Here's what I like most about Ubuntu Linux: the support we get from the factory, from the vendor.
I also find value in being able to test things with the community and learning more about the solutions that have been proven, so we can start to elaborate more and be able to pass to production more reliable information.
In the last five years, open source was not known, or it wasn't trusted much, but the industry knows that open source is the way to go as long as you work with the right vendor. Open source is here to stay, but it would always be dependent on the right partner, because there is a lot of open source software, but if they are not maintained, secured, or controlled, they are just like a train without a driver.
Based on all these, we love Linux, especially because of what we have learned in the last two months: We started to migrate customers from Microsoft SQL from Windows to Linux because the performance you can get from Linux with SQL from Microsoft, it's unparalleled.
You have more advantage from the power of Linux and you can cut off the cost of an operating system by using Linux with Microsoft SQL. We are starting to do that in the enterprise market because we believe that is a great step for them to reduce costs and to start making more powerful ETLs and queries, and faster processes, at a better price, because Linux is much more affordable.
Even in the Cloud, you'll find that a Windows instance is $400, while a Linux instance running SQL is $100 a month. It's a good thing here in Latin America, where we are going to start doing this. We are also seeing that there is a great opportunity in other countries in Europe.
We provide support for Red Hat, CentOS, and other distributions, but we have a solid relationship with Ubuntu Linux, with Canonical. We are a partner. With this solution, we are able to do more. We are able to explore a lot.
With Linux from Red Hat, we have been asked to manage, because in Peru, there were not many companies that had the knowledge to manage the workloads, but we prefer to use Canonical.
If a customer comes and asks us to manage their Red Hat Linux servers, we can definitely think about it because we have the skill. In our team, we have LPIC-1 engineers specialized in Linux so we can run any workloads over Linux.
Ubuntu Linux is very good.
There is a big advantage in security when using Ubuntu Linux that you will not have in the Windows environments short-term. When you have this relationship with Linux and you start working with very secure environments, there's less possibility of being directly attacked by a group of hackers. You will lose less data and you will have a more reliable ecosystem.
What I'd like to see included in the next release of Ubuntu Linux is for the interfaces to become richer, so they'll have the capability to absorb traditional and normal technologies.
For example, Canonical makes such high-end technologies to run and manage several servers at the same time, but they couldn't succeed because they were focusing all their efforts on just Linux-based systems.
If someone starts by providing things to assess and migrate the workloads you have in a data center where you have 200 to 300 Windows instances, you can provide some studies to these executive directors and say: "Today you're spending this money on licensing and operations, and you're getting 40% of your operational performance. If you start running Linux and you move your workloads into Linux, you'll be able to cut costs, and you'll get more out of your operational performance which you can present and provide data to your end customers more quickly and safely."
What we need to have is more tools to access the Windows environment of Ubuntu Linux, so we'll be able to say: "These are opportunities for your operational expenditure and cost cutting. These can help make your company better and allow you to provide more data to your customers more quickly."
Having more tools in the next release that can help provide information to executives: letting them know that there's money waiting in those opportunities for migration and change, is what I'd like to see.
I have more than 20 years experience in providing MSP services for enterprises and the government.
We've been exploring the market. We are located in Peru, so we designed our own technology and we've been exploring a number of technologies from several providers. What we did was to create technology locally and based that technology on the best practices of several brands.
For example, we talked with Silver Peak, we talked with Fortinet and Juniper and other providers, because the major problem in the industry was the pricing and the licensing models. What we did was to create our own technology in Peru, then we provide this technology as a service, as a managed service provider. That's what we've been doing.
We provide the consultancy, then we provide the hardware, then we manage, but they don't need to buy it. They just need to rent it for the period of time stated on their contract, then we provide the full managed services for that.
We started as a hardware appliance on-premises, but the time and the conditions of the market forced us to start preparing a virtual cloud appliance because as Amazon, IBM, and other companies were using the Cloud, we started to make these laboratories to enable our device to pass through traffic over the Cloud, on any cloud. We started to use a virtual appliance. We started to use the Cloud.
Now, we have a hybrid model where some of our main devices are located in the Cloud, but we have a satellite and it's called a hub. This hub is installed on the local data center and its availability is found on several other providers. In this way, we can start and we can continue to monitor everything without experiencing loss, because sometimes data centers have this downtime.
We can keep working with other devices that are connected. We made a load balancing with DNS. We have a DNS solution that provides this, so it also responds to unavailability. If we have a problem, what we do is we keep tracking, monitoring, and providing KPIs for customers, and if something happens, we can respond within 10 to 15 minutes.
Many companies have a monitoring system. They can use PRTG. They can use free open source devices, but they don't have awareness. They have the monitoring systems, but they don't have time to remain seated to watch all those KPIs and sensors. What we did was to create an escalating model where the most valuable information our customer needs is the availability of their core systems.
We always take great care and we provide notifications not only about the downtime. It's not about the values because there is a big difference between a DDoS attack and just another load of our applications. We know those patterns. We're usually notified about anomalous patterns, security, etc. Today, in both the government and private sectors, attackers are scanning all the time. As we have an IDS solution, we are able to detect some anomalous patterns on the main sites and on the application.
In applications, we have developed IM (identity management) solutions. This software also tracks all the users getting into an application. When we notice that there is anomalous pattern, we're notified, so we block because we are using the zero-trust concept.
The zero-trust concept is a concept that makes us more reliable, because if you are a collaborator, or an employee, and you have a computer, a tablet, and a mobile phone, and you have access to our applications, we will know that you are not connected to your device trying to get into an application. We will ask you if you are the person trying to get in and we will authorize and permit you to get into an application through this identity access management solution that provides you access to the applications, but at the same time, we are providing you access to parts of the application you have permission to access.
