VMware VeloCloud SD-WAN Reviews

Our only use cases are large multinational enterprise customers, as well as very large service providers. The solutions get deployed to the public cloud or to the private cloud depending on the user's requirements.

There are 400 functionalities of the system. Depending on the user's requirements, we may find one feature more important than the other.

Sometimes the room for improvement is designing it for a different customer base that they're not currently in. We match the right solution to the right user's requirements. Some users need a multitenant offering, whereas most large enterprises don't need that. But if you are talking to a service provider, some don't offer a multitenant solution. And so when the customer needs that, we would say use one with a multitenant, like VMware.

We have been using this solution for about two or three years.

The ones we work with are all very good.

The scalability of the solution is very good. We have plans to increase usage.

Some clients want to go very slowly and test out just one country first. If you're going into a lot of countries, you sometimes have regulatory problems with devices being imported and finding the right people to handle that.

I can't really comment because I haven't used it that much.

The initial setup is always pretty complex with what we're involved in.

The deployment time really depends on how conservative a customer is and how quickly they want it to go. Some of them are three months and some are a year.

The amount of staff we need for the deployment depends on whether you're deploying to 50 sites or 2,500 sites. 

I would rate this solution a nine out of ten. In order for it to be a ten out of ten, there should be more automation and self-service features.

Our main use cases have been for customers who have been traditionally using MPLS lines and are now moving into a lot more SaaS-based applications to transform the WAN infrastructure. A lot of our customers have end-of-life, end-of-support devices on the WAN and they are always looking for doing RFPs on cutting edge technology. Some of them happen to use a lot of VMware and Cisco portfolios in their data centers and they are constantly trying to see what solution fits best for them.

VMware has been chosen by many of our customers who do not want to do a lot of complex routing in their environment and want a very easy-to-use solution. The most valuable features are simple troubleshooting, Dynamic Multipath Optimization (DMPO), and cybersecurity. 

DMPO is a technology that is similar to a proprietary technology that VMware has. Customers probably do not understand too much about it in detail but they know about the benefits of it in general. What it does is constantly monitors all different areas, such as latency, jitter bandwidth, artificial intelligence, and machine learning. It makes sure that if there is a case where there is a networking jitter contained in the link, it figures out how to maneuver your traffic elsewhere. This is all done automatically without customers getting slowed down on those poor links. The customers do not want to spend too much time researching these complex technologies which VMware automatically takes care of them. It benefits many IT teams by taking a lot of the burden away who are constantly juggling a hundred different other things as well.

The cybersecurity component is very important. VMware also has a lot of security components embedded into the SD-WAN, especially with the traditionally SD-WAN, and now a lot more integration through secure access with partnerships companies, such as Zscaler and other SaaS-based solutions. VMware has a lot of portfolios within themselves that can support many security use cases and if they cannot they have partnerships with industry leaders who can. For example, the Zero Trust Network Access (ZTNA) solutions through the Zscaler partnerships can help customers who want to move from traditional VPN solutions towards ZTNA based ones.

Overall as an experience, I have found that customers like the UI/UX experience that they receive from VMware.

There are customers that have very large routing and segmentation operations who do a lot of segmentation within their network and have complex routing requirements. VMware does not provide the facilities in terms of doing a lot of operations with routing tables, such as complex routing policies. It is more about that out-of-pocket experience that customers get out of VMware. There are other technologies for this type of use case from solutions, such as Versa or other WANs. These are for customers who are wanting to maneuver or configure things themselves, similar to DIY solutions, for them going to technologies, for instance from Versa, makes much more sense because they can do a lot more with routing, complex segmentation, and complex configurations. For customers who have complex service provider requirements, VMware and other solutions cannot support a lot of those use cases.

In an upcoming release, they should allow customers the flexibility to use mobile applications where they can go and check on the information about their networks. A lot of vendors, such as Meraki, have a lot more integration with the use of portals to a mobile application. Having this feature as an alternative to logging into a laptop would be beneficial. If you are a SaaS-based company why not make a mobile application as well. You might not be able to do configurations but at least monitor while away or on vacation.

I have been using this solution for approximately four years.

The software itself is very stable which our customers really like. They do not want to have buggy software or problems. I see the customers choose VMware over others and over time because of positive experiences with use cases, it becomes a trusted vendor and they tend to trust in the name as a brand in the future.

