Steelhead Reviews

The solution is being used for optimization. 

The connectivity to speed is the valuable feature. 

The solution needs to have alert notifications. 

I have been using Steelhead for the last ten years. 

The solution is highly stable so it's a ten out of ten.

The initial setup is easy. The deployment takes eight hours time. 

The pricing is reasonable. 

Overall, I would rate Steelhead a nine out of ten. 

The primary use case of Steelhead is as a CDR application and a second use case is for optimization. We are global distributors for Riverbed and I'm a technical engineer.

Scalable data referencing is the most valuable feature developed by Riverbed. It's also great that you can optimize multiple applications and get better reporting. These days, everybody is looking for visibility and Steelhead enables you to see what's running on your network. You can easily filter the site, the name of an application, the port, source, or destination IP.

The application response time of the solution can be improved. 

I've been using this solution for 13 years. 

The stability is good overall. 

The solution is scalable, I've deployed it for more than 20,000 users.

We've had a good experience with technical support. They have a toll-free number and are happy to carry out a remote session to troubleshoot any issues.

Positive

The initial setup is very easy. 

Licensing costs are reasonable. 

We've compared with other vendors but none have the scalable data referencing (SDR) feature, the mobile client software to optimize for roaming users.

I rate the solution nine out of 10. 

We did a data center migration for one client between two countries in Latin America. The challenge for this company was that they needed to send a large amount of traffic from Argentina to Mexico in a very short time. Without Riverbed WAN Optimization, it took one hour to send one gig of traffic, but with Riverbed, one gig of traffic was sent in five minutes. From one hour to five minutes, it had a big impact.

A feature we like is data streamlining. Riverbed does packet deduplication, packet compression and uses something like simple data referencing to compress the data. If one gig of traffic is sent to one link, that gig of traffic coming from the packet is only 100 mbps, so the compression of Riverbed is very powerful.

Another is transport streamlining, which includes TCP optimization. The window size of the TCP protocol is achieved in order to transfer more data in the TCP section despite the endpoint. The endpoint transfer was very slow, but it doesn't matter because with Riverbed in the middle, it can handle more quantities of traffic.

For application streamlining, where the HTTP application previously had a lot of traffic, Riverbed recognizes HTTP protocols and it can optimize the process because it can create the HTTP server. Normally, the browser and web server agree and compress the traffic, but if Riverbed is in between, it strips the compression from the browser and Riverbed does its own compressions, which is more powerful, and hence, the application runs faster for the user.

Also, for HTTP, Riverbed does caching so it can store data from all the user traffic. It keeps track of the HTTP data locally. Traveling into the server isn't necessary. They can retrieve the information locally because there is a local Riverbed.

It's for a very specific use case. It's not for everything. Although the solution is very powerful, it doesn't apply to all enterprises.

There is only one feature that needs to be improved. In the case of a High Availability event, if we lose the primary Steelhead, all the connections need to be restarted. If we lose the primary Steelhead that handles the optimization, the secondary Steelhead cannot handle the same High Availability. Some applications will restart automatically, but for some other applications, like STP, the users need to start a new session. If we load a primary firewall, the secondary firewall usually handles all the active connections, but in Riverbed, this isn't the case. We lose all the active connections at the moment of failure.

I have been using Steelhead for five years. 

I would rate Steelhead a nine out of ten because the product is very stable; it never fails. It does compression and application optimization very well. 

Riverbed can be deployed on-premise or in the cloud as well. We can deploy it in Amazon, Azure, and Google Cloud Platform. The solution is in the marketplace of every car provider and it also optimizes Office 365 traffic, since there's a partnership with Riverbed and Microsoft. We can deploy on-premise, we can deploy on IaaS or we can deploy on PaaS.

In the past, the internet links of the MPLS link were full capacity (one or two Mbps) and it was very expensive. Nowadays, the prices of the MPLS links are cheaper and the capacity is increased.

You should buy SteelHead for WAN optimization because it delivers more. WAN optimization is the process of increasing data throughput and improving the network and application performance over the WAN. This is what Riverbed SteelHead helps with. If you face a latency issue or slow application performance, go for this solution.

