Cloud-powered agility for your data marketplace: matching on-premise security benchmarks
Our latest Q&A blog explains why businesses should choose a cloud-based option for their data product marketplace in order to increase consumption, minimize administration, effectively protect data, and scale to meet growing demand.
Harnessing the cloud is now central to the IT strategies of most organizations, providing flexibility, cost, and security benefits. However, given the importance of keeping close control over their data, some businesses may be wary about deploying a cloud-based data product marketplace. We spoke to David Thoumas, Huwise’s Chief Technology and Product Officer, and he explained how to overcome any concerns and why a cloud approach delivers business benefits while effectively protecting data.
Why did Huwise adopt a cloud-only model from the start, despite possible barriers to sharing private data?
Choosing the cloud was primarily a go-to-market decision for Huwise. When we began, we could see that our data marketplace platform delivered benefits to clients around the globe, across a range of different industries.
However, at the time, packaging and deploying on-premise software was much harder than it is today. Our objective was to be able to deploy our solution quickly anywhere in the world without having to deal with complex installation procedures on a customer’s premises.
We didn’t want to limit ourselves geographically or around deployment resources. Harnessing the cloud allowed us to make the solution available globally in a much simpler way.
Back when we were founded in 2012 this was actually a bold decision, because cloud adoption was not as widely accepted back then—especially in the public sector. But in the end, that choice proved to be the right one – we now have over 350 clients in 25 countries around the world, including leading businesses such as banks and financial services companies, as well as major public sector organizations.
Beyond the obvious, what benefits does the cloud offer that are sometimes underappreciated?
While cost was the original driver for many companies to adopt the cloud, what they are now realizing are key benefits around operations, scalability, agility and security. Cloud platforms dramatically reduce the effort needed for maintenance, infrastructure management, and monitoring. New features and capabilities can be easily added, without requiring lengthy downtime. Essentially it shifts investment from capital expenditure (CAPEX) to operating expenditure (OPEX).
As workloads increase, the cloud scales to cope. Some of our customers, such as Banque de France have data marketplaces that share huge volumes of data with thousands of users on a daily basis. During the Paris 2024 Olympics, the organizer’s Games Map, powered by Huwise, saw millions of unique users, generating spikes of more than 200 requests per second at its peak.
What is currently driving the increase in cloud adoption, particularly in sectors that have traditionally remained on-premise?
In an increasingly digital world, security and data protection is a key priority for every organization.
When it comes to infrastructure security, the major cloud providers—Microsoft, Google, and Amazon—have capabilities that no individual organization can realistically match. Their investments in security, compliance, and infrastructure protection are enormous.
At the same time, hackers are increasingly targeting public entities, such as municipalities or local government bodies, as they can lack the budget and the expertise to secure their infrastructure effectively. Over the past few years, many have suffered cyberattacks such as ransomware or distributed denial-of-service (DDoS) attacks.
In several cases involving our customers, their internal systems were taken down, but their public data marketplaces remained operational because they were hosted on secure cloud infrastructure.
However, it’s important to understand that simply relying on a cloud provider’s security measures is not enough. Even if the cloud infrastructure itself is secure, the software deployed on top of it is not automatically protected. Security still depends on how the application itself is designed and managed – something we spend a lot of time and resource on at Huwise.
Are there examples of where cloud implementations have overcome security challenges?
I can think of two examples – one of which is very current.
Cyber attack
Firstly, a few years ago, several French government websites were targeted by international hacking groups using distributed denial-of-service (DDoS) attacks. Some of the affected sites were operated on our platform, and while the attack caused initial disruption, we worked closely with the French cybersecurity authorities to quickly recover services and deploy countermeasures. Similar attacks occurred several times afterward before the group responsible was eventually stopped. Thanks to the countermeasures we had in place, these attacks had no impact on our service.
Physical attack
The current example is the military strikes that recently hit the data center of one of our partners in the Gulf region. One of our customers was hosted in that location meaning that we had to immediately activate our disaster recovery procedures. Thanks to our strategy of hosting backups across different geographic regions, we were able to restore the service from a data center in another region in less than 24 hours.
How can companies balance control of their most sensitive data with benefiting from the cloud?
While the majority of organizations are now embracing the cloud, others, such as those in defense or critical infrastructure have to operate differently. They may completely isolate their systems from the internet to prevent any possibility of unauthorized external access.
For these types of organizations, on-premise infrastructure will remain necessary for decades. To tap into the benefits of the cloud, the most realistic model for many of them is a hybrid approach, where:
- The most sensitive workloads remain on-premise
- Less critical workloads are moved to cloud environments
Our platform is moving to be Kubernetes-ready, allowing applications to be deployed in a cloud-native and infrastructure-agnostic way. This means it can run on major public clouds, private cloud environments, or on Kubernetes clusters running on-premise.
