http://www.sqlmag.com/authors/author/author/5086182/rsscomment/5086182en-USFri, 25 May 2012 10:47:14 GMTFri, 25 May 2012 10:47:14 GMThttp://www.sqlmag.com/article/sql-server/training-sql-server-newbies-avoid-basic-database-mistakes-143183#commentsAnchorThu, 24 May 2012 21:39:39 GMT
As database technolgy matures and settles out more (as it seems to be doing), perhaps more time and effort can be spent on training ofr good technique and system performance and less on learning the next new version or feature set. Many of the changes have been great, but there are limits to leanring new things. ]]>sjcomeauThu, 24 May 2012 21:39:39 GMThttp://www.sqlmag.com/article/sql-server/training-sql-server-newbies-avoid-basic-database-mistakes-143183#commentsAnchorhttp://www.sqlmag.com/article/sql-server/training-sql-server-newbies-avoid-basic-database-mistakes-143183#commentsAnchorThu, 24 May 2012 10:24:37 GMT
Great point! Thanks for highlighting this, as it is one of those "can't see the forest for the trees" issues. Vendors could do a great deal to solve this by coming up with "maintenance plans" which actually take real life into account, rather than just orchestrating the regular flipping and flopping of bits. ]]>anyoneisThu, 24 May 2012 10:24:37 GMThttp://www.sqlmag.com/article/sql-server/training-sql-server-newbies-avoid-basic-database-mistakes-143183#commentsAnchorhttp://www.sqlmag.com/article/commentary/shorter-sql-server-conference-sessions-142641#commentsAnchorSat, 24 Mar 2012 19:21:43 GMT
I'm with you, Brian -- 60 minutes is just too long for any presentation other than a class or workshop. Thirty minutes I can do, like you said, once or maybe even twice a week. I've lost count of the stack of links to presentations (most of which are by now probably invalid!) and even white papers that I've promised myself I would attend to tomorrow... Go ahead and start that Facebook page, I'll "like" it as soon as I hear about it. Cheers, Michelle]]>GoldenGalSat, 24 Mar 2012 19:21:43 GMThttp://www.sqlmag.com/article/commentary/shorter-sql-server-conference-sessions-142641#commentsAnchorhttp://www.sqlmag.com/article/commentary/shorter-sql-server-conference-sessions-142641#commentsAnchorFri, 23 Mar 2012 16:26:16 GMT
Instead of altering the 60 minute presentation, alter how you view it. The answer for me, is to skip the "live" version, and watch it later using software that allows me to watch it in 2x speed. http://www.enounce.com With that, I can watch a 60 minute training in 30 minutes :) (Once you get used to the software, 2.5 x is actually quite comfortable). If the content gets too thick/gooey, just slow down. Thanks for the article though, I do agree that "I will do that tomorrow" almost never happens.]]>TrogdorFri, 23 Mar 2012 16:26:16 GMThttp://www.sqlmag.com/article/commentary/shorter-sql-server-conference-sessions-142641#commentsAnchorhttp://www.sqlmag.com/article/commentary/shorter-sql-server-conference-sessions-142641#commentsAnchorFri, 23 Mar 2012 08:24:26 GMT
I don't think I agree. We can sit through a movie that's 90-120 minutes, why not a technical session of 60 minutes. The other issue is that, yes, there are times I think a presenter could have moved faster or omitted some part of his/her materials, but we also have to remember that not everyone in the room has the same knowledge and experience so someone may need that information I think could be skimmed over or omitted. Perhaps offering a combination of different length sessions might work as well.]]>unclebigunsFri, 23 Mar 2012 08:24:26 GMThttp://www.sqlmag.com/article/commentary/shorter-sql-server-conference-sessions-142641#commentsAnchorhttp://www.sqlmag.com/article/commentary/shorter-sql-server-conference-sessions-142641#commentsAnchorThu, 22 Mar 2012 18:23:52 GMT
I agree with you, Brian. Most speakers could present in 30 minutes. You could also allow speakers with a lot of real content to hold two consequtive sessions with a break. This would be flexible.]]>Andrew ZanevskyThu, 22 Mar 2012 18:23:52 GMThttp://www.sqlmag.com/article/commentary/shorter-sql-server-conference-sessions-142641#commentsAnchorhttp://www.sqlmag.com/article/commentary/shorter-sql-server-conference-sessions-142641#commentsAnchorThu, 22 Mar 2012 15:39:40 GMT
