Cover Story News

Is the Multi Cloud Model Too Much Too Soon for the industry?

According to IBM, 85% of enterprises are in a multi-cloud environment, but only 41% have a multi-cloud strategy. The challenges that the organizations are facing, outweigh the benefits they are accruing from the model. Is multi-cloud coming into an industry that is not yet ready to optimally utilize it?

A multi-cloud environment is a type of digital infrastructure where an enterprise distributes its cloud assets, software, applications and more across several cloud providers. Every cloud provider has unique things to offer, each has their different architectures, because of which they all offer different levels of security, adherence to compliances, managed services along with varying billing models and support for different technology platforms. It is on the part of the organizations to select the best suited cloud type for their business needs at a cost that fits their budget.

“Vendors have to buy-in to support interoperability to the highest degree possible, and co-exist & operate like a well-oiled machine to provide a seamless cloud experience to the client, thereby having organizations to really benefit from their multi cloud strategy.

Biswajit Mohapatra, Executive Director, IBM Cloud Migration Factory, Cloud Application Services.

Commenting on different perspectives that led to the adoption of the multi-cloud model, Subramanian Gopalaratnam, CTO, Resulticks says, “One school of thought holds that key factors in the market are resilience and, disaster recovery capabilities to ensure business continuity across the cloud. The second school of thought is to provide or consume specific enablers from each of it. For example, Amazon Pegasus provides a great stream processing ability for a company whereas they might want to use Azure’s database capabilities as they are a Microsoft house. This provides certain point of views on why one is using the multi-cloud model. But at the same time, there is also another school of thought which maintains that different clouds provide the same solution, which means, one is not utilizing the specific cloud’s native or proprietary offerings.”

“Organizations all around the globe are moving towards the adoption of multi-cloud approach to take better control of their cloud ecosystem. A multi-cloud strategy should be designed to simplify workload migrations, eliminating silos and cutting operational costs, thereby achieving a consistent cloud management experience.”

Surajit Sen, Chief of Staff, Modern Data Centre, APJ, Dell Technologies.

Multi-cloud environment does accrue cost and efficiency benefits, but it also complicates cloud management in the same process. Given that each individual cloud provider has its own siloed systems and configurations, the overall management of the cloud requires more resources and time. Visibility across clouds might be complicated. 

“Multicloud is  playing out in one of two ways: Organizations committing to a platform but still using differentiating services from any source; and Building core application development and deployment strategies utilizing Kubernetes and Serverless to stay vendor neutral,” adds Subramanian. 

With adoption of containers and open programming languages by cloud providers there is an opportunity to leverage multi-cloud for capabilities and features. “Multi cloud comes with the promise of agility and freedom. Shifts to modular application design, microservices, containers, cloud-native applications along with AI/ML, IOT, Blockchain and edge initiatives are enabling multi cloud adoption.” said Biswajit Mohapatra, Executive Director, IBM Cloud Migration Factory, Cloud Application Services.

“Cloud computing has reached a new stage of maturity characterized by movement away from infrastructure-as-a-service (IaaS) to the next level of platform-as-a-service (PaaS). This trend will evolve over the next five years, as PaaS becomes an entry point to the Cloud.”

Subramanian Gopalaratnam, CTO, Resulticks.

“With IP Lock-in and portability issues abating, customers like having best of breed options for choosing clouds for their unique differentiating advantages or workloads e.g. Azure for DevOps automation or GCP for AI/ML, AWS for number of services and availability zones etc. Today we have customers who are developing cloud native apps on their On-Prem PaaS, AWS for bursting scenarios and plan to get their insights from their data lake on GCP. Customers are often picking one dominant partner as default or go-to option and use others to augment the first option,” says Yesh Subramanian, SVP Digital, Mphasis. 

“As companies get more comfortable with the idea of utilizing a multiple public cloud strategy, they are able to make choices based on cost, performance and application capability – potentially moving between clouds as these dynamics change.”

Alex Li, GM, Alibaba Cloud South Asia.

Advantages of Multi Cloud 

Migration of application and workloads from on-site data centers to the cloud, as well as the development of cloud-ready and cloud-native applications are fuelling growth in cloud space. When they find their organizations so heavily dependent on cloud, CIOs are becoming cognizant of the fact that they shouldn’t be leaning on a single cloud service provider for the entirety of their workload. 

“Challenges around single cloud reliability, cloud service provider lock-in, regulatory compliance, price-sensitive deployments, data sovereignty, reality of shadow IT, latency/proximity concerns and resiliency are some of the driving factors leading to hybrid multi cloud model,” said Biswajit Mohapatra, Executive Director, IBM Cloud Migration Factory, Cloud Application Services. 

