← AWS Podcast![#691: [MIGRATION SPECIAL SERIES] AJE Group's Cloud Transformation Journey to AWS with CloudHesive — AWS Podcast cover](http://d3gih7jbfe3jlq.cloudfront.net/AWS-Podcast-Title-Art.jpg)
AWS Podcast
#691: [MIGRATION SPECIAL SERIES] AJE Group's Cloud Transformation Journey to AWS with CloudHesive
00:28:06
In this episode, we explore AJE Group's journey to the cloud, discussing the business challenges tha
Loading summary
Transcript27 lines
- [00:00]
Daniel Flores
This is episode 691 of the AWS.
- [00:03]
Sandra Woolcott
Podcast, released on October 24, 2024.
- [00:08]
Simon Lish
Hello everyone and welcome back to AWS Podcast. Simon Lish here with you of course, giving you a bit of an intro to a special series we have running, which is exploring real world experiences of businesses that have successfully migrated to Amazon Web Services with a partner. In each episode, we sit with customers and their AWS partners that have helped them transform the infrastructure. And our guests will share candid insights into the migration journeys, talking about the challenges they had, strategies they used, the benefits they got, from technical hurdles to organizational change management. This is the nitty gritty details of what can make or break a cloud migration. Now we have customers and partners all around the world. And so you'll be hearing from different hosts and different customers and getting a different view of what people are doing globally. Hopefully there's something in it for you. Hope you enjoy the series.
- [00:54]
Daniel Flores
Hello and welcome to the AWS Partner Migration Podcast channel. My name is Daniel Flores. I am a migration partner solutions architect based in Bogota, Colombia. And in today's episode we're going to take a look at Grupo AJE migration project. And we are joined by two guests. Sandra Woolcott, CIO of AJE Group and Federico El Campo, who is team leader of the Solutions Architect department in Cloudesif. Now we'd like our guests to introduce themselves and tell us a little bit about their companies. So Sandra, if you can introduce yourself, please.
- [01:28]
Sandra Woolcott
Thank you. Hi, Daniel Fernando. Thank you for the invitation. I am Sandra Woolcott, CIO and Group's global manager of digital transformation at AAHE. I've been working in IT industry for the last 30 years, 20 of them leading business intelligence teams and over a decade implementing digital transformation projects in large corporations. I've been over five years in AJE Group, a Peruvian owned multinational dedicated to food and beverage manufacturing. To be more precise, it is the fourth largest bottling company worldwide with operations in four continents. The company was founded in 1988 in a remote province of Peru during one of the most difficult moments in country history. Quoting the founder's words, opportunity is found in the face of adversity and thus aroused the company to democratize the consumption of drinks with a focus in the most vulnerable populations. It was born and grew up empirically. It currently employs 9,000 people and continues to expand operations and reinventing itself, seeking to create new opportunities for all. It is a case of success and pride for Peruvians.
- [02:51]
Daniel Flores
Okay, and now Federico, can you introduce yourself please?
- [02:54]
Federico Del Campo
Of course. Hello everybody and thank you. Daniel for inviting me to share our experience about the AJE Immigration project. As you said, my name is Federico Del Campo, I'm from Argentina and with over two decades of experience in the IT industry. My journey has been a rich tapestry of innovation and transformation. I spent many years working with on premise infrastructure before delving into the realm of Virtualized infrastructure with pimwarp and over the past five years I further specialized in cloud infrastructure focusing on aws. Currently I serve as a Team Leader of Architecture at cloudhesive and AWS Partner Premier with operations across the Americas and office in the United States, Argentina, Chile and Colombia. We have a far reaching influence in the cloud services arena. From my position I lead a highly skilled infrastructure architecture team. We handle a broad spectrum of tasks including the design and deployment of cloud architectures, automation and of course migrations. That's it.
- [04:13]
Daniel Flores
Thank you Federico. Now I would like to set like a context for our listeners. So Sandra, could you tell us a little bit about your technological stack before thinking about moving to aws?