The conference session was founded around the length of a University lecture (and a high school period) - because trial and error found that 50 minutes was the longest amount of time that you could deliver information before people became distracted. The big change is that many people no longer have the attention span they used to have. Watch the average millenial try to concentrate on the same subject for 50 minutes without doing six other things. 30 minutes is probably optimistic. 15 minutes is probably more realistic.]]>Orin ThomasThu, 22 Mar 2012 15:39:40 GMThttp://www.sqlmag.com/article/commentary/shorter-sql-server-conference-sessions-142641#commentsAnchorhttp://www.sqlmag.com/article/sql-server/apples-ibooks-2-revolutionize-sql-server-book-training-market-142041#commentsAnchorSat, 28 Jan 2012 15:48:55 GMT
Would I buy a technical book that wasn't offered in an electronic format? Probably not. Nearly every book offered is now bundled with an electronic version (MOBI, EPUB or PDF). If the book isn't offered in that format, I need a compelling reason to buy that particular book. I enjoy being able to have my entire technical library on a USB thumb drive, my various eReaders and my DropBox account. I want the same experiece as my music collection: if I want access to something, I have multiple ways of accessing it. The real value I find in books is not necessarily the content. I can find almost everything I've ever wanted to know through search engines. Most of the books I've read are restatements of the vendors own documentation (How many SQL books are word-for-word restatements of Books Online? Why buy a book that uses the various AdventureWorks databases as their sample? I may as well surf the examples at msdn.microsoft.com and technet.microsoft.com). The value I find in books is organization and logical presentation of the information. Will iBooks 2 enhance that ability and experience? I'd like the graphics in eBooks to render better on my eReaders. Do I see the value in an embedded video presentation? Not unless the annotation capabilities let me bookmark and and add notes to the video; something I haven't seen done well yet. Do I want sound in my books? I might, but the person next to me on the train/plane/bus doesn't. Does that mean that industry conferences will die or take a new form? I don't think so. I get invitiations to hundreds of "webinars" a year. Ask me how many I actually have the time to view or participate in. If you guess NONE, you are right. When in the office or at home, there will be something pressing that needs my attention. The real value of a conference is the time reservation. If I'm physically at a conference... across town or across the country, chances are I will be more likely to participate.]]>Marc JellinekSat, 28 Jan 2012 15:48:55 GMThttp://www.sqlmag.com/article/sql-server/apples-ibooks-2-revolutionize-sql-server-book-training-market-142041#commentsAnchorhttp://www.sqlmag.com/article/quering/estimating-query-costs#commentsAnchorMon, 21 Nov 2011 05:27:39 GMT
but at least is it possible to map estimated plan (its parts) to currently running query? for example i have a simple query has been running for 5 minutes already. And i have its estimated plan with 3 main parts like clustered index seek for one of join tables , table scan for the second join table and nested loops join operation that works with two inputs and outputs matching rows. can i figure out somehow what exact part of the plan SQL Server query executer is running currently - at this minute- for example, is reading data pages #n-#m through clustered index or/and matches key rows #x-#y. or my questions do not make sense ?]]>kimisMon, 21 Nov 2011 05:27:39 GMThttp://www.sqlmag.com/article/quering/estimating-query-costs#commentsAnchorhttp://www.sqlmag.com/blog/leap-think-54/sql-server/entrepreneurs-start-planning-key-hires-140767#commentsAnchorThu, 13 Oct 2011 13:33:20 GMT
Great reminder! Most project managers will tell you that picking a project team can make or break a project. Careful thought should be given to each candidates technical skills, work habits and social characteristics as the apply to that particular project. Unfortunately, this aspect of project planning is often rushed or directed without due consideration. ]]>Mitch BarbeyThu, 13 Oct 2011 13:33:20 GMThttp://www.sqlmag.com/blog/leap-think-54/sql-server/entrepreneurs-start-planning-key-hires-140767#commentsAnchorhttp://www.sqlmag.com/blog/leap-think-54/development/business-140807#commentsAnchorFri, 07 Oct 2011 09:29:43 GMT
Isn't the "You're fired" a perfect a perfect exmaple of that? You may work with a great guy but someone who is technically incompetent. After many attempts to bring them up to speed they continue to be incapable of doing their job efficiently. Having to let them go isn't a personal choice, but rather a business decision. In that scenario I would think the phrase could help put that into perspective?]]>MishbakerFri, 07 Oct 2011 09:29:43 GMThttp://www.sqlmag.com/blog/leap-think-54/development/business-140807#commentsAnchorhttp://www.sqlmag.com/article/Performancetuning/ssds-performance-tuning-experts-milkman-139591#commentsAnchorFri, 02 Sep 2011 15:01:32 GMT