IT vendors are getting much more stable with the code, but nowadays everybody has more attention towards SaaS technology. Networking will just become another bundle to a lot of those providing this type of solution. Many vendors are getting into SaaS solutions and in terms of documentation, code releases, and features, there are probably more advancements. In SD-WAN specifically, a lot more features are moving into SaaS-based solutions. These new solutions are going to get reviewed and push the industries even further towards SaaS solutions it is the future.

The technical support is good.

The solution is straightforward to install compared to other solutions, VMware makes setup simple. SD-WAN vendors have a kind of simplicity embedded into them, just because of the nature of the software-defined components into it. 

I have evaluated Zscaler and Versa.

The advice to others wanting to implement this solution is not to try and do everything yourself. A lot of customers choose this route and are negatively affected two years later. It is important to do a lot of proof of concepts and testing. Get the solution from a trusted service provider who can manage it for you because these technologies are really complex under the code. Having a trusted service provider who can work with you and can have a lot of benefits. If your company business is not running the WAN links, I would suggest just offload that burden to a trusted service provider because then you can focus on your business. Technologies are complex, and doing all of these operations through your own IP node is complex. A trusted service provider who knows what they are doing can offload a lot of that burden allowing for you to relax and actually focus on your main applications and business.

A lot of customers I have seen in the industry try to do all the operations themselves just because they think that it is good to insource everything. This is one of the areas that you should not be insourceed a hundred percent of your team because these technologies can be extremely difficult to do on a day-to-day basis. Service providers handle hundreds of customers and they know a lot more about what you are doing and can be very useful than if you did it by yourself.

I rate VMware SD-WAN an eight out of ten.

a customer seeking to deploy an SD-WAN environment via 2 alternate stable methods either in a virtual environment or a physical device depending upon their site-specific requirements.

It is fairly similar to other solutions. It has the capability for testing link connectivity, which is its unique feature from a control perspective.

I would like to see an on-premise NGFW embedded into the solution. They have a SASE solution that is part of the deployment model now but is more cloud-focused. It would be nice to have an on-prem all-in-one SDWAN/NGFW box for some smaller deployments that don't need or have a virtual environment nor have a need for a full SASE deployment model.

I have been using this solution for a couple of years. We generally have the version that is previous to the latest one. Our clients have cloud and on-premises deployments.

Early on, there were some issues with it, but now, it is fairly stable.

It works well, but as it gets larger, it becomes more complex and difficult to address. It is good for mid-market type enterprise deployments but not for large deployments. Our clients are small, medium, and large businesses.

I haven't used their technical support.

One of our clients was using Fortinet.

It is fairly straightforward. After you've done it once or twice, it is fairly simple.

I would advise others to understand what you're using SD-WAN for. It is not for replacing MPLS. It is about application performance, access control, and manageability of your sites. Most of our clients don't even understand why they need SD-WAN or what they're looking at in SD-WAN. They think they can just replace MPLS, which is not really the case for SD-WAN. They do it because they think they are lowering costs, which in some cases is true, but it depends on the application performance criteria and things of that nature. 

I would rate this solution an eight out of ten. They can do some improvements. There are other solutions out there that I would highly recommend over this solution, but it is not bad.

The most valuable features are the orchestration layer, plug and play capabilities, and also the fact that you can either configure it as a full mesh or hard install. Even though fully meshed is limited, it is still a good feature.

There are a number of customers we have that require a fully mashed SD-WAN on VeloCloud that is not available currently. If you compare the reporting tool to all other tools that you use for customers for MPLS, they are not as granular. Additionally, the security could be better and increasing the capabilities for fully meshed architectural typology would be beneficial.

I have been using the solution for approximately two years.

This solution is scalable, most of our clients are enterprise size companies. 

The technical support is there if we need them. We are a premium partner to VMware, if there are issues that you cannot resolve internally, we reach out to them and they give us the help we need.

We currently use a few solutions other than this one such as Meraki. When picking the best solution for our clients, the best depends on the use case. It depends on the customer requirements, what they are looking for and then we will recommend the best fit based on our offering.

The installation is easy.

The price is reasonable for this solution, but the only challenge for us is that the Rand currency fluctuates to the dollar and provides some complexities. The price might go up just because the Rand has actually lost value.

This solution is at a very high level, and quite good. However, it is not one platform that fits all, it depends on the use case and what the customer looking for. You might find customers that want security or other UTM based features, which VeloCloud does not offer. It weighs heavily on the customer requirements, but if security is what they are after then I would not generally recommend the VeloCloud offering.

I rate VMware SD-WAN an eight out of ten.

The primary use cases of this solution are global WAN refreshment, Campus-to-Campus connectivity, and because my client is a global customer, they want zero-touch provisioning.