- [04:26]
Sandra Woolcott
Sure. Our core business is supported by an ERP custom made using an Erelli platform called Mike that has been around for about 20 years and obviously has its limitations. When we started the migration the application was primarily running on Linux and Microsoft with SQL 2003 as a database and of course we have been in a continuous migration process ever since until we reach a much more up to date version. In addition to that, we currently have a number of peripheral systems with interfaces to this ERP in magic. As one of the stages of this migration, part of that system used to run in a data center that somehow works as a private cloud, but in reality it was an on premise hosting located in Dallas and managed by IBM, while another part was hosted in the AWS cloud. To manage this mixed architecture we use a VMware VCenter hypervisor and approximately 20 terminal servers and MySQL databases. In addition to these services and applications, we had our business intelligence solutions outside of AWS, using Tableau for data visualization and Active Directory for identity management and access. It's important to point out that at that time all our applications and services were deployed, considering only two main instances, development and production. It's clear that our technology was considerably behind the current standards given that our tools and solutions were exceeding 10 years of age when we started the migration accompanied by cloudlessive.
- [06:34]
Daniel Flores
Okay, and now can you tell us a little bit what were the drivers that took a group to think about moving to AWS completely?
- [06:44]
Sandra Woolcott
Yes, we had several reasons. First, we have a small Corporate team. So I prefer to focus on the critical business processes, focus all of their attention on strengthening our value proposition by implementing technological solutions and to delegate all all the operational support and infrastructure management to third party experts. On the other hand, thinking of the value we provide for the business, it is relevant to consider that the sale of drinks is very seasonal, it is very high, peaks in the summer or special events and then shrinks which leads us to seek the greatest possible flexibility to expand or reduce our infrastructure quickly and at a reasonable cost. In addition, the organization continues to grow, creating new product categories and expanding into various markets. We demand a higher capacity of data storage, speed of processing on demand scalability, leverage the strategy of digital transformation of the company and keep integrating new solutions to our technology architecture, making as much versatile as possible. While that we must improve the performance of the systems as it is key to the user's productivity and to the operational continuity of the business in general. In the massive consumer industry, a track that won't come out in time for distribution of their products becomes a sale lost. So we need all the systems clockworking, being very fast and assertive, delivering always in time and also we put the best infrastructure at work. We have certainly faced a few challenges especially due to connectivity issues. When we had all solutions hosted in America, our operations in Asia were sometimes impacted by intermittence and latency. Being now on AWS and having deployed the model designed by Cloud SIP has allowed us to distribute our apps living in America what America requires and migrating to Singapore all applications used by our operations in Asia. In conclusion, always expanding multinational SAHI group we need to provide infrastructure and services that ensure the agility, speed, scalability and security aligned to the business strategic plans while maintaining a cost control which is being achieved through the design of STAR model network deployed by cloudec since IT allows each account to have its own subnets centralized through a transit gateway. And finally, with the deployment of an account of Albon in aws we have been able to decrease that cost of networking.
- [10:20]
Daniel Flores
Thank you. And now Federico, can you tell me a little bit about how codesive helped a head group design the final architecture that was going to be used in AWS to deploy their software?
- [10:33]
Federico Del Campo
Sure Daniel, and in fact, thank you for asking. It's a really interesting question. Of course we were aware about the importance of a company like AHE and the importance of their IT infrastructure for their business. The task was clear demuration of resources and dependencies of the BigMagic application from the on premise facility in Dallas to Singapore AWS region with the objective of mitigate latency problems for for clients located in Asia. First of all, we had to recognize that this was a significant change in the infrastructure, not just a simple task and it was also time sensitive. Having said that, we made a complete assessment of the resources in Dallas and then we proceeded to redesign the network infrastructure and apply a multi account strategy to comply with the pillars of the AWS web architected framework. Furthermore, along with AJ engineers we both decide to take a LEAF and Jeep migration strategy in order to achieve quick wins within short timeframes. We also decided the transition from a mesh network topology to start model with subnet centralized through a transit gateway appliance. I think it was a very useful approach because it allowed us to improve the infrastructure security through the deployment of an inspection VPC and a network firewall inside for east west traffic inspection. In addition, I would like to say that we were also committed in the efforts to reduce network cost such as reducing the number of NAT gateways, but centralizing outgoing traffic via the out on account I guess. On the other hand we also had to set up the communication between different regions and to achieve that we have established a site to site VPN between a Fortinet device in Ohio and the transit gateway deployed in Singapore and then also a site to site VPN between another Fortinet device in Dallas and the transit gateway in Singapore. In addition, we have also configured two paths for communication between data centers in Thailand, India, Vietnam and Indonesia to Singapore AWS region. And to get that done we have established two site to site VPN per facility, the first one between the local Fortinet and the transit gateway in Singapore and the second one between the local fortinet in the facility and the Fortinet appliance in Ohio. I guess another significant point was the migration of 7 terabytes of data, including resources like databases and no less than 40 Linux and Windows servers. And we had to get that done in just 10 days. Honestly, I think we couldn't be able to achieve that without such a reliable service like the AWS application migration services. It was I guess, a real game changer. And last but not least, we put a lot of emphasis on replicating the backup plans they had on premises. And to achieve that we used the AWS backup services, which is a simple but powerful service, to keep your data safe.