Some queries execution times are growing exponentially with data size. Performance increase in hardware is no help in these cases. Some other performance problems are contention problems. Hardware doesn't help much but usually reduces contention, because individual transactions are faster. Some other performance problems are not on hardware at all! I mean human problems related to poorly understandable queries. The server get slow but much more the programmer doing maintenance on it! However SSD may get a big speed bumps and reliability and price drop with the advent of ReRam HP promise to delivers with Hynix next year. This Reram is a lot faster and smaller per bit and very reliable. So there is going to be less and less job in performing some actual slow queries, which will not be anymore. But there are going to be more and more monster queries on monster databases. Tanks to BI ! ]]>Maurice PelchatFri, 02 Sep 2011 15:01:32 GMThttp://www.sqlmag.com/article/Performancetuning/ssds-performance-tuning-experts-milkman-139591#commentsAnchorhttp://www.sqlmag.com/article/sql-server/software-patents-encouraging-litigation-innovation-140349#commentsAnchorFri, 26 Aug 2011 09:46:36 GMT
I created a software company 36 years ago with innovative technology protected by trade secret law. Since then we have seen patents issued for many trivial (non-novel) methods. The cost of defending yourself against patent suits, even when you have proof you are right, will quickly put you out of business. We now have to avoid innovation to protect ourselves. There are very rare cases where software patents are justified. I see them only helping attorneys and aggressive companies with deep pockets.]]>larrysFri, 26 Aug 2011 09:46:36 GMThttp://www.sqlmag.com/article/sql-server/software-patents-encouraging-litigation-innovation-140349#commentsAnchorhttp://www.sqlmag.com/article/sql-server/future-paid-sql-server-conference-model-140001#commentsAnchorSat, 06 Aug 2011 15:16:56 GMT
For about a decade our company no longer pays for conferences. I have had to be way more clever and proactive in finding resources to help me along the way. Free webinars from PragmaticWorks and attending my local SQL Server Users Groups (West Michigan and Detroit) have helped to build skills and to network. I was, however, able to successfully "lobby" management to book me a hotel and reimburse me for around 400 miles of round trip driving to attend one of the SQLSaturday events, which was well worth it. It was on my time (driving and attending) but at least the company met me part way. Few companies, in my estimate, are willing to fly employees for an entire week, put them in a hotel, and pay registration for big events any more. The successful events then, will either be sponsore, like SQL Saturday, or shorter, more regional events where someone may be able to drive and stay for 2-3 days instead of five.]]>GSquillaceSat, 06 Aug 2011 15:16:56 GMThttp://www.sqlmag.com/article/sql-server/future-paid-sql-server-conference-model-140001#commentsAnchorhttp://www.sqlmag.com/article/sql-server/future-paid-sql-server-conference-model-140001#commentsAnchorMon, 01 Aug 2011 18:58:13 GMT
SQL Solstice has posted a postponement notice on their website.]]>ryanbseattleMon, 01 Aug 2011 18:58:13 GMThttp://www.sqlmag.com/article/sql-server/future-paid-sql-server-conference-model-140001#commentsAnchorhttp://www.sqlmag.com/article/sql-server/future-paid-sql-server-conference-model-140001#commentsAnchorMon, 01 Aug 2011 11:48:38 GMT
This is eerily remindful of the discussions that went on about the now-defunct DECUS user group meetings - the meetings helped to pay for the various services to members. The demise of DEC was the main reason for its dissolution, but it had begun waning as new sources of advice and information began to show up on the internet for free.]]>Peter M. KehoeMon, 01 Aug 2011 11:48:38 GMThttp://www.sqlmag.com/article/sql-server/future-paid-sql-server-conference-model-140001#commentsAnchorhttp://www.sqlmag.com/article/Performancetuning/ssds-performance-tuning-experts-milkman-139591#commentsAnchorThu, 21 Jul 2011 23:11:22 GMT
I am tasked with querying multiple data marts with fact tables in the 100's of millions. My query tuning needs are not going to disappear just because of a 10X improvement in physical disk querying. The improvement would probably need to be 3 orders of magnitude, not 1. kj]]>kenambroseThu, 21 Jul 2011 23:11:22 GMThttp://www.sqlmag.com/article/Performancetuning/ssds-performance-tuning-experts-milkman-139591#commentsAnchorhttp://www.sqlmag.com/article/Performancetuning/ssds-performance-tuning-experts-milkman-139591#commentsAnchorSun, 10 Jul 2011 09:41:42 GMT
My Reference point for this argument is asking Paul Randal to run poorly performing cursor based code from http://ask.sqlservercentral.com/questions/826/the-fifo-stock-inventory-sql-problem on his SSD test rig , HDD = 60 mins , SSD = 45 Mins, tuning the code = 3 seconds. I rest my case. ;)]]>Sun, 10 Jul 2011 09:41:42 GMThttp://www.sqlmag.com/article/Performancetuning/ssds-performance-tuning-experts-milkman-139591#commentsAnchorhttp://www.sqlmag.com/article/sql-server/the-10k-server-of-the-future#commentsAnchorThu, 30 Jun 2011 21:40:55 GMT
Dear Kohlmiller, Your data must be very scattered and the access pattern is even more scattered. We use a clustered index based on the primary key (year,ID). We extract the data to SAS BI software and run analysis using SAS. Our GP encounter based dataset uses terminologies and classification systems that suits our analysis (eg. ICPC-2 PLUS) and we don't like free text guesswork. The datasets are arrayed for convenience. The number of columns read is reduced prior to further analysis. 1M encounters, 6GB. We only focus extraction of data for the current year with previous years static in SAS. ]]>tygrusThu, 30 Jun 2011 21:40:55 GMThttp://www.sqlmag.com/article/sql-server/the-10k-server-of-the-future#commentsAnchorhttp://www.sqlmag.com/article/Performancetuning/ssds-performance-tuning-experts-milkman-139591#commentsAnchorWed, 29 Jun 2011 20:36:31 GMT
I agree with John on "I personally believe the database design, the application code and accompanying SQL has much to do with what we are seeing." I would like to add that It's in the Application and its interaction with the database where most significant performance improvements can be made after a DBA has dealt with poor indexing, blocking queries, etc. Adding a faster disk subsystem does improve performance significantly but it has a limit that sometimes is not enough to satisfy customer needs.]]>Wed, 29 Jun 2011 20:36:31 GMThttp://www.sqlmag.com/article/Performancetuning/ssds-performance-tuning-experts-milkman-139591#commentsAnchorhttp://www.sqlmag.com/article/Performancetuning/ssds-performance-tuning-experts-milkman-139591#commentsAnchorSat, 25 Jun 2011 08:38:38 GMT
Bad software (table/index design, TSQL) beats good hardware. That will never change. On this we agree. With each boost in performance gain comes the new "gotta have" feature requiring more consumption of even greater amounts of resources. It is a cycle that will only end when man's curiosity and need to know ends.]]>ajsongySat, 25 Jun 2011 08:38:38 GMThttp://www.sqlmag.com/article/Performancetuning/ssds-performance-tuning-experts-milkman-139591#commentsAnchorhttp://www.sqlmag.com/article/Performancetuning/ssds-performance-tuning-experts-milkman-139591#commentsAnchorFri, 24 Jun 2011 08:15:48 GMT
There is also another solution utilizing flash memory that is a better alternative than SSDs. Attaching an SSD to the server, or even in a SAN still limits the access speed to the speed of the Disk I/O channel. Consider that the PCI bus in servers can transmit data at many times the speed of Disk I/O. There is at least one company out there (FusionIO) that provides flash memory devices that plug directly into the backplane of a server to provide up to terabytes of flash memory "disk" storage. The value proposition to solving I/O related performance problems with a solution like this is not just the time saved paying tuning experts, but in scaling up the hardware solution to achieve the required I/O speeds, and avoiding throwing massive amounts of hardware at the problem.]]>Clayton GroomFri, 24 Jun 2011 08:15:48 GMThttp://www.sqlmag.com/article/Performancetuning/ssds-performance-tuning-experts-milkman-139591#commentsAnchorhttp://www.sqlmag.com/article/Performancetuning/ssds-performance-tuning-experts-milkman-139591#commentsAnchorFri, 24 Jun 2011 06:04:27 GMT
