Scality Predictions for 2019 and Beyond

Scality, Inc.‘s CTO Giorgio Regni gazes into his crystal ball to reveal his forecast for the year ahead, drawing on trends in cloud, AI, edge computing and open source.

There will be cloud repatriation
The mobility of workloads and data will increase in importance as organizations that have leapt into the public cloud are now beginning to understand the reality of its cost, complexity and potential for lock-in. Enterprises will not repatriate the entirety of their workloads back to on-premises though, as the value that cloud services add in terms of agility, flexibility and functionality remains significant. Organizations will change their perspective from being ‘all-in with cloud’ to wanting the ability to move between cloud and on-premises. Even staunch mono-cloud defender AWS has now announced their on-premises offering called Outpost.

The new norm will be visibility of resources across on-premises and public clouds, independent of location, in addition to the ability to move resources quickly and efficiently to the right location on-demand, based on business and/or technical requirements.

Cloud providers will continue to make acquisitions
to build offerings in the media and entertainment industry
The past few years have seen many cloud providers invest heavily in solutions aimed at the media and entertainment (M&E) industry; for example, AWS’s acquisition of Elemental in September 2015, Google’s acquisition of Avanto in July 2016 and Azure’s purchase of Avere in January 2018.

M&E organizations create substantial volumes of digital content and high-definition programming, generating enormous quantities of CPU/GPU (central processing unit/graphics processing unit) cycles. The sheer size of storage capacity required by the M&E industry could potentially make it a massive consumer of cloud-based solutions and the associated services, ranging from archive and encoding/transcoding to content distribution and artificial intelligence.

Additionally, there is now a trend of telco and network providers acquiring content companies, bringing content catalogues and massive communication and media distribution networks together into potential media megaliths. In June 2018 AT&T acquired Time Warner, and in September 2018 Comcast acquired SkyTV in a blind auction bid.

So what will 2019 and beyond bring? AWS, Azure, and Google will continue to make major acquisitions to build out their M&E portfolios. Amazon (and by extension AWS) will continue to move into the content creation and delivery space, blurring the lines between cloud, network and content services. The balance of power in the content creation and distribution race will continue to shift from Hollywood to Silicon Valley and Seattle.

Digital transformation trend continues transition to cloud-native
2019 will reveal the transition of applications from monolithic to modular or micro services. Many of these applications still store data via file-based interfaces, but the progression to RESTful-based object APIs will accelerate for the stateless nature of micro-services. When this changeover happens, it will help simplify data management and reduce application code at scale. Solutions with both native file and object interfaces are uniquely positioned to handle the transition from traditional monolith to new architectures of modular/micro services under one platform. We will see digital transformation and cloud native application deployments growing at ten times the rate of traditional applications. Amidst this transitional growth, driving down cost and complexity will continue to be central to success.

AI storage-hungry, but you cannot use what you cannot find
The ability to aggregate and orchestrate data generated by AI and machine learning (ML) across a variety of sources will be critical. Meeting that need, robust scale-out storage solutions will fuel the momentum of AI and ML by paving the way for storing and processing the vast volumes of unstructured data that they generate. Metadata search, which helps users locate relevant information and spans a single, navigable namespace and automatic metadata tagging, will emerge for its cornerstone capabilities.

‘Edge’ continues to grow in size, and importance
Consolidation, IoT and the working from home (WFH) trend are leading to an overall growth of ‘edge’ cases. Careful planning is required for the storage and protection of data that is used, modified and/or generated in edge and ROBO environments or other distributed locations. As remote data collection and generation become more prevalent, centralized secure storage and high-performance networks are becoming ever more critical since data is pulled from remote to central datacenters. This will be the case whether the data is retrieved via public or private cloud, or the more straightforward datacenter route.

Geo-distributed storage with efficient resiliency models that exploit advantages such as those found in cross-geo erasure coding (such as durability against datacenter failure and reduced storage costs), will make data safer, and also more readily available whenever and wherever needed by authorized users.

Centralized backups and recoverability will require solid network infrastructure, as well as resilient and efficient central storage to make performing backups and retrieving backed-up data painless for remote users and central administrators alike.

When it comes to valuable data, economies (of OPEX as well as of CAPEX) have to be strong at the data collection, retrieval and long-term retention stages. Centralized IT teams will look for low-overhead distributed solutions, while overall business value will continue to be key in the years ahead.

‘Cloud’ becomes ‘clouds’
Multi- and hybrid clouds will become the new ‘cloud’ as the awareness of data as a major enterprise asset continues to increase. Today’s common model is a hybrid IT approach: organizations selectively and consciously choose to use their on-premises IT resources for specific tasks, while moving other workloads and data to a public cloud, or clouds.

Recognizing that one single cloud does not necessarily offer the most effective option across the board, enterprises will continue to use multiple public cloud services across their applications and workloads. Vertical specialization of cloud services and regional services will make it commonplace for enterprises to have workloads and data managed across multiple clouds, both public and private. Such a move will require a solution that simplifies and automates data workflows across the multi-cloud environment.

Open source is enterprise requirements list
Enterprises have been embracing open source on account of its two key virtues: encouraging and accelerating innovation, and providing an exit route from vendor lock-in. Open source has gained even more enterprise credibility in 4Q18, IBM’s acquisition of Red Hat being a key example. Organizations will continue to evaluate products of all kinds in terms of their features, functions, support and roadmaps; but open source will move from fringe to RFP as enterprise development teams look to collaborate with developers and, if necessary, take the lead in managing their mission-critical applications. More acquisitions are likely to take place over the next couple of years as vendors strive to add open source to their competencies in a bid to boost revenue.

With Red Hat now acquired and Ubuntu looking for external investors, it is looking as though, finally, the Linux desktop will no longer be the main topic of discussion.