1 edit
3 recommendations |
to Iscream
Re: CC DisasterI've lurked for a long time in the various CC threads (concerning both the DDoS and the hurricane). I've seen a lot of accusations about anonymous attacks, underhanded business practices, etc.
What I can say is this. I've been a CallCentric customer for a couple years. And I've recently submitted a request to port my DID to Voip.ms.
My reasons are as follows. I will note that, as an overarching point, the tone of IScream's posts has not exactly played well in my mind. In the cold light of posts on a forum, without the subtle nuances of speech, his posts come across as presumptive, arrogant, and dismissive to what I see, as a consumer, as being very very legitimate concerns. But it also was not ultimately a factor I considered in switching. This is more just a global comment that companies should be aware of the representatives they send out to forums, and the tone of the posts that 'official' representatives provide.
1. Georedundancy: I don't really care if I can select multiple servers, or if that is transparent to me, or what not. What I do care about is a provider having in place systems so that there is no single point of physical failure -- or providing substantial mitigation if there is a single point (see below). From everything I've read, CC really did put all its eggs in one basket, without even a cold backup location that could be spun up. Will there be another hurricane of this size to hit NYC? Probably not anytime soon. But could there be a steampipe explosion, an inadvertent sprinkler activation, etc.? Assuredly. It seems like many other providers talked about in these forums do have some sort of redundancy (I've seen Voip.ms, Future9, and others discussed -- a non-exhaustive list for sure).
2. Mitigation of failures: IScream talks about CallCentric being in a "bullet-proof . . . Tier-1 data-center of its own in the telecom building". But as we found out, this "bullet-proof" location did not have diesel generators and relied only on short-term UPS backup. Similarly situated providers and other data centers were able to switch to diesel and continue running despite the loss of mains power. Yes, I've seen the posts about how NYC bans tenants from installing generators, and only allows landlords to do so. But that doesn't seem like it's an excuse, that to me is a sign of additional failure to mitigate -- CC chose to situate itself in a building without industry-standard power protection equipment (aka diesel).
3. Uptime matters: I disagree with IScream, "voice quality" is not the number one concern. Uptime is. Or, if you want to put it another way, there is no voice quality without a powered server.
Ultimately I chose to switch to Voip.ms because they had a few features I found attractive, and their distributed server system makes more sense to me in a period of time when natural disasters (at least appear) to be getting more severe. Living in California, I'm all too aware of the dangers mother nature poses (Earthquake Country!). NYC is subject to hurricanes and terrorists, the midwest to tornados and floods, the west coast to earthquakes and wildfires. But the odds of multiple disasters hitting multiple regions simultaneously seems low.
I sympathize with the employees of CC who have had a rough month, but other providers trucked through. A company's disaster recovery, data backup, etc. plans are only as good as the first time they are tested. CC's failed, other's did not. I hope it's a learning experience for CC, as I think this marketplace could use as many vibrant competitors as possible. |