What was the urge to avoid the nightmare of vendor lock-in, has paved way to the adoption of the multi cloud model that at the face of it offers better disaster management. As the industry started dividing up its workload to different CSPs, it realized that breaking free of the vendor lock-in is merely the tip of the iceberg. 

As the organizations gain a deeper understanding of the cloud services they use to run their workloads, they have started realizing that they don’t necessarily have to associate themselves to a single cloud provider. Organizations are becoming cognizant that a multi-cloud strategy allows them to select solutions from different providers for different needs, based on the provider’s merits. By using multiple clouds, organizations can avoid vendor lock-in. This essentially means that they are benefitting from the strengths of every provider without limiting themselves to a single provider’s portfolio. 

Using multiple clouds allow organizations to carry out Flexible Architecture Design, which enables them to customize their deployment to best suit their existing setup. This makes it easier for organizations to integrate legacy infrastructure. 

In a public cloud model, organizations have the ability to spend on what they need at the time they need it. A multi cloud model considerably enhances an organization’s On-demand Scalability. Managed correctly, public clouds cost lesser compared to private cloud, especially when the low capital outlay is factored in. 

A multi cloud enables the user to own the foundational infrastructure while leasing more resources to deal with spikes in traffic, ensuring they only pay for actual consumption. This way organizations can meet financial targets with the help of pay-as-you-go billing,” says Alex Li, GM, Alibaba Cloud South Asia. 

Multi-cloud provides more options and gives us the ability to balance certain regional obligations from a compliance angle. 

“Increased agility enabling workload mobility between heterogeneous cloud platforms, improved resiliency, robust security, competitive pricing model with cost-performance optimization , reduced risk of service disruption, driving innovation through multi cloud management and monitoring process and freedom form cloud service provider lock-in are the advantages experienced in the adoption of multi cloud model. I strongly believe faster time to market, monetized data for new revenue streams, optimized supply chain, enhanced market reach, superior customer experience and cross organization collaboration can be achieved through hybrid multi cloud adoption. Fit-for-purpose cloud architecture, compatibility with emerging partner ecosystem, competitive pricing model, avoiding cloud service provider lock-in, ability to transition seamlessly for better alternatives, security and privacy are at the top of my list” said Biswajit Mohapatra, Executive Director, IBM Cloud Migration Factory, Cloud Application Services.

“Taking the multi cloud approach help in achieve increased resiliency, agility, performance, cost savings, resource optimization, and adherence to compliances as compared to being tied up with a single vendor.”

Dr. Rajeev Papneja, Chief Growth Officer, ESDS Software Solution Pvt. Ltd

Changing Dynamics Between the Cloud Service Providers

For a service model to not just sustain, but to thrive, it has to drive growth for both parties involved. Popularity of multi cloud is an opportunity and a driving force for the vendors to deliver a more robust and wonderful customer experience. Commenting on how a multi cloud model leads to vendor growth, Subramanian Gopalaratnam, CTO, Resulticks says, “It forces the vendors to provide solutions capable of scaling across continents and countries without violating any of the regionalized compliance requirements. This is where a multi-cloud option opens up greater opportunities for vendors to provide an enhanced solution.”

True to its nature, competition is one of the drivers of innovation for cloud service providers. The advent of multi cloud is forcing cloud service providers to partner with each other for providing a seamless client experience. Competition is resolving itself in favour of collaboration. From pitting the vendors against each other for a larger share of the market, organizations now require them to work in coalition so that they can reap the benefits of the unique strengths of different cloud providers. While some cloud provider would be strong in providing IaaS, others would have great business solutions, vendors collaborating to enhance the multi cloud model is how the customers can meet the entire gamut of their business cloud requirements.

In many ways, the multi cloud revolution re-constituting the relationship between the organizations and their Cloud service providers. Dr. Rajeev Papneja, Chief Growth Officer, ESDS Software Solution Pvt. Ltd, elaborates on this. “A key component of multi cloud model is the Multi cloud management platform. Every Cloud Service Provider had Cloud management platform to manage their own platform and with the advent of multi cloud models, the first thing every vendor had had to do is redesign their CMP to make it MCM platform. This move has made the cloud service providers as competitors to software vendors who only developed multi cloud platforms. These third party vendors will now be forced to come up with some greater value additions to be in business. Secondly, organizations are expecting the CSP’s to be their one stop solution provider for all their transitions and management thereafter, which is a blessing in disguise. On one hand it is an added opportunity, while on the other hand this means ramping up the IT skillset and manpower along with various technology tools to assist the businesses embrace multi cloud models and get the best of all clouds. As glittery as it looks, CSP’s are under pressure to bundle different solutions from different cloud providers and offer it tailormade on customer demand. In doing so it is becoming difficult more than ever to maintain its own unique proposition as highlight for the customer,” he says. 