This is a great accomplishment in Peru because we created something very competitive, in terms of Okta or AWS Cognito: it's their standard solution. We created that here in Peru, so we are trying to push this technology outside to make the people know about it. It's a mix of things because if you try to make just a firewall and start checking just the IDS and IPS, and you don't start checking the application itself, plus you don't start checking other patterns, you will have less information. What we are trying to do is to be more holistic on how a person works in the company to protect both their information and their access to the applications.
It's very holistic. We are mixing bare metal security. We are using a WAF (web application firewall) that we made here. It's a universal thing. At the same time, we are using the identity management platform. We made it for protection at that level. We are making several layers for the security, and also to provide the whole holistic pattern to our customers.
This is why our customers stay longer with us. Each customer we have today has been with us for more than five years, and they renew their contracts with us because they feel so comfortable with us, and we are well-trusted.
Customers just need to ask us if they can do something specific, if they can explore, because what we usually do and build for them are laboratories. We are making proof of concepts of new products because customers want to move forward and try new products. This is what enables us to keep the customers and have them renew their contracts, so they can move forward with new products. This is a good thing for us, and we are able to retain customers who have been with us in the last 10 years, for example.
We didn't experience many issues with Linux because we started using it early. The first solution I provided was 15 years ago, when we worked with a telco and this telco wanted to go to the enterprise market to sell IT services. What I proposed was to build a Platform as a Service to protect their information real-time. It's called continuous data protection. We installed all those in Linux.
We have extensive experience doing that and we started with data continuity for data centers. We started to replicate data a lot, even for a core bank located here in Peru. They were one of our first big customers and we had a five-year contract with them.
We didn't find many challenges at the beginning with Linux, because we started to build software over there, but then when we started to manage very big logs, we decided to build another software, in another instance, to start distributing the data and have more information and visibility for our customers.
We also developed a software over Linux to compress the traffic in transit. We made a lot of those. We didn't face any challenges because we have been working a lot with Linux.
We learned a lot. We learned how to build software over Linux and in several languages because we needed to build interfaces for end users. We also needed to build the backend. Our backend technology today has not yet been used a lot. We are using HTML technologies for the frontend. We have a team to do that. We also managed several, general things for any of our distributions. It depends on the function we would like to add to these appliances.
For example, for a network appliance, we are making our own distribution with Ubuntu Linux, but with some customizations to make it work lighter and easier on our dashboards. It's networking.
When we want to use Ubuntu Linux to develop software, we prepare those devices to run those workloads and make good backend servers and frontend servers. For example, for the IM (identity management) solution we made, we built it over Ubuntu Linux, but we are not yet using it at the backend. We are using the standard Open IDZ, but we made our own version for it to run effectively and be able to integrate this granular part of the permissions, because that is a simple way to make an identity management solution.
When you need to connect applications and provide granular permissions to the applications based on profiles and start from the FAP: If you have a new employee and this employee needs to have a lifecycle, plus permissions to applications, there is another integration we make with our software. We've been working a lot on that part to create this fully integrated software for identity management and application permission management. It's very nice.
The advice I would give to others who are looking into implementing Ubuntu Linux is that they need to start working with a partner. They need to start working with laboratories and start assessing. They need to start assessing what the company pain is because a few years ago we had been invited to talk for a country who was submerged in several taxes by paying Microsoft and some other companies with proprietary software. We saw that the companies in this country were spending a lot of money on Active Directory, SQL servers, and other technologies that the customers have been using for so many years. When we showed them how they can transform this into an open source technology package of assorted tools, and that they were easy to manage and to learn, they started to study it.
If a company wants to move forward with the world of Linux for improvements, savings, and start operating differently, they need to start working with a peer who has then done this for years, to make it aspirational, who would be able to tell them that they were using this technology, that they migrated their infrastructure and their solutions to open source, to Linux, because they needed to do that, and they succeed. They should start from there. They would want to start from laboratories and start passing to production the things that they can manage. First, with a partner, then they can run a team made up of people who can manage this new technology.
It's the right path. It's what they need to look at. If Microsoft didn't want to put SQL or Linux, they would be closing too many doors that they have today. Microsoft has grown bigger because they are open to the world of Linux. Microsoft has been saying that they are using Linux on the network and on other parts. It's the future. For example, we are using MongoDB. It's a great document database and doesn't have anything to do with SQL servers. They have no relation.
Companies need to explore. They need to start exploring new things and make these laboratories. If they start making these laboratories, they'll have opportunities to save money and make their operational performance better. They'll have a great migration to a new set of technologies.
Ratings for Ubuntu Linux will depend on the country and the culture. There are some companies who want to invest on the subscriptions. At the beginning, it would be best to invest on a partner instead of the subscription, because you'll really be able to take advantage of a subscription when you have an understanding of Linux. If you don't understand Linux yet, you should invest more in a partner who really knows about it, who can start traveling with you in this journey of migration.
Once you have everything up and running, that's the time you can select which subscriptions you need, but if you have a good partner, your partner will be able to give support about the subscription. Ubuntu Linux is the only one that doesn't need a subscription to work in an LTS version, unlike Red Hat and other distributions that where a subscription is mandatory. For Ubuntu Linux, you just need to buy a subscription on an LTS version in Canonical when you really need it.
I prioritized having a good partner first, then we went for a subscription to provide compliance to my operating systems that needed that compliance grade. You don't need to waste time and money in a subscription if you have the right partner on your side. There are two ways to look at this: the subscription cost and your partner who can manage everything properly.
Ubuntu Linux is a great solution, so if I'll rate it from one to ten, with one being the worst and ten being the best, it's a ten for me.