There's a bigger picture quite apart from performance tuning. So much of the way SQL DBMSs work and what we do with them is based around the needs of spinning disk, especially the need to minimise disk seek times and favour sequential access over random access. Expect a radical replumbing of DBMS internals as more customers move onto SSD. Not only indexing strategies but the whole concept of how data maps into storage should start to change. Some consequences might be new levels of abstraction, support for richer data types and the increasing obsolescence of some old concepts and assumptions. For example the fallacy of "denormalization for performance", the balance of tradeoffs between storage and computation, artificial sequential keys, physical partitioning, clustered indexes, etc.]]>dportasFri, 24 Jun 2011 06:04:27 GMThttp://www.sqlmag.com/article/Performancetuning/ssds-performance-tuning-experts-milkman-139591#commentsAnchorhttp://www.sqlmag.com/article/Performancetuning/ssds-performance-tuning-experts-milkman-139591#commentsAnchorThu, 23 Jun 2011 21:42:38 GMT
It depends. We're testing on a database server (Windows 2008, SQL 2008 R2, 110 GB RAM, 2 - 6 core CPUs, SSD disk storage). The third-party ERP we run for our particular industry is running less robustly there than in the current prod environment (Windows 2003, SQL 2005, traditional disk, 80 GB RAM, 4 quad CPUs). I personally believe the database design, the application code and accompanying SQL has much to do with what we are seeing. ]]>John LangstonThu, 23 Jun 2011 21:42:38 GMThttp://www.sqlmag.com/article/Performancetuning/ssds-performance-tuning-experts-milkman-139591#commentsAnchorhttp://www.sqlmag.com/article/Performancetuning/ssds-performance-tuning-experts-milkman-139591#commentsAnchorThu, 23 Jun 2011 17:56:32 GMT
Brian, you have a point when you say "this is the first time Ive started to see a pattern of throwing hardware at the problem being a decent answer." Hardware, not only SSDs, is really starting to become more than just utility components of the DBMS. Processors that encrypt on-board data are available at very reasonable prices -- and these could be the answer to your "data @ rest encryption" requirements. I think we're going to see more of these type of building blocks for all DBMSs. And, as for the quintessential question...does a tree falling in the forest still produce pressure waves in the air, sound will happen as soon as those waves hit a receiver (an ear, for instance). ;-)]]>GoldenGalThu, 23 Jun 2011 17:56:32 GMThttp://www.sqlmag.com/article/Performancetuning/ssds-performance-tuning-experts-milkman-139591#commentsAnchorhttp://www.sqlmag.com/article/microsoft-sql-azure-database/database-cloud-computing-cloudy-chance-meatballs-135994#commentsAnchorThu, 05 May 2011 16:54:32 GMT
I wish "cpagani" would elaborate with implementation examples and/or case studies.]]>ryanbseattleThu, 05 May 2011 16:54:32 GMThttp://www.sqlmag.com/article/microsoft-sql-azure-database/database-cloud-computing-cloudy-chance-meatballs-135994#commentsAnchorhttp://www.sqlmag.com/article/microsoft-sql-azure-database/database-cloud-computing-cloudy-chance-meatballs-135994#commentsAnchorThu, 28 Apr 2011 13:47:49 GMT
Quote "I usually like to avoid writing about things that I dont know too much aboutespecially when Im being paid to write it!" Sorry Brian, You should heed your first sentence. Cloud database has great uses today, especially for companies with globally distributed locations!]]>cpaganiThu, 28 Apr 2011 13:47:49 GMThttp://www.sqlmag.com/article/microsoft-sql-azure-database/database-cloud-computing-cloudy-chance-meatballs-135994#commentsAnchorhttp://www.sqlmag.com/article/microsoft-sql-azure-database/database-cloud-computing-cloudy-chance-meatballs-135994#commentsAnchorWed, 27 Apr 2011 21:00:19 GMT
Wow! Excellent article. I agree with each and every sentence! You have accurately portrayed the state of the cloud today. But I think that the cloud will very rapidly evolve to what it should really be and so will our notion of it. You use the phrase short term but you do not quantitatively define it, which is OK, of course. I personally expect this short term to be less than 2 years.]]>Dimitrios KalemisWed, 27 Apr 2011 21:00:19 GMThttp://www.sqlmag.com/article/microsoft-sql-azure-database/database-cloud-computing-cloudy-chance-meatballs-135994#commentsAnchorhttp://www.sqlmag.com/article/sql-server/the-10k-server-of-the-future#commentsAnchorTue, 12 Apr 2011 12:01:22 GMT
If a $5k new laptop can run on 1 TB of SSD today, why wouldnt it be possible to run a $10k database server on it today??]]>ComfiTue, 12 Apr 2011 12:01:22 GMThttp://www.sqlmag.com/article/sql-server/the-10k-server-of-the-future#commentsAnchorhttp://www.sqlmag.com/article/sql-server/the-10k-server-of-the-future#commentsAnchorFri, 25 Mar 2011 11:55:08 GMT
Great Post Brian.