The first feature that is valuable is SDR, Scalable Data Referencing. Suppose you have two offices, one based in India and the other one is in the U.S., which is where the data center is. If an engineer is working from India and accessing an application from there, downloading or uploading files or folders over the file server or other applications, SDR will cache the data on the SteelHead device itself. The second time, he won't need to go via the WAN. He will be able to download the file locally. He will feel he is getting LAN-like speed over the WAN.

The second valuable feature is the compression.

The third valuable feature is the TCP optimization. Every time protocols are communicating with each other, every time they establish a connection, three is a three-way handshake. For UDP there is no handshaking because it is connection-less. But TCP does the three-way handshake before delivering any information to the application. The optimization feature caches a particular TCP connection and the next time a user uses that connection he will reach the destination easily.

Application response time and network performance could be improved.

Allot, an Israel-based organization, is offering DDoS attack prevention, not mitigation but prevention; threat response. If Riverbed would offer DDoS or IPS-type stuff, that would be great. Because we are deploying Riverbed in inline mode, we can easily take advantage of these kinds of features.

I have been working with Riverbed SteelHead for more than 15 years. I have a lot of experience with it. I have implemented it many times, both on-prem and in the cloud, around the world.

It is very stable. If you check the industry reviews, you will see that it is number-one for WAN optimization.

It is easily scalable. I have implemented it in SMBs and in enterprise-level companies.

Overall, Riverbed provides satisfactory support.

The initial setup is very straightforward. It is not complex. It is a very simple device. It is like an L2 device, and you can easily install it into a network. It can be deployed in inline mode, but there are some methods for deploying this appliance into a network. The first is the physical, meaning inline mode. The second is virtual. And the third is the server-side, where you can deploy this device with the switch nearest to the server, and the client-side is deployed in in-line mode.

Every SteelHead deployment always involves deploying two appliances. One appliance on its own will not work because SteelHead always works in pairs. If your organization has five branch offices, and one is the data center, you will need to deploy a SteelHead in each branch office to get the benefit of WAN optimization. 

The number of people needed for deployment depends on how far apart the locations are. If the locations are not very far apart, one person can fully deploy the product. If there are distant remote locations, you will need a person in each location.

The deployment time is three to four hours, maximum. It's very easy to install. You only need to run some commands, then go to the UI and do some configuration.

I would 100 percent advise others to implement this product. When potential customers ask us why they should buy SteelHead instead of just increasing their bandwidth from 10 Mbps to 20 Mbps, we tell them that doing so does not eliminate protocol chattiness. The reason behind this is that the TCP three-way handshake will happen at 10 Mbps and at 20 Mbps as well. If you have 20 applications in your environment and your users are accessing those applications, there is a lot of protocol chattiness going on. Upgrading bandwidth will not eliminate it. If you buy SteelHead, you can eliminate the protocol chattiness, and all the different kinds of handshaking. 

Every protocol is very chatty. CIFS, Word, Excel, PowerPoint, applications that are running on port 445, use a protocol that is very chatty as well.

Also, Riverbed recently started offering SD-WAN. The SteelConnect appliance delivers SD-WAN service and can easily provide routing, application visibility, security services, and WAN optimization. Riverbed has also started delivering a next-generation firewall which is completely an application firewall.

First of all we are great fan of Riverbed WAN optimization feature which I would say is the most valuable for us as we have couple of low bandwidth scenarios. Apart from this we are using specific Office 365 optimization since we are using Exchange online. We also use EX functionality of the box where we host our VM servers.

Many of our offshore installation are on low bandwidth but this has really helped to optimize the network which basically means we are able to achieve faster data connection or link.

There are issues with the tech support and scalability.

We have had no issues with the deployment.

We have had no issues with the stability.

The device mostly comes with a standard configuration with no options to upgrade the hardware. This means that we need to go ahead with a new or high end model in case the infra changes happen.

Riverbed needs to really work on local support when it comes to technical knowledge transfer. They need to have teams who can give proper knowledge transfer to customers so that they can use the device in best ways

Riverbed is great product and base functionality of these devices which is WAN optimization is handled by them in best and intelligent way.

The data reduction was fantastic, saved the company I was working for at the time a lot of time and money in upgrading circuits.