- [14:32]
Daniel Flores
Okay, and now Sandra, can you tell us a little bit about the benefits that a head group obtained after migrating to aws?
- [14:40]
Sandra Woolcott
Yes, the migration to AWS provides a number of benefits that you can see reflected in several areas of the operation in the first place and considering the current state of the company, one of the main concerns of east without doubt security, especially because some of our brands are sponsoring international events such as the Davis cup or soccer sport teams such as Paris Saint Germain which put put us frequently in the crosshairs of hackers that try to perpetrate a cyber attack. Fortunately, so far we haven't been successfully attacked. On the other hand, as I mentioned earlier, our technology has not been properly updated for years as it grew empirically and we are still in the process of catching up while the business maintains its growth plans and we must support the organization in the process. Which leads me to highlight that a cloud such as AWS provides with many services for its clients which allows us to test solutions, test of concepts, make several MVPs and as soon as one suits our needs we are able to implement it and scale it quickly which comes to us handy. While we are able to take exponential leaps in our process of technological advancement and digital transformation, network performance is very critical for us, especially because we have worldwide operation. That is where the START model designed by cloudec has allowed us to move some services to Asia to to the AWS Singapore's cloud, fixing most of the connectivity problems for users in Thailand, Indonesia and Vietnam and generating a very well received improvement in the performance of the applications and the processing of their reports which led to the country managers of this region to consider this migration as a successful case before the board of the company. Last but not the least, there is the issue of cost. Whoever that consider that going to the cloud is expensive is not considering the long term nor the operation cost of on premise services to support the growth of the company. We need flexibility to increase our services, to keep us updated and to simplify the government of the solutions. We are not a technology company, we are in the business of manufacturing food and beverage. So I prefer to have my team focus on delivering value to the core business and not distracted by activities that offer no real app value such as standing apps, upgrading licenses, backing up data. We leave those activities for the experts whose main task is precisely that to maintain the infrastructure always updated and optimized. The process of migration is not a one time effort, it is an evolving process as we are always going forward implementing and adopting new solutions, optimizing what we already have and trying to do everything as quickly as possible as the speed, scalability and security are key in our day to day operation. This is where it becomes very important to have the best strategic partners at our site Having the same sense of urgency we do. That is why we have chosen the team of AWS and cloudesive. We have without labs, improved the user experience. We have also gained agility, speed, scalability, security, simplicity in the government of the solutions. And also we are far from finishing our job. We are well on track.
- [19:30]
Daniel Flores
Okay, and now can you go a little deeper in what were the benefits that AK Group obtained by working with Caldesive as a consulting partner?
- [19:40]
Sandra Woolcott
Yes, I can say that Cloudesive is one of our main partners and without them we wouldn't be able to move forward as we have done. We have an excellent communication, we are supported 247 and I know that they are committed with our business needs. When we started the migration, there were not experts in cloud in our team and it is there where Cloud SAF played a very prominent role, bringing all its knowledge, experience and methodology to carry out the project in the shortest time possible without affecting the quality cloud. As if suggested design and the selection of the tools together we planned the activities, define the objective, identify the risks and elaborate a well defined contingency plan including different scenarios in order to ensure the business continuity of the company. That is why we were able to transfer more than seven terabytes of information in a short time. With the peace of mind of having a sound plan and the proper support. Thanks to the guidance and assistance of the experts of cloudusive.
