Sigma Technology supports customers with digital transformations and upgrades of the technology portfolios. We aim to help our customers gain business resiliency together with technological excellence, building a solid background for them to make a competitive leap forward and operate “like never before.” Cloud computing technologies enable the most innovative and boldest solutions, making the path towards digitalization smoother. So today, we are talking with Jimmy Dahlqvist, Head of AWS at Sigma Technology Digital & Cloud Solutions, about cloud computing and new trends that will certainly facilitate how companies utilize their cloud ecosystems.
Recent RedHat research of the post-COVID-19 work showed that though the right technologies at the right place are the third most crucial pillar of the digital transformation, they influence the success of transformation on the corporate culture and operations levels. The dramatic changes in the way of working caused even more intensive usage of cloud capacity or urgent migrations for those who were still using traditional data centers. For example, according to Contino’s report on data maturity in the public cloud, 32% of technical use cases in 2021 is cloud data migration. This is clear evidence that a fair amount of work is still going on in moving to the cloud. So let’s sort the things out and discover how you can squeeze the most out of the cloud advantages!
Why do you think cloud computing has become the “must-have” rather than just an option?
Cloud development brings so much speed and agility into the development process, making it impossible not to adopt. Development teams can easily create and manage the infrastructure needed for their application, and the business itself benefits from a “pay as you go” billing model, where an organization avoids the risks of under- or overprovisioning. For example, now teams can create a new server in the other part of the world just in a matter of minutes. Basically, you could create a working prototype running in a different part of the world in less than a day. Something that was unbelievable that long ago.
Moreover, DevOps philosophy is becoming more and more popular development methodology because of its significant advantage to streamline the delivery, align all teams of the production lifecycle and break the siloes in communication. And here, cloud technologies play a vital role, enabling leveraging DevOps tools at scale. That’s why here cloud computing becomes the “must-have” instead of the “nice-to-have,” opening the gates to working like “you build it, you run it.”
What types of organizations benefit first from cloud solutions?
This is an interesting question. I think that most organizations benefit from cloud solutions. What has been seen during the last years is that start-ups that are born in the cloud quickly can become a threat to major enterprises. We can also see that so many Independent Software Vendors (ISVs) migrate away from the traditional packaged software model towards a SaaS (Software as a Service) solution based in the cloud. There has also been a boom in FinTech and EdTech industries, much due to cloud solutions. But which organization will benefit first? I believe that most organizations benefit from the cloud, so very hard to say who would benefit first.
Does it have any restrictions? For example, the medical sector in the USA is strictly regulated, and data privacy is the first issue towards cloud transformation of the industry.
The short answer is YES.
There are so many rules and regulations that you need to consider during a cloud migration. For example, HIPAA (Health Insurance Portability and Accountability Act) is a US regulation in the health care industry, and PCI DSS (Payment Card Industry Data Security Standard) is a regulation to secure credit card payments, just to mention a few. But Most services, offered by public cloud vendors are certified for a bunch of regulations. That doesn’t mean that you fulfill the requirements just by using these services. It means that you can still meet the requirements when you are using the services.
When it comes to something a bit closer to most companies in Europe, GDPR (General Data Protection Regulation) is a regulation that you as an organization must follow. So you need to keep that in mind when designing a new cloud solution. Right now, the discussion is much about the CLOUD Act (Clarifying Lawful Overseas Use of Data) and Schrems II and if the public sector can use US-based cloud vendors.
So yes, rules and regulations affect cloud services and the way how to build your cloud solution. This is something that must be considered when planning a cloud migration or a new cloud project.
What trends do you see for the next few years? Does it require significant changes for those who are already operating in the cloud?
I would say that serverless development is a trend that is already here. In the coming years, I think that it will only grow stronger and stronger. Indeed, the term “serverless” doesn’t mean that no servers are used or managed, this process is just hidden from developers, and vendors take over the management responsibility.
Serverless computing comes with such great benefits as:
- only paying for exactly what you use
- no-Ops autoscaling of the resources, almost towards infinity
- high fault tolerance and 99,9% service availability
- quicker deployment since all changes are applied through calling API
- greater testability and opportunity to see how changes would affect your current infrastructure
- easier integration thanks to Backend as a Service, special services to perform specific tasks
- enablement of building efficient microservices-based architectures
The main benefit of allowing organizations to focus on solving business problems and spend less time managing servers is a strong driver to leverage serverless architecture. We shouldn’t also forget that serverless cloud computing is a perfect match for event-driven architectures. Companies that are already in the cloud that like to adopt serverless need to train their personnel and start shifting mindset from servers to thinking serverless.