Management of the device is simple, but has all the technical aspects of a powerful appliance.

Enabled the company I was working for then the opportunity to centralise file storage, which reduced the cost of Windows server licenses by a massive amount. It also meant that collaboration across branch offices was vastly improved, as it was possible for documents to be stored centrally and open the documents as fast as if they were local... even very large technical drawings.

The version of RIOS that I used was very good, and I can't think of anything that needed major improvement at the time. We were using most of the functionality of the appliances, including Citrix optimizations, and it was all very good.

It was a few years ago that I was using it, but I was using it for about 5 years

None, the performance increase was straight away, the second and subsequent time of opening any document was markedly faster than the first, once the users understood this, it wasn't a problem. Also we learnt to 'pre-warm' the box before deployment, meaning that we could get round the 'first opening delay' somewhat.

These boxes were always up, very stable, and even upgrades were smooth. I can't think of an instance when the appliance wasn't doing it's job as expected.

None, the licenses were easy to deploy to upgrade and new appliances were just a drop in replacement. We scaled this up from an initial PoC to a fully fledged production appliance without any issues.

We were with Teneo, and their customer service department was very good, although we dealt mainly with the sales and technical teams.

Teneos knowledge in house was very good, both the sales force and technical support were brilliant, and when they couldn't deal with it they didn't have a problem referring it direct to Riverbed.

No, we went from bare WAN connections to Steelheads, we were running out of capacity on the WAN connections, users were complaining and we needed to centralise their data, we needed to either upgrade or optimize.

The initial setup was very straight forward, we followed the initial advice of Teneo to start with and then went back and optimized to our own policy over time, as we found better ways of optimizing certain traffic.

Through a vendor initially (Teneo), and then continued deployment in house once we were familiar. Teneos knowledge was excellent and they were extremely helpful.

It was very good, we ended up getting about 3x the average bandwidth increase (up to 10x at peak), and to replace with the equivalent IPClear circuits at the time would have meant increasing a majority of lines from 2Mb->5-10Mb. Back when we implemented it this would have run into the realms of about £50,000 plus running costs annually, so over the first year we easily saved that money.

Don't be scared of the pricing, they seem expensive, but in reality the performance of these appliances is astounding. Even a circuit of the equivalent speed would not be as fast due to Riverbeds ability to optimize TCP windows to get 90%+ utilization of a circuit. Licensing was a little confusing when we purchased it, so take advise of what you really need before jumping in and purchasing.

We did look at Cisco WAAS at the time, but it was very early technology, and I didn't have confidence in it. We also looked at BlueCoat, but it didn't offer the full acceleration of all the applications that we were after.

Riverbed Steelhead is a brilliant concept, and well executed appliance. It really does do it's job very well and worth the investment.

I've found the appliance to be the most valuable. Also, it's easy to use and configure.

It's helped to reduce WAN traffic by approximately 60%.

I would like to see more IP services, perhaps aligned better Netflow stats. The Central Management Console (CMC) newly named SteelCentral Controller (SCC) has rudimentary netflow capabilities, but it doesn’t have particularly in-depth data. I realize the need for a dedicated NetFlow environment but because the SteelHead technology deal with flow and not IP this is not capable.

We've had no issues with deployment.

We've had no issues with stability.

We've had no issues with scalability.

The SteelHead product is very good, easy to configure, however a home page which can be altered would enhance the gui 

• Easy deployment
• Fantastic bandwidth optimization
• Low maintenance requirements

It's improved the stability and throughput of the WAN, especially for remote small branches with low link speed. Reduced impact to network performance from Windows updates deployment on users PCs.

Better logic of the web GUI.

I've been using it for three years.

We have had no issues with the deployment.

We had some stability problems with a few devices. Probably RAM leaking issue by some processes which caused problem with remote access to the devices trough SSH or web.

We have had no issues scaling it for our needs.

I haven't had much experience with Riverbed Technical Support since our devices are managed by outsourcing company.

We previously used Cisco WAAS solution.

The deployment of the Riverbed products is very easy and straightforward.

I implemented it myself. The advice is just to follow the vendor instructions.

Check your network requirements before deploying one of the Riverbed devices. CX255M is more than enough for most of the branch sites with 10-50 people.