This will allow customers to host their own internal data marketplace on their infrastructure if needed, sharing data within the organization while meeting their security objectives.
Why do some customers still prefer on-premise deployments? How can these concerns be overcome?
We typically see two main reasons – organizational history and data sovereignty.
Some industries—such as banking, finance, and defense—have historically built their entire infrastructure around on-premise security models. Moving everything to the cloud is not always realistic given the size and complexity of their infrastructure. Instead, they gradually move certain workloads to the cloud while keeping others on-premise. French bank Groupe BPCE uses Huwise to share less sensitive data internally and across its ecosystem, increasing collaboration and improving internal efficiency.
In terms of data sovereignty some organizations want guarantees that their data will remain under specific national or regional jurisdictions. Even if the data is not extremely sensitive, they may want to avoid situations where foreign governments could potentially access it. That means they require cloud hosting that is guaranteed to be in specific countries, rather than internationally.
How challenging is data sovereignty today? How do you support clients in managing it?
To deliver data sovereignty safeguards, you need to be able to offer a choice of cloud provider. From the beginning, we designed our architecture to be cloud-provider-agnostic. This is different to many SaaS vendors who built their platforms using proprietary services from a single cloud provider. That makes migrating to other platforms down the line difficult.
We intentionally relied on services that exist across multiple providers. As a result, we’ve been able to deploy our platform across multiple infrastructures over time, including:
- AWS
- Microsoft Azure
- Google Cloud
- Local French cloud providers
This multi-cloud capability allows us to host data in specific countries when required—for example, hosting data in France with a local provider for customers who need that level of sovereignty.
While data sovereignty is a legitimate and growing concern, it is important to balance it with the fact that cloud hyperscalers offer unparalleled reach, performance, resilience and security. This means that organizations must decide:
- Which data is truly critical
- What must remain sovereign
- and What can be processed in global cloud infrastructures
Techniques like anonymization, encryption, and segmentation can also help protect sensitive information even when using external infrastructure.
How do you work with organizations that cannot move their data to the cloud?
One key point is that the data marketplace does not need to store all the data itself – it can just be a central access point for users.
In many cases, the marketplace simply manages metadata, while the actual data remains in secure internal systems, and is accessed via the marketplace through self-service.
For example, some customers just use the platform to expose metadata about datasets stored in their internal data lakes.
Organizations should also consider that not all their data is equally sensitive. Often, a portion of the data they want to share internally can safely be handled through a cloud platform, whatever their wider security policies.
Another common starting point is external data purchased from third-party providers. This type of data usually has fewer security constraints and can easily be distributed through a marketplace, which makes it much more accessible and cost-effective.
What benefits do customers gain from multi-cloud deployments?
As data becomes ever more central to businesses, what we are now seeing more frequently is a focus on backup strategies and disaster recovery plans. Customers want to understand where backups are stored, how quickly services can be restored, and how resilience is ensured. Requests for the ability to instantly switch between hyperscalers are still rare, but interest in robust recovery strategies is definitely increasing.
What would you say to organizations looking to deploy a data product marketplace but who are not yet sure about moving to the cloud?
Based on its research and market understanding, Gartner believes that cloud computing will be a necessary component for maintaining business competitiveness for all organizations by 2028.
Those looking to benefit from the increased data consumption and value that a marketplace brings, but worried about the cloud should focus on three areas:
The cloud is flexible
Not all data or applications need to be hosted on the public cloud. There’s now much more flexibility around deployments. For example, the data product marketplace can be cloud-based, with data remaining on-premise, or with information split depending on its sensitivity. Private and hybrid clouds can provide additional safeguards as required.
Huwise supports your cloud success
We are dedicated to helping our clients maximize the use of their data to create value, as seamlessly as possible. Not only is our platform easy to administer, it is backed by the highest level of security. Alongside the protection provided by cloud providers, we have ISO27001 certification, a dedicated security team and policies, as well as proven experience in crisis or attack management. All of this minimizes any concerns you may have about security, taking a load off your mind.
The cloud is a low risk option
We’ve seen a number of projects where companies have aimed to create their own on-premise marketplace. Generally, while they deliver a usable solution, it can prove to be a more complex, costly, and longer process than first imagined. Using the cloud instead accelerates deployments and ensures intuitive access to data, lowering risks.
Organizations understand the vital importance of harnessing their data to improve business performance through data product marketplaces. Embracing the cloud for these key platforms is critical to creating the foundation for secure, self-service access to data for humans and AI, increasing consumption and delivering ROI.
Share this post:
Articles on the same topic: