Well, it appears that the problems from earlier, remain unabated despite efforts from Linden Lab's Ops team. Having logged on, I'm faced with much the same problems as when I logged off 8 hours ago.

Sorting through the message inundation that occurs when I log in doesn't indicate that things got any better today, though eyeballing the concurrency curves for the day, there seems to have been a period where things were working relatively smoothly, despite the lack of search and recurring issues with routing or filtering at Level 3.

There are still issues with transactions, search, teleportation, group IMs, and assets. Technology is awesome!

Frontier Linden advises that in order to address issues with the grid (we presume that these are among the issues to which they are referring) Linden Lab has shut down Search functions for Places and Groups as of 1:40AM SLT (US Pacific).

Additionally, many European users (all the affected ones seem to be routing through Level 3) are once again unable to access Second Life web servers or the Second Life grid. The first major occurrence of that was on Friday during the time that the Support Portal was offline, then a protracted period on Saturday, and now we are informed, that they have been locked out since some time before midnight Saturday, and that those connectivity problems are ongoing.

[Update: 5PM SLT - Search is back on, having been off for the busiest day of the week - but no word from Linden Lab as to whether the problems that they were trying to solve are themselves resolved]

Certainly some of the recent peak, and near-peak load problems we've seen in Second Life are related to data connectivity between the two Second Life colocation facilities (San Francisco and Dallas). It's no mystery to anyone who has been following the trials and tribulations over the last four or five months that the data paths between the two facilities have been an ongoing source of woe.

Enter data carrier Level Three, who have fat data pipes all over, that sport 10 gigabit per second pipes and tier 1 infrastructure. Level 3 are a big data carrier, and yesterday it was announced that Linden Lab has chosen them to carry data between the two colocation facilities. Could this spell an overall improvement to grid scalability? Or is this just planning for the future and we're not presently choking? Unless someone asks Cory tomorrow, I doubt we'll have the answers very soon.

It is not presently clear as to whether the cutover will be seamless, bumpy or have to take place during a maintenance downtime window.