The one thing that I think will continue to grow with processing power is demand for storage. They seem to stay neck and neck - to the point where I’m not sure performance tuning will go by the wayside.

In other words, today’s hardware would have made yesterday’s performance problems all look like a complete joke. Yet, we still have GOBS of performance problems today - because demand and dependency have increased correspondingly.

Then again, maybe that’s just my agenda (as a performance tuner) speaking ;)

]]>Michael K. CampbellFri, 25 Mar 2011 11:55:08 GMThttp://www.sqlmag.com/article/sql-server/the-10k-server-of-the-future#commentsAnchorhttp://www.sqlmag.com/article/sql-server/the-10k-server-of-the-future#commentsAnchorThu, 24 Mar 2011 19:13:00 GMT
Back in the 1980’s I worked on an optimizing compiler. We fought to get every millisecond of performance. Now it is a valueless skill - one of many I’ve collected over the years. But today I’m working on a brand new server and staring at a query that took 5 hours before I killed it and the query analyzer has been chewing on it for 2 more hours. We use data from hospitals and we see 1 million records for one table every day and that is still not capturing everything. I’m afraid a 1TB SSD will be a nice little C: drive but where’s my petabyte SAN with a 5Tb cache on the controller?]]>KohlmillerThu, 24 Mar 2011 19:13:00 GMThttp://www.sqlmag.com/article/sql-server/the-10k-server-of-the-future#commentsAnchorhttp://www.sqlmag.com/article/training-and-certification2/improving-the-sql-server-training-ecosystem#commentsAnchorTue, 15 Mar 2011 20:09:13 GMT
In person training can be incredibly valuable, particularly when it is with an expert who can engage in an interesting discussion, and when you are able to ask questions specific to your environment and situation.

