In this podcast, the Radio Free HPC team looks at a controversial study of hard drive reliability by Backblaze. Henry Newman was so overwrought over their lack of technical rigor that he wrote up a rant in Enterprise Storage Forum. Then Slashdot picked it up and all Hell broke loose.
Of course, Rich and Dan can’t help but fan the fire. The question is: Can you really complain when you use consumer hard drive technology for enterprise applications?
We recently talked with Doug Kothe, who has served in senior management ….
When building large-scale AI GPU clusters for training or inference, the backend network should be high-performance, lossless, and predictable to ensure maximum GPU utilization. This is hard to achieve when using Ethernet for the back-end network. This guide showcases a high-level reference design for an 8,192 GPU cluster, describing how it can be achieved with […]
Featured RSS Feed More News from insideAI News