I can't see that ever happening. It would cost a fortune, and be a logistical nightmare. What on earth would they do with the ECI DSLAM's they removed? Who's gonna pay for all the new Huawei DSLAM's.
I also find the title of that article extremely misleading. It may give better latency, better error correction, more stability. But where does the claim of "better speed" come from. I've never seen a single direct comparison of the 2 DSLAM's. If the ECI performs 10% better than a Huawei then that would more than wipe out any gains G.INP gives the Huawei. ECI cabinets will be getting G.INP eventually anyway. The vectoring argument doesn't wash either, as OpenReach don't deploy it on a large scale.
Hopefully on Tuesday I'll have a direct comparison with my own line. I'm hoping it will be connected to our new Huawei DSLAM, which I will be able to compare with my stats from a few weeks ago with an ECI DSLAM.
I'd confidently say OpenReach won't be deploying any more ECI cabinets. Even existing ECI's which are at capacity are being added to with Huawei's now. There's not the slightest possibility though that OpenReach will swap existing ECI's .