As I mentioned in my earlier blog on “Banking on a Hybrid Cloud world”, its accepted that hybrid cloud adoption in financial services outpaces all other industries. Whether it is purely with the public cloud providers, or more typically in a hybrid cloud model, utilising a mix of private cloud capabilities and multiple public cloud providers is more focused on agility and speed of delivering change - whether as a FinTech with no legacy, or an incumbent firm. However, whilst most financial services firms have adopted some form of hybrid cloud, the focus is now turning to multi-cloud.
In a recent Flexera survey, 84% of respondents said they are working with two or more of the public cloud providers and actively have a multi-cloud strategy. However there is still much debate as to what multi-cloud actually is – ask any two firms who have a multi-cloud strategy what it means to them, and you will likely get two different answers. Multi-cloud could be where more than one of the public cloud providers are used by a firm, with each being used for specific workload types, or it could be where the same workloads are run across multiple public cloud providers. In the former, the drivers are typically to leverage specific IP available from different public cloud providers to enable a firm to accelerate a business idea or offering. In the later, the driver is typically regulatory or risk related.
With wider adoption of the public cloud providers by financial services firms, the various regulators have all started to focus on how to understand and mitigate the risk posed by multiple institutions relying on the same underlying providers. One example of this is the Financial Stability Board (FSB), an international body of the G20 central banks and supervisory organisations, which noted is concerns about the concentration risk of cloud services in the financial markets in a report issued earlier this year. Similar to this view point, the UK Governments Treasury Committee recently released a report that explored IT Failures in the Financial Services sector. In this report they too highlighted the potential concentration risk that the large public cloud providers pose, with a recommendation that the UK Government should urgent consider how best to regulate cloud service providers, and that regulating them as a critical infrastructure (the same as payment processing companies such as Vocalink) may be necessary.
Another example would be the advice given by the UK’s FCA which treats the use of public cloud by regulated firms the same as it treats outsourcing the provision of any services; a key requirement being that “firms should have exit plans and termination agreement that are understood, documented and fully tested” as well as knowing “how [a firm] would transition to an alternative service provider and maintain business continuity”.
There are many challenges to effectively supporting a multi-cloud strategy within a firm – these range from how do you ensure you have the right skills in your workforce, to selecting the right tools to effectively manage services on multiple cloud providers environments, to effectively managing security and compliance across both your private cloud, as well as the environments managed within multiple cloud providers.
Firms that are looking to adopt a multi-cloud strategy to support portability of applications across multiple cloud providers, have typically already embarked on an approach to modernize their applications and leverage architectures such as micro-services, container deployment patterns etc. In April 2019, Google Cloud announced their hybrid-multi cloud product set – Google Cloud Anthos. One of the key capabilities of Anthos is that it allows organisations to deploy both virtual machines as well as container based applications in Google Kubernetes Engine (GKE) on Google Cloud Platform, or on GKE On-Prem within a customer’s data centre. In addition, support for GKE on both Microsoft Azure and Amazon Web Services is also planned, thus providing the ability for firms to be able to do deploy the same application, using the same tool set across multiple public cloud providers.
As part of the Google Cloud Anthos announcement a number of platform partners were announced, including HPE. HPE has a number of validated designs supporting different use cases ranging from edge to core predictable workloads at remote sites, un-predictable workloads that require data mobility as well as application mobility, and finally platforms that can run both containerised, virtual and bare-metal workloads. All of which utilise AI powered infrastructure, that can be consumed however a customer wishes (be that via a traditional capital investment model, or utilising HPE GreenLake to benefit from cloud economics for on-premises deployment).
Whilst Google Anthos helps support the application mobility, applications are only as powerful as the data they access. To support data mobility and consistent data services, HPE has also integrated HPE Cloud Volumes with Google Cloud (as well as with Microsoft Azure and Amazon Web Services). HPE Cloud Volumes is a cloud data service that is integrated with on-premises HPE storage arrays delivering the flexibility of bi-directional data mobility and data protection. The integration with the three public cloud providers means that replicated or migrating data to and from the public cloud does not incur data egress free, which can some-times be significant.
Whether a firm is looking to adopt multi-cloud to leverage best of breed IP on different public cloud providers, or they are looking to adopt multi-cloud to meet regulatory requirements, what is key is both ensuring that they have the right operating model and skills to support multiple cloud providers, as well as understanding what services to deploy where. At HPE, our Pointnext Services utilise the HPE Right Mix Advisor to deliver a data-driven approach to help organisations develop their hybrid cloud strategies and to understand where best to place workloads and applications according to the individual organisational needs and drivers, as well as assessing how to modernize applications.
To conclude, over the last twelve months, more and more banks are talking publicly about what they are doing in the hybrid cloud space; they are getting confidence both to look to change, but also to talk about their experiences. I think over the next twelve months we will see more of a shift in conversation around how they are successfully adopting multi-cloud strategies. However it is important that we do not forget that moving to any cloud model is not the ultimate outcome – it is about enabling what the business is overall trying to achieve.
Chris Ibbitson
Hewlett Packard Enterprise
twitter.com/Ibbitsc
linkedin.com/in/chrisibbitson/
hpe.com/uk
Chris_Ibbitson
Chris is a Chief Technologist for HPE, focused on the Financial Services industry. Before joining HPE, Chris has worked at both a Global Systems Integrator, as well as at a Global Bank in a variety of senior architectural roles.