It's true, zero-touch provisioning. Anyone can migrate over to VeloCloud, it's that simple, just plug in the necessary cables and it's done.

Security needs improvement.

Also, it needs better scalability. There are certain products that if you need to create extra, the company expands and they will bring on extra branch-to-branch connectivity. There is a point where you will have to move to new hardware. We would need a larger scope for scalability.

I don't want to deploy a number of VeloClouds, and then in one year after business acquisitions, they need scalability. The only way to do that would be to remove the hardware and replace it with the new hardware.

I have been using VeloCloud, on and off for 18 months.

We are using the latest version, but not in a deep dive method.

It's stable, definitely.

It's a scalable solution. We have approximately 700 sites at the moment, and it's growing more as we adopt Microsoft Azure for this particular customer. So, we will need to spin up virtual instances of VeloCloud in Azure Regions.

The whole company is going through a restructuring where they're going to segregate portions of the business. They'll have their own VeloCloud SD-WAN overlay instance, integrating into the Cloud.

We have not contacted technical support. We have someone in our company that addresses anything obscure. He has a fair bit of experience that began prior to our engagement.

The initial setup is simple. SD-WAN is the simplest, and anyone can do it.

To deploy VeloCloud itself takes approximately one hour.

In IT, things don't always go to plan, but for the most part, it's an hour.

I used a consultant to help with deployment.

For the maintenance of each region, there will be a network SME to make sure that everything goes smoothly.

I will continue to use this solution, we have another customer that wants to use SD-WAN, but they are better suited for Cisco Meraki, so we are just doing a proof of concept with them.

I would definitely recommend this solution to others. 100%. That said, it needs better security. There is always room to improve.

I would rate this solution an eight out of ten.

We use it for connecting multiple sites through the internet.

Typically, our clients are medium-sized businesses.

VeloCloud developed DMPO to increase performance and connectivity.

VeloCloud provides very good performance, support, portal configuration, and service integration — I think that it's perfect.

VeloCloud could improve its integration capabilities with other solutions. VeloCloud just provides connectivity, but what about advanced security services or administration service providers, and HPP for voice solutions? I think that VeloCloud could improve this kind of service integration. I know it's not their main use case, but it's part of their portfolio.

In the next release, I would like to see better support for IPv6.

I also think that a CPE that supports WiFi 6 is necessary.

I am a reseller — I have been using this solution for three years.

The stability is good, but it could be better.

Scalability-wise, I think it's pretty good. It can support various numbers of CPEs and VMs for different uses in order to grow the capacity of the platform.

Our customers don't have any complaints surrounding technical support. They don't often need help as there is plenty of reliable documentation online surrounding VeloCloud. We have yet to come across a use case that VeloCloud doesn't have documentation for.

The initial setup is very simple. You can auto-connect the CPEs. Once you activate the CPEs, the CPEs can then connect to other CPEs that are in the network.

For reliable connectivity, I would definitely recommend VeloCloud; however, if my customers want another kind of service in addition to connectivity, then I would recommend another kind of solution.

On a scale from one to ten, I would give this solution a rating of nine.

There is a cloud gateway feature that centrally locates every space, which you don't get with Cisco. The cloud gateway is a very good feature for scaling purposes, etc. It's a very simple feature. In fact, everything about VeloCloud is quite simple, including the concept of the circuits, and, from a  technical perspective, it's not as challenging as Cisco VPNs. They have VPN 0, VPN 1, VPN 2, VPN 3, but they're very simple in that approach — the concept of circuits. Overall, I think both products are good. 

Cisco has begun to push its SD-WAN code within all of its ISR routers. This has made deployment quite challenging. With VeloCloud, deployment is much easier because they include all of their own hardware.

Cloud-based, it's okay because they roll up the device and provide you with a link for access. In regards to in-house, when you want to deploy the orchestrator, it becomes very difficult. Currently, I don't see any such good documentation compared to their competitors, like Cisco, etc.

Also, If you look at Cisco, just Google it and you'll get every detail: what to do, what specific system, what server, how much RAM, how much storage, all the details — it's just much easier.

If a customer has an optimization solution within their network, then you have to be very careful when designing — optimization and all. This can make your design very complex. If the customer has an existing optimization solution, then you have to be very careful when designing any part of the SD-WAN solution — Cisco or Velo.

They should provide us with the flexibility to scale up.

I have been using this solution for two and a half to three years.

VeloCloud is very stable — it's a very good solution.

I have not done a scale with something like 6,000 to 7,000 sites; however, if you look at the pure design, phase-wise, then you have to be very particular about new designs. Velo is purely based on design. I don't think the SD-WAN is that new and scaling that amount of sites requires a simple design rather than a complex solution.

With Cisco, we were working on a huge scale — I think it was around 1700 sites for that customer. After 900 sites, it started having problems. The controller was not sending the new site updates to the hub and it was not reflecting in the routing table; however, in the end, it scaled, but there were some challenges that we had to overcome.

There are not enough people within their support team. I would give them a rating of three out of five.

They have to improve, they have to be quicker and understand the types of problems customers face. 

The initial setup is straightforward. The initial setup is very simple compared to Cisco. Cisco claims to be Zero Touch Provisioning, but I think they have a lot of complexity surrounding that Zero Touch. With VeloCloud, you can generate and send emails, and the receiver just has to connect to the device, open up the email and the configuration improves. In this way, I think Velo is good with Zero Touch Provisioning.

Deployment time really depends on what you deploy. If you want to roll out a small site on a single device, then the designing and the policies are all done — it's a really quick job. Your circuit and site will rollout quickly, everything will be up and running roughly within two to two and a half hours. If you're looking at a complex site, then of course, the complexity increases.

VeloCloud is a good solution. They are only lacking in regards to the in-house version. 

On a scale from one to ten, I would give VeloCloud a rating of seven.

I work for a supply chain and we have a site where we want to offload the internet traffic and get to the application, etc. Some of the cloud applications go out directly and automatically through the SD-WAN device to the internet, so it is local and does not go to the MPLS. We can load balance those applications.

Their QOE, or Quality Of Experience, is the most valuable feature. Basically, VeloCloud measures the latency and the chatter and everything on a voice application, and it just routes the traffic or duplicates the packets according to that chatter. 

They have a lot of built-ins. There are so many applications defined inside the solution. It makes it so that we don't have to create some policies. Each and every application is already defined in it.

The solution doesn't offer failover between their own devices. If I want to, I should be able to put the two devices together, so there can be a failover. 

They need to come out with a new device, such as a 510 LTE which would have LTE capabilities.

On the 610 devices, they're saying they're going to have LTE in it. They don't have it. They don't even have a POE feature inside the VeloCloud devices. Therefore, if I want to set up a POE, just the one device with everything in it, I actually don't have a POE support on it.

I've been using the solution for about six months now.

Stability-wise the solution is quite good. It's reliable. It doesn't crash or freeze. It's not buggy at all.

The scalability is giving us problems at the moment. We want to put this as, for example, a primary MPLS, with internet secondary, and cellular tertiary. The LTE device doesn't even failover between them, however. Therefore, there is no scalability there.

We don't deal with technical support, due to the fact that we have AT&T. Since we deal with AT&T, we don't deal with them directly. Therefore, I can't speak to their level of knowledge or responsiveness.

Our company is actually many companies. At another subsidiary, they are doing a POC with Cisco. Cisco is providing them with better pricing. On top of that, they have good availability and failover capabilities on LTE. The Cisco router can do the secondary subnet on the same VLAN, which VeloCloud cannot do.

They have a lot of features we technically need to have but don't on VeloCloud.

The initial setup isn't too complex. It's pretty straightforward.

In terms of the deployment, we designed everything properly, and due to the fact that we took the time to design it correctly, it didn't take a long time. Strategy-wise, we have AT&T as our provider. That said, we have set up a reference architecture in such a way that it just copies every site, according to our site size and the requirements.

If you compare VeloCloud to, for example, Cisco, the pricing is almost the same. However, Cisco offers more features that are integral to the way we need the solution to operate. In that sense, it's better value for money. For example, now we need to have an extra LTE modem outside of VeloCloud, and it's going to add to the cost, which ultimately makes it more expensive.

We're just customers. We don't have a business relationship with the company.

We are not using the newest version of the solution. They already have a newer version, however, we have everything outsourced to AT&T, and they have not upgraded the new version on it yet. The version we have on the VeloCloud is 3.4.3, and the newest version, which, from my understanding, is four, and is already out there.

The greatest advice I have for potential new users is this: when you do deploy the VeloCloud, it's pretty straightforward. However, the only thing you have to remember is to make sure that you have the site requirements in terms of the primary and secondary circuits. For us, the last mile circuit is always a requirement, so we have a tertiary as a cellular. That way, even if the primary and secondary go down, you are okay. Also, you need to make sure to size your VeloCloud equipment or VeloCloud model according to the size of your needs.

Overall, on a scale from one to ten, I'd rate the solution at a nine.