Virtualization means different things to different people. But viewed across IT infrastructure as a whole, virtualization has been a trend that has been ongoing in IT for decades.
Gradually, more and more areas of IT have become abstracted. Storage, servers, compute, data, networking, and even memory have been part of the virtualization trend.
See below for some of the top trends companies and IT teams are seeing in the virtualization market:
1. Cloud Migration
The ultimate in virtualization is sending data and applications from on-premises to the cloud. And that is driving further forms of virtualization such as data virtualization.
“As cloud storage and cloud computing become more cost-effective and widely available in the market, more organizations are migrating their warehoused data and their data processing to the cloud,” said Jerod Johnson, technology evangelist, CData.
“Cloud migration leads to the unique problem of data silos: data silos. Fortunately, data virtualization solves that problem by connecting disparate applications and systems within an organization’s cloud ecosystem.”
2. Bringing Algorithms to Datasets
A company known as Devron is on the forefront of providing algorithms that are brought directly to datasets as opposed to companies making copies of and centralizing the data to run it through the algorithms.
“The latter approach is the primary cause of data privacy breaches,” said Kartik Chopra, founder and CEO, Devron.
“Virtualization of data is leading to data privacy issues around the world; firms and governments can alleviate these challenges using AI and machine learning.”
3. Data Center Liberation
Today’s users want both logical and physical consolidation with the fewest restrictions or the ability to stack database instances and containers, not machines. The goal is to stack workloads, not operating systems.
Users also wish to eliminate OS sprawl and hardware limitations, while enabling greater capabilities around high availability (HA) and disaster recovery (DR).
At the same time, they seek to consolidate Microsoft SQL Server environments; whether they have taken an unlimited virtualization approach, with all cores licensed for SQL Server Enterprise edition or a per-virtual CPU (vCPU) approach.
And finally, they want to be able to pull an already-consolidated Docker environment into a highly availability framework and enable their stateful containers to fail over between hosts while all persistent data remains intact. This is leading to increased demand for data center liberation.
“Users wish to enjoy top performance, availability, agility and cost savings across any mix of bare metal, virtual, or cloud,” said Don Boxley, co-founder and CEO, DH2i.
“IT professionals should seek abstraction technology solutions that enable them to decouple Windows and Linux application instances, containers and file share from the underlying OS and IT infrastructure.”
Boxley advocates vendor-agnostic infrastructure that enables users to avoid ecosystem lock-in. In doing so, users can enjoy top performance, HA, DR, agility, and cost savings on any mix of bare-metal, virtual or cloud.
This is also manifesting in greater reliance on edge computing.
“By adopting edge virtualization, a business can use smaller-scale virtual machines to process data for any individual source,” Johnson said. “Instead of investing in the creation, adoption, and maintenance of bespoke systems, IT teams can configure these virtual systems to work with any device in any environment.”
4. Cloud Data Warehouses are the Engine of Data Virtualization
Cloud data warehouses are driving significant changes in how companies manage and retrieve data.
As cloud-native technologies, these systems can abstract a variety of data sources behind familiar tables and views.
Snowflake’s Data Sharing, for example, is a marketplace for external datasets. Similarly, Databricks’ Lakehouse is a vendor-agnostic data storage architecture; and Starburst is a next-generation federated database system.
“While today each vendor has its own innovative features, by the end of the decade, we’ll see a definitive convergence towards a feature set that blurs the line between data stored in the warehouse and data the lives outside of it,” said Max Seiden, principal engineer, Sigma Computing.
“SQL will also play a major role here, as it enables both people and systems to query both structured (and semi-structured) data, without caring about the technical details of the underlying systems. These cloud data warehouses will continue to deliver great performance, infinite scalability, enterprise-grade security, and governance controls.”
5. Multi-Cloud Virtualization
Organizations typically have applications in more than one cloud.
“Users are adopting a multicloud approach and will use the cloud that offers the best price and optimal performance for the specific application,” said David McNerney, spokesperson, Virtana.
“That means that we will see increased signs of cloud movement; if one cloud fails to hit price or performance goals, customers will switch to the cloud that is more reliable. With frequent cloud movement and a multicloud strategy also comes the need for cost accounting. Containers and PaS capabilities require the need to allocate costs, and without cost accounting, there is a lack of responsibility and visibility throughout cloud management.”