When data gravity is a factor, what is a best practice for architecture placement of compute and storage?

When data gravity is a factor, what is a best practice for architecture placement of compute and storage?

• A. Move data to where compute resides
• B. Remove data to avoid gravity
• C. Place compute near the data
• D. Data gravity has no impact

Answer: (C)

Data gravity describes how large datasets pull workloads toward them because moving that data is expensive in terms of bandwidth, latency, and cost. When data gravity is a factor, the best architectural move is to place compute near the data so processing happens where the data resides. This minimizes cross-network transfers, reducing latency and bandwidth usage while lowering data-egress costs, which is crucial for analytics and big data workloads. Trying to move the data to the compute location would still incur substantial data transfer overhead as datasets grow, making it less scalable. Suggesting that data gravity has no impact ignores real constraints in cloud environments, and simply removing data isn’t a practical or scalable solution.

Data gravity describes how large datasets pull workloads toward them because moving that data is expensive in terms of bandwidth, latency, and cost. When data gravity is a factor, the best architectural move is to place compute near the data so processing happens where the data resides. This minimizes cross-network transfers, reducing latency and bandwidth usage while lowering data-egress costs, which is crucial for analytics and big data workloads. Trying to move the data to the compute location would still incur substantial data transfer overhead as datasets grow, making it less scalable. Suggesting that data gravity has no impact ignores real constraints in cloud environments, and simply removing data isn’t a practical or scalable solution.