The Challenges of Cloud Native Development

One of the hot topics in the developer world at the moment is building microservice-based cloud native applications and the pervasive nature of this approach in today’s development ecosystem. However, a significant number of developers have adopted a different approach. 

 

Many of the developers I have spoken to recently have had a couple of things in common; firstly, they all come from organisations that own or process huge volumes of data on a daily basis. Secondly, all need to create a process involving both analytical and transactional workloads with that data using a variety of different technologies. 

 

These analytical workflows are varied in nature. Many have found that being able to run SQL, for example, isn’t enough to solve the functional and non-functional requirements of queries in an adequate manner.  

 

Consequently, a number of developers have been looking at non-traditional, bespoke data management solutions, usually running in a private cloud and often on specialist hardware. Given there is such a wide variety of well-known cloud native and on-premise platforms out there, in both the SQL and NoSQL spaces, it is interesting to see the number of organisations feeling the need to do this. 

 

This is notable as while it may seem like the most cost-effective option to begin with, it is likely to turn out to be far less economic in the long-term. Besides, there is considerable risk inherent in this approach. 

 

For these organisations, all will be well if the developers building the solutions remain with them. However, if they decide to leave – and let’s face it the competition for developers couldn’t be stronger – then, their employer has to find a way to retain them or face a knowledge vacuum surrounding the platform. 

 

The other issue is a matter of functionality. Once the organisation wants to do something extra with the platform they will need to set up a data pipeline and replicate it in another data store, reconstructing it along the way. Before they know it, they have the same data replicated in four or five different structures. Suddenly, what started out as a cost-effective platform developed for a particular purpose has become both expensive and complex. 

 

Interestingly, concerns about complexity are increasingly evident. When the number of discrete elements to maintain, monitor and manage increases, the complexity of doing so can scale in a non-linear manner. People are recognising that this can be a significant barrier to supporting and enhancing systems that have been deployed in a highly granular architecture and are looking for alternatives; this is one aspect deterring developers from going cloud native.  

 

Additionally, if you consider the lower cost storage options on AWS, or any of the other cloud platforms, they are not fast mechanisms. If a business is data-intensive, then it does beg the question as to whether cloud native is the right route, as the data replication inherent in satisfying transactional and analytical workloads in a cloud native environment becomes increasingly expensive.  

 

Of course, there are positives, especially if a developer makes use of all available services offered by the cloud provider. This can significantly reduce the time to build and deploy a solution – and ensures it is highly scalable. However, this does tie the organisation to a particular cloud provider; if a solution is built as a native application, for example in GCP, it can be difficult to refactor to move to another cloud environment.  

 

It’s not surprising that many developers are building their own cloud native applications. However, it is surprising that so many, from such a technically inquisitive and inventive group, are implementing their own data platforms. They are forced to do so because of the shortcomings of cloud native architectures that scale well for relatively simple use cases that can easily be decomposed into small, functional units, but aren’t as good of a fit for more technically challenging requirements. This all means that new market developments which are addressing these challenges in different ways are certainly moving in the right direction. 

 

To find out more about the UK’s future cloud ecosystem, visit our landing page by clicking here!

  • Tom Henderson

    Tom Henderson

    Programme Manager | Smart Cities and IoT
    T 020 7331 2043

Share this

FROM SOCIAL MEDIA

Last chance to register for our Monday event on how AI can help us adapt to climate change. Awesome panel with our… https://t.co/gmPF7VyvE9
This Monday, join our webinar on ethnic #diversity in #healthtech & help us identify how we can do more to ensure d… https://t.co/pvfJLruEKB
Featuring techUK, @gov_procurement & @DWPDigital @GovNetTech will explore the value of public services investing in… https://t.co/b9jaaoDBJ5
Read the guest blog from @VipaDigital on why #Criticalnationalinfrastructure (#CNI status) for #datacentres is not… https://t.co/Lx0x8V0IGa
Join other local authorities & industry for 'Digital Insight and Business Intelligence in Local Gov 2020,' webinar… https://t.co/d1cwNpCQAs
Learn what makes #SMEs unique and gives them a competitive edge in the Defence and Security market, in a series of… https://t.co/u4CoBLxDqJ
Research published by @OctopusVentures earlier this year found that investment in #mentalhealthtech surged from £12… https://t.co/gqjp4PwBdA
Become a Member
×

Become a techUK Member

By becoming a techUK member we will help you grow through:

Click here to learn more...