Ipanema’s Grenot on WAN Optimisation, Governance, SLA, SaaS and the future of Cloud computing

ITProPortal recently sat down with cloud specialist Ipanema Technologies which is based in France, to talk about the concept of WAN optimization and what it takes to make sure that the Cloud doesn’t fall over. On that occasion, we interviewed Thierry Grenot, the EVP and co-founder of the company.

Can you briefly introduce Ipanema to our readers?

Ipanema provides technology that guarantees application performance. We deliver visibility into how applications are used and what for, over the corporate network (including services like Salesforce or Gmail). Companies can then prioritise business-critical applications, see where issues arise, and generally manage network activity in an automated manner. We call this entire process WAN Governance.

At the core of Ipanema's offering is the concept of WAN Optimization. What is it all about?

WAN Optimisation is a small but critical part of what we do. It deals with accelerating and compressing applications across the network in a transparent way and in real time. Think of applications as cars, and networks as roads. WAN Optimisation is about speeding up certain cars, or making some of them smaller so they travel faster.

Yet what we do extends beyond WAN Optimisation and into WAN Governance. This involves additional features such as quality of service and control. Rather than just speeding up traffic, companies can manage it proactively. They become traffic wardens, directing cars according to the needs of users and the businesses’ objectives.

What happens if you fail to hit a client's Guaranteed Application Performance target?

First, it’s important to note that we do everything we can to make sure no issues arise. Should something happen, we have several tools and methods to handle the situation promptly. Our process differs depending on whether our technology is used directly by the customer or not.

It can either be operated by (and be in the hands of) the customer; or it can be delivered as a service, meaning it’s being provided either by us or by a value added reseller, often a telecoms operator.

In the first instance, when the technology is being consumed as a product, the first thing we would do is make sure the customer has the right bandwidth in place, a process which we call “right sizing”.

Sometimes applications fail because bandwidth isn’t sufficient and an upgrade is required. This happens when the company’s network doesn’t have the right capacity, or if too many users have been added.

What we do is deliver reports explaining the state of the network to the customer, the steps that they need to take to make sure the applications work as expected, and so forth. We explain to the company that “if you do x, y, and z, then we can guarantee all your applications”. We always do this when we first start working with a company, but sometimes they add additional users/applications which change things. So it’s part of a continuous learning process.

We handle things differently when we’re working with telcos as VARs, which is where the “Ipanema as a Service” offer comes in. Here we’re enabling telcos to provide WAN Governance as a managed service. Companies may then look to their Service Level Agreements, as the telcos will have some guaranteed application contract that’s been signed with the customer.

So when this happens, there is a discussion. Let’s assume the application quality score (which is what we use to measure how applications are performing) isn’t the right one for a critical application. Then there’s a first level of analysis: is it because the network and the Ipanema system? Or is it because the traffic that the customer has put on the network? If it is due to very high traffic levels then the service provider will use a right-sizing report to explain things to customer.

If the performance isn’t obtained because of the network itself then usually it’s the responsibility of the service provider, and they’ll provide the compensation that’s been negotiated within the SLA. Inside the Ipanema system, we provide all the tools, not only to monitor the performance but also to allocate the responsibility to either the provider or the operator.

The difficulty usually with SLAs isn’t really in knowing that something hasn’t been reached. Rather, the difficulty lies in figuring out why. Why are there failures? Caused by what? Caused by whom? It’s very important that companies can provide this information in a report to understand who is responsible for the breach of performance.

While the bulk of your customers are from enterprise, what's your strategy for small businesses given how ubiquitous cloud computing is?

We now have a dedicated service, called AppsWork, that addresses small and medium businesses. We offer cloud based services to this market. SMBs are using cloud services everywhere. AppsWork exists to guarantee the performance of those cloud applications in a simpler, cost-effective way.

We’ve produced AppsWork in order to align our business model with the priorities of SMBs. We see that SMBs have the same needs as large enterprises, mainly because they’re using more and more cloud applications. At the same time, SMBs don’t have the resources or the budget of large enterprises, so they need something easy to use and cost efficient.

Where does Ipanema and its offering sit with regards to (a) Big Data and the BYOD paradigm (b) unified communications?

To put things simply and from a networking perspective, Big Data and BYOD (Bring Your Own Device) can be perilous for businesses. With BYOD, you have the common security concerns associated with a technical device: spyware, malware, viruses, and so forth. These are all amplified by the fact that BYOD involves foreign devices coming into the office in an uncontrollable way. Employees are bringing their own computers and tablets. Suddenly, system administrators have to factor variables like applications and antivirus software. With Big Data, you have a lot of traffic flowing suddenly over the network, often in peaks and troughs which can potentially impact the performance of other applications.

Companies must be prepared to handle these challenges before they rush after Big Data and BYOD. They must understand how employees are using their networks, and prioritise business-critical applications above others. That’s what we do. Our tools enable companies to slow down applications such as Facebook and YouTube in order to allocate more bandwidth and network space to business-critical operations, reducing IT risks. Companies can see what’s flowing over their networks. With Big Data, they can use our tools to make sure their networks are ready to handle file transfers – and that business-critical applications won’t crash.

One of the really pressing things for companies these days is also Unified Communications. More and more enterprises want to use UC, as do smaller companies. Operating with voice and video, UC isn’t forgiving to delays or jitters. These might occur when there are too many applications flowing across a network. In the same way our tools allow companies to manage BYOD and Big Data, we also help manage UC.

To this extent, we recently announced a partnership with Polycom, the global leader in open standards-based UC. This partnership was focused on making sure UC streams were understood and visible; that they were controlled effectively; and that they were routed along the appropriate network.