The move to cloud might seem an obscure and almost irrelevant topic for most people in scientific R&D. However, several key areas of impact are worth considering before choosing the right solutions for your scientific organization.

For the greater part of history, computing involved large machines housed in dedicated on-site data centers. Before the internet, there was no concept of external networks, and software access was exclusively from in-house terminals and PCs. 

The software itself was purchased as a capitalized expense, with fees and support based on users, processors, or transactions – or some heady mix of all three. Upgrades and new versions, often every two or three years apart, usually entail a significant administration workload to deploy, configure, and maintain. As the workload grew, organizations invested in more processing and storage capacity to enable greater performance.

Among the cost, complexity, and effort, in-house control of the software and systems offered a significant advantage: users could heavily customize their solutions to provide highly tailored solutions.

And then came ‘Business at the Speed of Thought’ 

Fast-forward to today's environment, and the talk focuses almost exclusively on the move to the cloud, including terms such as ‘software as a service’ (SaaS) and ‘cloud native.’ But what does cloud mean for science, and why is it important?
The cloud methodology is only important because of the benefits that the approach offers rather than the underlying technologies. But those benefits flow only if the claims of the cloud architecture match up to reality. 

To start, it is worth unpacking the terminology. Essentially, cloud-native refers to software that was designed for cloud operations, almost the opposite of the design techniques deployed for on-premises systems. The technologies include containers, microservices, orchestration, and more. The methodologies and development approaches for cloud-native software lead directly to enhanced security, scalability, and reliability, combined with significantly lower operational costs.

For cloud solutions, ‘SaaS’ refers to a delivery model where the vendor is responsible for building, operating, updating, and maintaining the software, which is then provided as a service. For a real-world analogy, think of Uber and Lyft offering ‘transportation as a service,’ while the drivers own, manage, and maintain the vehicles.

Cloud-native solutions are designed to be delivered on the SaaS model, paid as a monthly or yearly subscription with no capital outlay or ownership. Cloud-native design allows for rapid releases of new features, better quality, security best-practices, high availability, accessibility from any location, compatibility across devices, and convenience, to name a few key benefits.

Not All Cloud Solutions are the Same

Converting or migrating from older systems architectures to cloud-native solutions can be a costly, slow, and painful process for the software industry. There are many examples where vendors have done little more than move their existing solutions from in-house data centers to public hosting providers, such as Amazon Web Services, Google Cloud, and Microsoft Azure. Without cloud-native design, these solutions will not bring the benefits of cloud computing.

Naturally, there are some trade-offs with the cloud-native, SaaS, subscription approach: while the solution can be extensively configured, it cannot be fully customized. To take a consumer example, everyone using Gmail can configure and personalize the interface, and add integrations with multiple Google and other services. However, users cannot change the inner workings, such as how Gmail handles, stores, archives, and retrieves messages.

The SaaS approach means that all users access the same software, configured as each user prefers. With only one version to develop and maintain, vendors of SaaS software find that they can release updates and upgrades on a regular, frequent basis, offering service improvements and enhancements at monthly, weekly or even faster cadence. 

What Cloud Looks Like for Scientific R & D

In scientific R & D, cloud-native software offers distinct advantages. One of the major challenges facing research and clinical organizations is the exponential growth in data and the need for rapid analysis. This surge in workload can easily overwhelm existing on-premises systems, leading to costly and nearly continuous investments in new processing power and storage. These expenses divert funds away from essential science. In the past, the expense of investing in suitable high-performance systems formed a significant barrier for smaller organizations. Larger companies, too, might struggle to find the balance between systems capable of serving peak workload and not over-investing in capacity that would be idle during business-as-usual.

Cloud solutions have solved many of these issues of performance, cost, and scalability, and in doing so have democratized research and analysis, bringing near-limitless capacity within reach of almost any budget. Rather than being tied to substantial capital investments, SaaS makes the power and scalability of cloud solutions accessible to organizations of any size.

State of the Software Solutions Market 

As the commercial cloud marketplace has matured, software solutions have naturally also continued to evolve. An example of this is the portfolio of software solutions offered by Revvity Signals which have been designed and coded as fully cloud-native software solutions, covering the full gamut of research and clinical development, and are among a few (and arguably the best) cloud-native solutions in this sector. All users access the same software version, with new features and capabilities released monthly. The cloud subscription model allows users to scale up for intensive analytics with near-limitless capacity, and scale back when needed. Revvity Signals handles every aspect of software development and maintenance, enabling researchers to focus on their core mission: developing and exploring new drug therapies.

Finally, Artificial Intelligence (AI) and Cloud

With sufficient training data and processing power, AI offers immense pattern-recognition potential, now fully enabled by the capacity and scalability of the cloud. Since research in many fields depends on discovering patterns, AI could soon transform the discovery landscape. To the point: cloud-native solutions are ideally placed to exploit the AI world. Vast datasets are readily available, the cloud ecosystem provides secure, reliable infrastructure, and AI services can be connected and scaled according to demand. Cloud-native solutions such as those offered by Revvity Signals are able to access cloud-based AI rapidly and easily.

For forward-thinking researchers, choosing a truly cloud-native solution will significantly impact today’s research and tomorrow’s yet-to-be-imagined workloads. The Revvity Signals platform is prepared to support both.