Well, I'm back in the office now - I arrived home about 9:30 last night after heading down to Sydney for last weekend's SMBiT Pro Security and Optimization workshop. The workshop went well with both Dana Epp and Susan Bradley managing to deliver presentations from their offices over Live (sic) Meeting, before Live (sic) Meeting (notice the word "Live" here...) went offline for its regularly scheduled 6 hour weekly maintenance that occurs each Saturday at 10:00 AM AEST - smack bang in the middle of the most popular conference day in Australia.
Now, after a long 3-way conversation with 2 Microsoft folks in the US about the status of Live (sic) Meeting and their 6 hours a week of scheduled offline maintenance (which makes this online service have an availability of around 96%), I was told that "Apparently there is no scheduled downtime, so you are good to go as per last time" so I believed Microsoft. "Last time" refers to the last Sydney session that I confirmed was not going to be affected by the weekly scheduled offline maintenance window, and wasn't affected - that scheduled maintenance window went unused. This one didn't.
Well, after Susan's presentation ended we had a coffee break whilst the Trend guys prepared for their presentation, and during this period, Live (sic) Meeting went offline. Both Singapore and US data centers were unreachable. During our lunch break the servers came partially back online and I was able to get *most* of the remaining content recorded locally (as the "Record to Service" option wasn't available - that part obviously hadn't come back online as yet).
Remember that Live (sic) Meeting is a part of BPOS which stands for Business Productivity (sic) Online (sic) Suite. And Microsoft wants to have us move our clients into their cloud and live with not only 96% uptime, but downtime of 6 hours right through the middle of every Saturday here in Australia.
Well, as I could build a half dozen crappy white box desktops, call them servers and offer a much, much better uptime than Microsoft seems to be able to with all of their resources, I don't know how many of our clients will accept the appalling and unacceptable downtime that BPOS and Live (sic) Meeting offer. Not that I'd do that with crappy white box desktops acting as servers, but I'm just proving a point.
And Steve Ballmer said Microsoft is "all in" with BPOS. If this is "all in" then we'd better all start looking elsewhere for our future OS, application and data hosting providers! :(
Now, to add salt to the wound, Microsoft recently told us all how they were Maintaining High Availability for Microsoft.com. Well, maybe someone from their Microsoft.com team needs to talk to their BPOS team and share some of their secret sauce, as this sort of performance is **DESPERATELY** needed if BPOS is to be taken seriously. 96% uptime for a cloud data and application provider is an utter, utter joke. This is worse than a particular cloud backup provider who didn't understand that RAID isn't a backup technology and lost a significant amount of their clients' data because they didn't understand much about what they were doing.
We'll be looking to Goto Meeting, Goto Webinar, WebEx, DimDim or something else in the future as those applications all have the reliability that Live (sic) Meeting is sadly lacking. if anyone has some suggestions other than these, or has had good/bad experiences with these options, please feel free to comment in here and/or email me privately with your thoughts and experiences.
Also, have a read of Susan's thoughts on this issue...
Regards,
The Outspoken Wookie
3 comments:
I'm not sure how you did your calculation, but I arrived at 96.4% uptime. While not god, not as bad as the 91% you are talking about.
Oops, bad calc. Updated the figures. Still 96% is almost equally shit! :)
That sort of uptime is definitely unacceptable in this day and age!
You might also want to look at Genesys conferencing. I have no idea on pricing...but we've been using it for some time, and it compares favourably to LiveMeeting.
Post a Comment