“The early adoption of cloud was led by organizations lifting and shifting workloads onto the public cloud driven by the need to cut costs and increase availability. The next phase of the cloud is where we see serious enterprise workloads also getting ready to move to the cloud by investing in cloud-native development approaches and architecture. A hybrid multi-cloud approach offers them the flexibility to quickly move or migrate workloads across clouds.”

Seema Kumar, Technical Director, Cloud and Cognitive Software, IBM India/SA

Challenges in Multi Cloud 

The Multi Cloud model is still in its initial stages of evolution. As with every other technology adoption, the nascent stages have its own challenges in terms of adoption, implementation and sometimes even arriving at a nuanced understanding of the technology. 

According to IBM, 85% of enterprises are in a multicloud environment, but only 41% have a multicloud strategy. Further, according to a study by the IBM Institute of Business Value, more than 60 percent of customers lack the tools and procedures to manage and operate in a complex multicloud environment. Hence, it is to be expected that organizations will face challenges early on in their multi cloud journey. 

Each cloud service provider has its own way of doing things, and so expertise in one cloud environment does not necessarily translate to expertise with another. When embarking on their multi cloud journey, some enterprises resort to employing a single-cloud specific workforce that is gradually trained to garner multiple cloud skills. 

This inadvertently results in sub-optimal utilization of not just the talent in the organization, but also of the multi cloud environment. That can only be rectified either with experience or guided training. Therefore, ensuring efficient and optimum use of cloud resources becomes a key challenge in adoption of multi cloud environment, as the justification of adopting this, or any other model for that matter is incumbent upon optimum utilization of resources. 

To tackle this challenge, Alex Li, GM, Alibaba Cloud South Asia suggests, “The members of an organization’s IT staff will need to have a good working knowledge of the cloud platforms that are being used. Not every IT staff member needs an intricate knowledge of every cloud platform, but the IT department should collectively have the knowledge required to support each cloud that is being used.” The organizations should have an in-depth understanding of different cloud providers, especially at the level of architecture. When working with multiple cloud providers, it becomes necessary to understand the similarities and differences of their design architectures, deployment and management models. Only then can an organization make an informed decision when it comes to choosing different cloud providers. 

“Multiplying cloud environments typically means multiplying what’s required to effectively run those environments over the long haul. That’s true both for the internal teams and external resources. The higher the number of IaaS, PaaS, and SaaS solutions leveraged, the higher the number of [in-house] technology skills that must be developed and maintained,” says Prabhakar Jayakumar – Country Director, India at DigitalOcean.

 So, another challenge faced by the CIOs in operating on a multi cloud model is the added complexity in managing IT infrastructure. To tackle the complexity of having to redesigning the network for multiple clouds, Biswajit Mohapatra, Executive Director, IBM Cloud Migration Factory, Cloud Application Services suggests CIOs to have “a unified IT infrastructure strategy and execution across the clouds from security and cost optimization perspective in a multi cloud environment.”

With multiple clouds coming into the picture, effectively managing security and privacy across cloud environments is another crucial aspect of the movement to the multi cloud model. “Organizations that have stringent network compliance requirements and are tied up to specific networking devices, cannot impose the needs to all different cloud providers. How is the same level of security maintained across different cloud providers at each layer as each cloud provider will have its own framework in place, remains a big scar to be patched in the overall multi-cloud strategy,” says Dr. Rajeev Papneja, Chief Growth Officer, ESDS Software Solution Pvt. Ltd. 

A key problem that can come out of a multi-cloud environment is Data Synchronization ability or Interoperability. This includes the ability to manage the connectivity between the cloud environments, application portability and migration challenges across different cloud environments, among other complexities.

“With containers being mainstream, developing containerised applications to a great extent assist with lift and shift of environment from one cloud provider to another. If proprietary platform are being used from any cloud service provider, then in any case the organization is tied up to the vendor. Developing cloudagnostic applications to work on all different platforms using their technologies is not a very viable option either. There is still lot of work that needs to be done before true interoperability can be achieved,” comments Dr. Rajeev Papneja, Chief Growth Officer, ESDS Software Solution Pvt. Ltd. 

“Strategic coherence, Enhanced agility, Radical interoperability, Boundless application and data portability, Expanded ecosystem, Enterprise grade security, Compliance certifications, Established governance model and Improved cloud management are the factors to be taken into account while selecting a multi cloud service provider. Compliance and regulatory requirements, reliability, security and privacy are the most prioritized factors to be considered in a multi cloud adoption journey.” said Biswajit Mohapatra, Executive Director, IBM Cloud Migration Factory, Cloud Application Services.

