Today’s keynote with Dr. DeWitt starts at 8:15am EDT. Click here to watch today’s keynote live. For information on today’s keynote, titled “Big Data – What is the Big Deal?“, see the SQLPASS.org page on Keynotes.

If you wanna hear about the first part of the keynote, you can see it here.  I introduce the topic of big data there.

So here we are listening to Dr. DeWitt talk about big data and specifically he’s talking about data loss and node failures in HDFS.  And in his usual fashion, this is a very detailed accounting of how the file system performs its redundancy.  I’m not going to go into detail because I really can’t, but I believe this discussion will be available online at the URL above.  However, I will say that the most interesting thing about HDFS is that it gets it redundancy without any use of mirroring or RAID.  The file system itself takes care of everything.

I’m going to break away from talking about the keynote and give an update on the rest on the Summit.  I personally manned the Enterprise Mgmt expert pod in 2-4hr shifts this week, and I fielded dozens and questions about everything from replication and backup, to how to do things in Powershell.  So it was an excellent opportunity for me to either give some users good answers to their questions, or bring their prod boxes down.  Only time will tell which one it was.

There have been lots of vendor parties as usual and so far the most noteable to me was the Idera party.  Now, what makes a noteable party to me?  The food of course!  And Idera had their party at the Tap House and they had these lovely pulled pork spicy sliders. 

Right after the Idera party was the big Summit party at GameWorks.  Frankly I’m not a very big game guy so I’m not usually interested in playing the games… I’m in it for the food.  And the food at GameWorks was horrible.  There was only one thing there was was done any good at all and it was actually done very well.  That of course is the roast beef.  For some reason, while the macaroni was mushy and bland, the salad dressing was bitter, and the mashed potatoes looked like instant, the roast beef on the slicing table was perfectly medium rare, and seasoned well.  I guess the cook doesn’t like that other stuff cause he never bothered tasting them.

I blogged on one of the other days that the best part of the Summit is getting to see our friends that live all over the world, and that still holds true.  And we were even talking yesterday with some of the top-level MS guys that the Oracle conference doesn’t have anything that even resembles this level of community.  If you go to OpenWorld, you’ll be surrounded by people in suits and none of the vendors are really that interested in talking to you unless there’s a real possibility of getting any money from you.  So all of us are pretty glad we’re on the MS side of things, and we’re thrilled to be able to come to the Summit and take part in all of this.  Now, I read everywhere for one event or another where people like meeting new friends, etc, but that really takes on a different meaning here.  Those of us who come every year are so close and we watch out for each other so well, that I think regular conference friends in other technologies would be surprised to see how we relate.  It’s an amazing group of folks we hangout with here… and yes, I’ve actually made some new friends this year that I really think will fold into the group just fine if they keep coming.

Today’s keynote with Quentin Clark starts at 8:15am EDT. Click here to watch today’s keynote live. For information on today’s keynote (“a demo-packed adventure across SQL Server solutions that span traditional boxed software, appliances and cloud”) see the SQLPASS.org page on Keynotes.

 Ok, here we are on day 2 of the PASS Summit 2011.  If you want to know about the first part of the keynote you can see that here. 

Now, I know in my past post that they were starting on appliances and I wasn’t interested in writing about them, but they’re up on stage now pimping the hardware itself and the machines are very impressive-looking machines.  It’s good to be Dell right now to be able to get your hardware on stage like that.  I dind’t notice before but they’re also showing some HP boxes too, so it’s not gonna just be a big Dell-fest.

Talk about a big box though… the new HP has 2TB of RAM, 196 cores, and enough spare drives that “you won’t have to worry if one of your drives fails”.  It’s their new internal cloud fast track, and that’s another blog entirely, but basically the internal cloud offering is this HP box that comes configured with Hyper-V, System Center, and some other cloud management stuff.  It comes basically ready to plug in so once you bring it in the door you’re ready to start building cloud machines.  Now, this is really just a big VM server that you spin-up VM guests in, so don’ t get too excited.  It comes with some specialized mgmt stuff I believe, but overall you could do all of this yourself if you prefer.

