call this backup as a service but generally speaking backup as a service means the provider has some level of management and control over job schedules retention policies and so forth to get started with Cloud Connect backup the provider simply can carve you up a quota on their repository let's say you need 20 terabytes they carve up 20 terabytes once you make that connection of over the single port and establish that you're simply going to see another repository with whatever name they give it on their end and your overall capacity that's it from that point on the management lies over here with you as the customer with regards to what you send to the cloud what retention you put in the cloud are those backups encrypted or not so that's going to fall entirely on your shoulders now there are options depending on the provider where you can elevate your level of service where they will have more of a managerial presence in your infrastructure but to get started initially it's simply backup storage as a service now when you contrast that to Cloud Connect replication so let's put CCB which is what we just covered and then ccr4 Cloud Connect replication the story changes so if we're going to call Cloud Connect backup backup storage as a service or perhaps backup as a Service Cloud Connect replication is absolutely dr as a service now let's take a step back and remind you exactly what cook what replication means to veem right because there's several different definitions of replication when we look at replication we're talking about taking this VM over here on your side and creating a fully functioning copy over here at a target side now whether that's in a V CSP data center or your own dr site it doesn't matter at this point when we talk about regular replication within Veen the argument is we're not replicating your backups from repository a to repository B we're actually making fully-functioning copies ready to go if this whole data center were to go down or offline you can run business in this data center that's what veem replication does now when you look at Cloud Connect replication this offers a nice advantage over the typical replication to your dr site in the form of you don't have to manage the infrastructure you simply pay for the compute resources that you need and the provider will create what's called a hardware plan which will scope out exactly what you demand from a compute standpoint as well as storage and then that is presented to you as the customer as a cloud host target so when you're going through building a replication job after you've set this up on the provider side when you get to the target you're no longer actually browsing your own topology you're simply connecting to what they've already established for you so from a management standpoint that's entirely on the V TSP to make sure that that platform is operational and online and on your side you simply worry about scheduling the replication jobs based on the frequency that you would like those replicas updated right so our Pio recovery point objectives how much data loss could you afford should you have to failover to the secondary location now a little bit more technical just quickly the way the provider side handles this they are using VLANs on their side to ensure that no customer traffic data overlaps the next customer or tenant now these VLANs are non routed VLANs meaning they're configured on the v switches so the distributed virtual switches in a vSphere environment and the way that this works is they're not configured anywhere else outside of the DVS they're simply plugged in you don't have to make any changes on the physical hardware and this is the way that the provider keeps everything compartmentalized within their data center now when they first set this up they will provide a pool of VLANs based on your demand as a customer how many networks you have how many public facing static IPS you have they will configure that on their end now the one other variable behind this from a Veen component standpoint is known as the NEA now the NEA stands for network extension appliance and for lack of a better description it's a virtual router slash firewall the way that this operates is with a customer that just has a single network you'll have a single NEA on your side and you will have a matching NEA on the provider side now for every additional routable network on your side meaning the network actually has a functioning gateway you will increase the number of any A's you deploy source side so if you have four networks that actually have a gateway you will have for any A's source side while at the provider side the NEA count will never increase it's one NEA pertinent now the way that these NEA s operate is when it's time to do a full site failover let's say this particular data center is offline or has suffered a catastrophe natural disaster what have you when you do a failover this NEA handles the routing internally within that providers data center based on the way you've pre-configured it so if you do have the static IP need that I mentioned earlier for something like a web server or an email server that NEA already knows the routes and the rules because you've set that up ahead of time so when the failover occurs this NEA now becomes your own personal router based on the way that you've pre-designed it when you look at the partial site failover so to me this is a huge advantage with Cloud Connect replication because when you look at dr as a service so we have full and partial failover with full failover this is assuming the entire production data center is either offline has suffered a major outage and you need to failover everything that you've replicated ahead of time so that your business can continue within their providers data center with that there is a self-service portal that you can access as a customer so that you can trigger a predefined failover plan or multiple failover plans that you've built ahead of time so you control the order at which the VMS come online and that would be accessible to you via the public internet the partial is a little bit different the partial site failover assumes that part of your production infrastructure is still online and running perfectly fine but for whatever reason you need to failover a subset now potential use cases here could be you don't have the luxury of having a ton of extra resources and you need to do some hardware upgrades refreshes you need to take a cluster offline for management you know and patching or what have you you can shut down that subset of the environment initiate the partial site failover to those particular vm's that you've already replicated to the provider side and what happens that in EA's will build a layer tool and extension so the Nea will go out and this is using OpenVPN it will create a connection from the source side NEA cross the Internet to the target side NEA now this creates transparent networking for you as a customer so if you were hitting that web server on 10.0.0.0 is with regards to DNS so the resolution will be transparent to you as a customer because of that layer two extension now the NEA s are an optional component most of the service providers who don't already use with VMware vCloud director will opt to leverage the any ace now we do support vCloud director so we'll label that as V CD for short so if the service provider is using vCloud director for management the NEA s likely won't be needed because they'll be managing networking already using NSX or some other capability that they're familiar with so the NEA is an optional component to help get started with building that layer - extension so just a regroup on everything we've covered because this is certainly a very busy light board because there's a lot of things going on with Cloud Connect we've got two flavors of Cloud Connect you've got Cloud Connect backup and Cloud Connect replication backup enables our Veen customers to take backups from on-site repositories using a single ssl connection with a single port number going through the internet with no VPN tools require no MPLS networks needed and then establish a connection with one of our V CSP provider partners and land those backups on to their disk-based storage with the option of doing tape out so it's a very great option if you already have a relationship with a Veeam cloud and service provider partner or if you're looking to set up an active account with one of our VCS piece one of the key advantages that we need to remember with Cloud Connect replication which is the secondary option is when you look at replicating to a location that you do not own there's a certain level of permission and acts that you have to have to do host-based replication because we're talking to the hypervisor via api's we're doing snapshot calls we're writing data on to data stores as native files like VM decays or VHDs VHD X's so in order to replicate to a cloud-based environment currently the theme cloud and service provider community are the only cloud-based targets that you can truly replicate to using bean that provides two different options for failing over so that offers a key advantage if you're already looking at doing backups to the cloud but you could see yourself also needing dr as a service the VCS p can provide both of those with a single partnership so they can offer you dr as a service compute and storage resources as well as backup storage as a service or a cloud connect backup simply using whatever on-site disk-based storage they have an extra level of data protection using tape as a service and tape out so that concludes the tutorial around what Cloud Connect provides you as a customer so let's take a look inside the actual console at where these two features are configured thank you so much for watching you

Channel: Veeam

Views: 9,283

Keywords: Veeam, Veeam Software, Virtualization, Backup, Disaster Recovery, Availability, Recovery, Replication, Data Availability, Digital life, data protection, Backup and restore, Backup and recovery, Data recovery, Veeam Availability Suite, Veeam Availability Platform, business continuity, digital transformation, business resiliency, veeam cloud connect, DRaaS

Id: tLULNUPlFO0

Length: 17min 47sec (1067 seconds)

Published: Fri May 01 2020