“One key factor executives decided migrating to cloud was to reduce the cost of managing licenses, and periodic upgrades. Post migration, companies realize that the Cloud Service Providers have incorporated lock-in periods; resulting in additional cost due to inability to address the challenges of Flexibility and Complexity. Migrating to multi cloud solutions addresses this challenge.”

Saighiridhar V V, Senior Director of Engg, TechChefs

USE CASE 

Biswajit Mohapatra, Executive Director, IBM Cloud Migration Factory, Cloud Application Services shares a recent example of a client who is a mid-sized Bank.

The bank is facing multiple business challenges attributed to emergence of Fintech economy. The bank is lagging behind competition, customer retention and revenue is declining.

Creating a cognitive banking roadmap powered by multi cloud adoption strategy is the need of the hour for the bank. Three key use cases identified for quick wins are:

1. The loan origination process at the bank took weeks from initiation to disbursement. This was too long especially with respect to competition

2. Bank’s Anti money laundering systems had a huge scope for improvement.

3. The customer retention rate was a low 76%

A 4-pronged hybrid multi cloud transformation approach was used to provide accelerated and effective value delivery to the Bank on its cloud journey. During the Advisory phase, multi cloud strategy and roadmap was created for the bank performing business, technology and cost value assessments. The roadmap drives one of the migrate, modernize or cloud native application development paths in a hybrid multi cloud environment with a default containerization platform.

Finally, approach to manage the workloads in cloud leveraged advanced monitoring, using intelligent solutions to keep the systems healthy, and identify issues early.

A SaaS product from a public cloud provider was leveraged to have a seamless user experience and gather 360-degree view of customer, another SaaS product from second public cloud provider was leveraged to enable voice first banking and cognitive analytics product from third public cloud provider was leveraged to drive the insights and actionable patterns for corrective and preventive models for customer retention. Anti-money laundering product was deployed in Bank’s private dedicated cloud for proactive detection of fraud in commercial transactions.

All the structured and unstructured data generated was analysed, mined and processed with insightful, pattern-based business models to provide augmented on-demand service to customers. Bank’s on-premise monolith was broken into microservices and seamless integration between on-premise applications, private dedicated cloud and public cloud from 3 cloud service providers was established.

We are beginning to see the results and they look encouraging. The most important part is Bank had to make much lesser capital expenditure compared to traditional or single-cloud models. Without a multi cloud model, it is extremely difficult to implement such complex solutions in a matter of weeks and start seeing the business benefits.

Example 1 

Subramanian Gopalaratnam, CTO, Resulticks shares his multi cloud strategy with an example. 

“At Resulticks, we look at different clouds for the capability each provides. For instance, we use Azure because it can provide in-country or localized data protection. We use IBM SoftLayer because it provides us with more options for cost-effective, team-based solutions for hosting certain types of technology that are still not realisable. This way we benefit from the best of both worlds.”

Example 2

Alex Li, GM, Alibaba Cloud South Asia shares, 

“We at Alibaba Cloud work closely with our customers and partners by offering the proven products and solutions used by Alibaba Group for Indian companies. Alibaba Cloud as a vendor in this case not only provide our customers with technical support, we also advise them on the business aspects of cloud computing, given our success experience with Alibaba Group and the global ecosystem we have been building, ranging from e-commerce, Fin-tech, logistics, media and entertainment and digital marketing.”

Example 3

Saighiridhar V V, Senior Director of Engg, TechChefs

“Given that there are multiple Cloud vendors showcasing services in different domains, Organizations are eager to tap into different cloud services to address their specific requirements. For example: Sys Admins may prefer to manage data on AWS, while engineers may prefer to utilize the Analytics engine in GCP and the CIO may observe the Windows licensing is reasonable on Azure. A mix & match of such an architecture is possible within the multi cloud paradigm.”

Finally…

With the Multi Cloud model coming to the forefront of organizations’ cloud approach, it is not just enhancing the way organizations operate on the cloud, but is also paving the way for the service providers to expand their portfolio. Another aspect of the multi cloud makeover is the changing relationship dynamics between the different service providers and their role in the market. But the challenge for the organizations is to devise strategies to optimally and efficiently use the model, instead of merely jumping on the multi cloud bandwagon for the sake of it.

Related posts

New Relic Named a Cloud Observability Leader by GigaOm

enterpriseitworld

Quantum Announces the Scalar i7 RAPTOR for Data Lakes

enterpriseitworld

HPE Leverages GenAI to Enhance AIOps Capabilities of HPE Aruba Networking Central Platform

enterpriseitworld