>>14261393Research from Lake Vostof in Antarctica is proving that the half-life of DNA is roughly 1 million years when properly iced. 1 gram of DNA can store something to the tune of a petabyte of data. The only limitation currently is the read speed from that data pool, which currently is something like 1KB/s. Which is basically 0 when dealing with peta and exabyte scale storage volumes. What will define DNA computing to be a potential replacement of classic storage mediums is if we can get read/write bandwidth up to 2020 PCI Gen5 standards to the tune of 32GB simultaneous read/writes per second or 64GB/s one way. That will probably take the better part of the next decade, but even so that much bandwidth is sufficient for 90% of all general purpose applications. A 4K MPEG-2 raw is roughly 80GB. So it would take you 1.5 seconds to read that from a DNA storage medium if you can achieve 64G/s read.
Even if you went out a decade to 2030, the probability of 8 or 16K being common videos sources is statistically low as most of the market will finally have transitioned to 4K base. All that said, nanotech's evolution over the next decade can easily augment the limitation of low i/o potential that is currently the bottleneck.