Now, the really cool part of this for me is the new ODBC drivers for linux for CDC and Oracle.  What this means is that now you’ll be able to pull data out of the Oracle logs themselves, and put them right into your SQL Server DB.  So you can literally use CDC with Oracle now.

Now Michael Rys is on stage talking about semantic search.  I don’t really know enough about it to put into words what he’s doing, but I’ll just say that it’s really cool.  I know that’s a lame description, but it’s almost the best I can do.  I will describe one thing though.  He ran a semantic search against a folder of pdf docs he had and the semantic search was smart enough to know the general subjects of the books, and only showed him the titles that met his criteria.  There’s a lot that goes on behind the scenes there that I can’t speak to, but I do know enough to get a sense of what a leap that is for search. 

Now we’re talking about SmartTVs.  I’m not sure why yet though.  Oh, I see, they’re on Azure to sync the TV data.  Ok, they’re talking a lot about clouds so that’s it for today.

Today’s keynote with Ted Kummert starts at 8:15am EDT. Click here to watch today’s keynote live. For information on today’s keynote, see the SQLPASS.org page on Keynotes.

I’ve already blogged about some Summit topics here, but I wanted to continue the discussion on this blog.

Ok, during this opening keynote, Ted just announced the official name for Denali and it’s going to be SQL Server 2012 and it’s exact release date is in the first half of 2012.  We’re alredy in CTP 3 so MS has some work to do before now and then.

One of the big themes in Denali is the cloud.  MS is investing heavily in the cloud and until now the message has been very militant.  For the longest time now we’ve been getting the message that we’re all going to be out of a job if we  don’t transfer our skills to the cloud within the next 2-5yrs.  Now they’ve backed off quite a bit and are saying that the cloud is just oging to enhance your current skills and just give you another avenue when it comes to deciding where your data’s going to go.  Frankly I’m glad to see them calm down finally because it was making a lot of people very angry.  Now, just because MS says it that doesn’t make it so, so why was everyone getting so angry?  Well, in a word because of the offering.  They’ve been telling us that we’re all in trouble if we move our skills to the cloud, but their cloud offering is anything but complete.  So when there are such big holes in their offering, people get upset when you tell them their jobs are going away because it makes the scenario impossible.  So anyway, calm down guys, your job isn’t going anywhere.

Ted also just announced that SQL Server will be supporting Hadoop for ultra-large data sets.  They will also be making submissions to Hadoop to make it run properly for SQL Server.  Now, I know what you’re thinking… you’re a SQL Server guy, not a Hadoop guy.  Well, so am I, but it looks like we may have to at least get up to speed at least to a perfunctory level because we’re going to start seeing it in our DBs.  So to get you up and going on Hadoop, have a look at some of these links:

Hadoop Tutorial

Microsoft is getting in to BIG DATA.  One of their new goals going forward is to be able to work with the super-large datasets from many different sources.  A lot of this data is going to be unstructured, and that can be the hardest data to get at so bringing on products like Hadoop will help with that.  Because let’s face it… as DBAs we really want for everything to have a nice queryable structure, but it’s not always stored like that.

Data Explorer is an interesting topic too.  Microsoft has a new Azure Marketplace and apparently they’ll not only offer an app marketplace like we’re all used to seeing, but they’re also going to offer it as an internal service.  What it sounds like this means is that they’ll allow companies to have their own app marketplace that they can use to give their internal people  apps.  Now, this is a new market for sure, but how useful is it?  From a technology standpoint I get where they’re coming from, but I just don’t know how much of a use there is for something like that.  If it does get picked up then it’ll mean a big paradigm shift for pretty much every company in the world.  But this is definitely something to watch out for because it’s an interesting concept.  I remember when MS came out with web services.  They didn’t really know what to do with it, but they stood in front of their customers and said here’s this new technology, now we’d like for you to go see what you can do with it.  I’m waiting to see if this’ll be like that again, and I’m also interested to see if this has any real teeth.