- [21:08]
Daniel Flores
We know that a migration project of this size can be really challenging. So Federico, can you tell us a little bit about those challenges that you face during this migration and how you solve them?
- [21:21]
Federico Del Campo
Sure, Daniel, of course. As expected, there was many challenging situations, but I guess I can summarize them in three. Firstly, I would say the challenge of synchronizing seven terabyte of data was not so much due the amount of information to be migrated, but rather the set time frame. Fortunately, we already had experience in this kind of project and we were able to resolve it efficiently in accordance with the established timeline using the AWS application migration service. Secondly, I guess the challenge of designing the networking architecture to handle traffic efficiently and securely. This was achieved with the STAR network model and the use of Transit Gateway, which increases scalability and reduces complexity in both design and operation. Thirdly, I think the challenge of managing communication between different regions and data centers. This was resolved by implementing a variety of site to site VPNs between different types of appliances like the AWS Transit Gateway and Fortinet devices.
- [22:51]
Daniel Flores
Okay, now looking back at the project after you have finished, are there anything you would have done differently or any lessons you learned during this process?
- [23:03]
Federico Del Campo
Yes, for sure. In first place I would say the learning about the importance of a solid and efficient network design from the outset as reflected in the implementation of the START network model with the Transit gateway as a core appliance. Secondly, I must say the acknowledgement of the challenge of synchronizing large amount of data and how services like AWS Snowball could have accelerated the process. Thirdly, I guess the consideration for future project of implement AWS Direct Connect for more reliable and faster connectivity between regions and data centers. And I guess and last but not least, the focus on learning and continuous improvement from each project to continue providing the best possible service to our customers.
- [24:07]
Daniel Flores
And now this question. I want to make it for both of you. What advice would you give any organization that is considering a migration to aws? So Sandra, if you want to start.
- [24:19]
Sandra Woolcott
Thank you. Yes, based on our experience, what I will suggest is first of all, take your time and prepare your team ensuring that they have the required skills that will make process a lot smoother. Then be sure that you have clearly defined objectives as well as clear technical, functional and businesses expectations. After that, select the proper expert. In our case it was cloudsive. Be sure that your partner has the right communication skills and can understand the needs and urgencies of your business. And finally, planning conjunction with the business partner all activities, identifying any critical success factors and risks and do not start without having a clear idea of the technical requirements and the required performance. And it's very important at the same time develop the contingency plan and finally execute the plan, carefully monitoring each step. Federico?
- [25:31]
Federico Del Campo
Yes, for sure. In first place I would recommend them the importance of a detailed analysis of the current infrastructure, the needs and the goals in order to take the best decisions for the project and immigration itself. I also I will I would put emphasis in working with experienced professionals, whether in house or external because it allows you to make plans based on experience rather than theory. I guess another significant point could be the importance of a well throughout network design for efficient and secure infrastructure. And lastly but really important, I would say the need to plan the migration in phases, conduct tests at each stage and adjust the implementation as needed for for a smooth transition and minimize downtime.
- [26:37]
Daniel Flores
Okay, and now I have a last question for Sandra. During this migration project, were there any funny stories that you want to share with our listeners?
- [26:48]
Sandra Woolcott
Yes, at the beginning of the project we were very enthusiastic about the flexibility of the platform. So when we started finding the right sizes, we include so many just in cases to ensure that we have enough room for tests without really understanding how the cloud worked. Soon enough we received our first bill and we almost died on the spot, since the number basically invalidates all the economic arguments. Fortunately, AWS understood our predicament, saved the day by making a discount, and Cloudesive was instrumental in helping us to understand what we did wrong and how to properly optimize the use of the services and avoid dying of a heart attack. So in a way, they also save our lives and our jobs, so to speak. Thanks.
- [27:54]
Daniel Flores
Well now I wanted to thank you both for joining us today and all our listeners, and we hope you found this episode interesting and that you will join us in the next episode.