Virtualization Technology News and Information
Article
RSS
Velostrata 2018 Predictions: Mass Migrations, Multi-cloud Architectures, Modern Cloud Migration

VMblog Predictions 2018

Industry executives and experts share their predictions for 2018.  Read them in this 10th annual VMblog.com series exclusive.

Contributed by Issy Ben-Shaul, co-founder and CEO of Velostrata

Mass migrations, multi-cloud architectures, and adoption of modern cloud migration solutions accelerate in parallel at record pace

While public cloud in general has been experiencing explosive growth for several years in a row, it has been most leverage by small and medium businesses. With enterprises, on the other hand, using the cloud for experimentation and non-production workloads at the IaaS / PaaS level, or using SaaS applications.

However, towards the end of 2017, the ground has shifted, and my prediction is that during 2018 we will see a large number of enterprises conducting massive migrations of thousands of workloads per enterprise into the public cloud. The rationale behind this exodus from on-premises to the public cloud consists of three realizations:

1.       Opportunity Cost - Enterprises realize that if they can shift their resources away from maintaining and managing massive data-centers into focusing their resources on their core competency, they would be much more competitive in their respective markets, and eventually more cost-effective as well. Clearly, they have to be convinced that the outsourced alternative is stable, secure, and performant. The good news is that the cloud has reached the required levels of maturity and in some cases, has even surpassed the levels they had on-premises. In addition, enterprises have also had the time to validate, experiment and fine tune their processes to operate in the public cloud - and are now ready for the big move.

2.       Pace of Innovation in infrastructure and services - There have been numerous enterprises that followed the above guideline but shifted to hosted data-centers (aka private clouds), vs. moving to public clouds. However, the second important realization that leads to abandoning hosted data-centers is that the pace of innovation is much faster in the public cloud. It's easy to realize that the sheer amount of new technologies and startup innovation is in the public cloud environment. Furthermore, IT talent and engineers are also shifting to the public cloud. This means that an enterprise that has not established its infrastructure in the public cloud will find itself lagging behind their competition (and thus suffering the market drawbacks).

3.       The public cloud brings important operational efficiencies - The availability of countless new services that can be leveraged by the enterprise, the agility to expand or shrink quickly to keep pace with the needs of the enterprise, and the global presence and scale of public clouds that enables enterprise to run workloads anywhere in the world with reasonable latency. Plus, the inevitable advantage of economies of scale that hyper clouds offer, and which will eventually lead to better pricing. All of these ultimately lead to a superior alternative to private or hosted data-centers - both economically and functionally.

With the realization that enterprises will migrate at scale, they will need migration solutions that are optimized for large-scale migration, vs. merely repurposing backup/DR tools for the task. These include:

1.       Minimizing the overhead per server - with 50-100 workloads, spending several hours per server is tolerable. When it comes to 5,000 servers (or more!), it's critical to offer an automated framework the minimizes per-server overhead, such as eliminating things like the installation of replication agents. As an example, agentless solutions tend to save 3-4 hours per server in labor, and with a 5,000-server migration, this can amount to millions of dollars in savings, and the elimination of months' worth of extra cost and time.

2.       Automation and customizations frameworks - Being able to migrate complex, multi-tier applications with network and security dependencies - this includes tools that capture and map complex configurations to cloud equivalent configurations.

3.       Maximizing the speed of migration - With a small number of migrations, speed is not a big factor. With thousands of migrations at stake, a fast migration solution means reducing large scale migration projects from years to months.

My second prediction is that the majority of enterprises that will migrate to cloud at scale, will employ a multi-cloud strategy. More specifically, they will split their production workloads across more than one public cloud. At first thought, this might sound non-intuitive, given the learning curve and large effort involved in migrating to one public cloud - why bother with multiple clouds from day one?

The main rationale behind this decision is that enterprises are terrified by the concept of being locked-in for several reasons:

1.       Cost - Cloud wars will result in cost wars. If an enterprise can get significant cost reduction on infrastructure, this can mean millions of dollars in savings a year.

2.       Functionality - Different clouds offer different innovation and functionality. Enterprises would like to use best of breed for the different workloads to take advantage of what all clouds have to offer. 

3.       Compliance - Some industries (such as finance and healthcare, for example) are mandated to have an alternative cloud to run on. Others just want to make sure they can switch in case of security breaches, outages, or other issues that might affect a specific cloud. 

In order to realize the great potential behind a multi-cloud strategy, enterprise customers will need to plan ahead for their cloud deployment and architect the workloads for cloud mobility. These include:

1.       Minimize or avoid cloud-specific services - These would effectively lock you in to the cloud. Instead, look for services that are available on multiple clouds and require minimal migration effort. Or, consider the impact of keeping the service running in its original cloud while running the related workloads on another cloud.

2.       Leverage cloud mobility technologies and architectures - employ techniques that allow for the simple and fast mobilization of workloads across clouds with minimal disruption to downtime or re-configuration/re-write efforts.

3.       Unified, cloud-agnostic management and configuration - Use orchestration and management frameworks that interface to the native cloud management systems but operate uniformly across clouds.

And, with that, these are my predictions for the public cloud, cloud migration, and multi-cloud in 2018. We're excited, as always, to see what 2018 (and beyond) brings.

##

About the Author

Issy Ben Shaul 

Issy is the CEO and Co-Founder of Velostrata and a serial entrepreneur. Prior to co-founding Velostrata, Issy was the CTO for VMware's Mirage desktop virtualization product. Issy joined VMware through the acquisition of Wanova, a desktop virtualization company he co-founded, and acquired by VMware in 2012. Prior to Wanova, Issy was CTO for the Application Delivery Business Unit at Cisco. Issy joined Cisco in 2004 via the acquisition of Actona Technologies, where he was Co-Founder, CTO and Vice President of Engineering. Before founding Actona, Issy was a tenured faculty member at the Technion, Israel Institute of Technology, where he worked on design and implementation of wide-area distributed systems. Prior to the Technion, he was a staff member at the IBM Haifa Research Laboratory. Issy has published over 40 papers and holds 15 patents in the areas of WAN optimization and distributed systems. He holds a PhD and master's degree in computer science from Columbia University and holds a bachelor's degree in mathematics and computer science from Tel Aviv University.

Published Tuesday, November 14, 2017 7:21 AM by David Marshall
Comments
There are no comments for this post.
To post a comment, you must be a registered user. Registration is free and easy! Sign up now!
Calendar
<November 2017>
SuMoTuWeThFrSa
2930311234
567891011
12131415161718
19202122232425
262728293012
3456789