The shift to cloud computing in enterprises has been largely driven by the physical limits of on-premises storage and compute capacity. Chalan Aras, senior vice-president and general manager of acceleration at Riverbed, noted that the cloud offers users access to “practically infinite” resources. As a result, vast quantities of data—often in petabytes—are now stored in cloud environments, and organizations are increasingly interested in leveraging artificial intelligence (AI) to extract insights from this wealth of information.
However, the data’s location poses significant challenges for AI processing. Under a multi-cloud strategy, data is often distributed across multiple providers, complicating access. Even when all necessary data for an AI project is confined to a single cloud, it may be stored in a region where operating costs, such as power, are high, limiting access to the essential graphics processing units (GPUs) required for AI workloads. Hence, enterprises must confront the formidable task of transferring large volumes of data, which can be prohibitively expensive.
Data transfer can incur costs of up to $80,000 per petabyte in egress fees, even within a single cloud provider. Moreover, such transfers require stringent governance to ensure that the correct data reaches its intended destination intact. Speed is another critical factor; transferring just 1 petabyte over a 10Gbps connection takes approximately nine days. Given the necessity to continually update AI models with recent data—often on a daily basis—the urgency for efficient data transfers becomes more pronounced, despite smaller volumes.
Riverbed is leveraging its 25 years of experience in data movement to address these challenges in cloud environments. Aras described the company’s approach as one of “serving it on a plate,” where data is extracted from storage and optimized for network transfer. In a notable case, an organization needed to move 1 petabyte of data for AI training. The organization’s existing processes projected a 12-day transfer, yet Riverbed completed the task in just three to four weeks, allowing the company to utilize precious GPU time without delays.
In another instance within the financial services sector, a company required the transfer of roughly 30 petabytes of data between clouds. Riverbed managed to complete this migration in just over a month while adhering to required governance standards. Such efficiency highlights the growing need for rapid data transfer capabilities, as organizations navigate a complex mix of on-premises data centers, multiple cloud environments, and various software-as-a-service (SaaS) applications.
While it is feasible to consolidate to a single cloud provider, companies must carefully assess whether one provider can meet all their diverse needs. Aras pointed out that even the largest hyperscalers do not have a presence in every geographical region, necessitating a secondary provider for many businesses. This complexity typically only becomes apparent when large datasets need to be aggregated into one location, a requirement that is increasingly common as AI adoption expands.
As enterprises move towards employing agentic AI—which requires access to information from a multitude of sources in order to provide quick responses—the ongoing movement of data becomes critical. “This is great for users, as they can get very quick answers, but it does require the frequent movement of data,” Aras explained. Until recently, Riverbed primarily focused on facilitating one-time data transfers, such as system migrations from on-premises setups to the cloud. However, the company has pivoted to address the evolving needs of customers seeking to move substantial amounts of data continuously to support their AI initiatives.
This shift underscores a broader trend in the industry: as organizations increasingly rely on AI for decision-making and operational efficiency, the need for seamless and cost-effective data movement becomes paramount. The ability to adapt to these challenges effectively may soon define the leaders in the cloud and AI sectors, making Riverbed’s innovative approach a timely response to a rapidly changing landscape.
See also
Bank of America Warns of Wage Concerns Amid AI Spending Surge
OpenAI Restructures Amid Record Losses, Eyes 2030 Vision
Global Spending on AI Data Centers Surpasses Oil Investments in 2025
Rigetti CEO Signals Caution with $11 Million Stock Sale Amid Quantum Surge
Investors Must Adapt to New Multipolar World Dynamics