Along with serverless computing, there is a dramatic uptick in using the multi-cloud strategy. According to BUSINESS WIRE’S multi-cloud research, the multi-cloud management market is expected to grow to USD 4,492.7 Million by 2022 – just imagine, in 2017, the market size was only USD 1,169.5 Million! Out of experience, multiple public cloud solutions and hybrid cloud are the top trends nowadays, though some enterprises still opt for multiple private cloud solutions. I think the evolvement of IoT and AI/ML technologies considerably drives the development of multi-cloud strategies.
One more trend that I’m seeing is around Cloud Economics and FinOps. Many organizations have moved to the cloud, and it occurred to be not so cost-efficient as they thought. It can be everything from overprovisioning to incorrectly using the services. That’s why FinOps consulting is one of the most trending services among IT companies.
When multi-cloud approach suits best? What constraints can organizations meet when they decided to adopt it? Do you have an example when an organization boosted its IT delivery thank to the multi-cloud approach?
Multi-Cloud has really picked up traction lately. There are a couple of scenarios when Multi-Cloud is a good approach. One is when you like to use services from different cloud vendors since vendors are good at other things, and vendor lock-in is not an issue for you. Another is regulatory and business (customer) requirements.
I think it’s a poor strategy if you are trying to avoid the vendor lock-in. For me, this often results in the situation when your development teams are not able to use managed and vendor-specific services. You simply have to use the least common denominator, which often is virtual machines. In my eyes, there is no vendor lock-in. You can always change, but change comes with a cost. So you could either pay over and over again for something that might happen, switching cloud provider, or you can pay for it if and when it happens. For me, the correct choice is to pay for it if and when it happens.
The best use case for multi-cloud I know is an organization running a multi-cloud setup with AWS and Azure with great success. The company uses AWS to ingest and transform data. They are then running its BI tooling in Azure. The success story here is that AWS was the better choice when it came to data ingestion and transformation, while Azure was the natural choice for the BI tooling setup.
Why do you highlight serverless development as trend #1? What problems can be met with serverless development? And what’s the difference from the simple resource allocation in a server?
I actually started my cloud journey in a serverless way. One of the first things I did, was the creation of an AWS Lambda function, it stuck with me after that. Serverless enabled my team and me at Sony to focus on the business problem we wanted to solve. We didn’t have to be worried about auto-scaling or patching servers. We got all the solutions out of the box. For me, this is the most significant advantage of serverless, focus more on the business problems and less on infrastructure management.
If you, as an organization, want to adopt serverless and come from a traditional server development, you need to rethink how you build your solution. With serverless, your infrastructure quickly looks complex with so many moving parts, so this will be a challenge. You must also rethink how you deploy the solution since it’s very different.
The biggest difference in resource allocation is that with serverless, you pay for what you use if you invoke a Lambda function 100 times per second, you pay for that, if the function is not invoked, you pay nothing. You will not have to pay for any extra allocated headroom that you will with traditional server allocation.
What would you recommend for companies wishing to go to the cloud? What should they know?
Companies should start by asking the Why and What questions.
Why are they migrating to the cloud? What are their motivations? What challenges do they face? What is the expected outcome? They must make sure decision-makers and leaders, both official and unofficial, are committed and onboard. They should also know that the cloud adoption journey is not a straight line. That’s why road mapping, thorough planning, and risk management are the keys to successful cloud migration.
My last advice would be to work with a cloud partner, the partner can come with advice and guidance, and in the long run, this will generate a smoother ride with a lower cost.
Cloud technologies enable digital transformation and become the “must-have” for those wishing to reach the edge of innovation. Though cloud technologies may seem like a silver bullet, transition to the cloud requires a thorough discovery phase of infrastructure and business layers since the simplest “lift’n’shift” strategy doesn’t work anymore. It’s important to consider the fact that all IT infrastructures are unique and require a custom, data-driven approach and solid know-how in the background. Rash decisions may cause uncontrolled spendings and technical debt. To avoid this, companies have to take care of FinOps and financial operations after the migration to the cloud and before.