Really good training is like a partnership with someone who has more experience than you do. It also includes a lot of information that helps inform you as to WHY you would make decisions, be they in the design phase or in working with production systems.

I agree that hands-on learning is also valuable, and I have taught myself quite a bit from on the job experience and from research.

However there is a great deal that I have learned from in-person training that I doubt I would ever have learned from research: having experts help me create a context for the information has really shaped the way I think.]]>Kendra LittleTue, 15 Mar 2011 20:09:13 GMThttp://www.sqlmag.com/article/training-and-certification2/improving-the-sql-server-training-ecosystem#commentsAnchorhttp://www.sqlmag.com/article/training-and-certification2/improving-the-sql-server-training-ecosystem#commentsAnchorTue, 08 Mar 2011 03:35:09 GMT
I agree with Koko. The very best way to learn is dealing with trouble. I’ve been luycky enough to have been exposed to countless anomalies early on in my career, and every incident helped to sharpen my "SQL instincts", so to speak. Sure, if your databases never got corrupted, never went into suspect state, and your servers never crashed, that’s awesome, but you’d at the same time be deprived of a lot of fundamental knowledge.

Similar to a student of a military academy, your knowledge is only part of what you strive and are trained to be. Another is how you can handle yourself under stress.

With that said, I do have great experience with instructor led classes as a crash-course to learning the basics of a product, and when I have the choice and financial means, I most definitely will find the leading expert, and be taught by someone I know has been "battle-scarred" using the product in more than just the textbook ways. Using an expert allows me to ask questions that are not typically in a textbook, such as specifics relating to the specific environment I might be working with.

Taking an actual course in a training center is also fundamentally different than following an onluine training at home or in the office, as a learning center typically will ensure that distractions from the training are minimal, thus allowing you to soak up the knowledge more efficiently.

If training is not an option, I will refer to any material I can get my hands on. Books, online, you name it. I hardly ever take what I read online for granted, but typically will test the accuracy of the information.

The one thing to keep in mind is that your primary goal to training should always be a wish to improve yourself. You should never undertake training or certification for anyone other than yourself. ]]>Peter SchmitzTue, 08 Mar 2011 03:35:09 GMThttp://www.sqlmag.com/article/training-and-certification2/improving-the-sql-server-training-ecosystem#commentsAnchorhttp://www.sqlmag.com/article/training-and-certification2/improving-the-sql-server-training-ecosystem#commentsAnchorSat, 26 Feb 2011 14:27:06 GMT
The best way to learn is to face and solve issues on your own - doing your own research, making your own conclusions and finding the solution yourself.
It is time consuming but the gains are truly impressive over time.]]>KokoSat, 26 Feb 2011 14:27:06 GMThttp://www.sqlmag.com/article/training-and-certification2/improving-the-sql-server-training-ecosystem#commentsAnchorhttp://www.sqlmag.com/article/training-and-certification2/improving-the-sql-server-training-ecosystem#commentsAnchorFri, 25 Feb 2011 03:06:00 GMT
The best ways to learn new IT skills for me, in order of effectiveness:

a) Working hands on with an expert - overwhelmingly so.
b) Self-study including online (because motivation enhances learning).
c) Classroom courses.

Also there is often a lack of basic common sense troubleshooting methodology in highly certified people I have worked with.

Most Microsoft products are rich in features and I would suggest that not all organisations use all the features so it may be better to think about task or scenario based training rather than product based training. Good article.]]>GreensladeFri, 25 Feb 2011 03:06:00 GMThttp://www.sqlmag.com/article/training-and-certification2/improving-the-sql-server-training-ecosystem#commentsAnchorhttp://www.sqlmag.com/article/sql-server/rdbms-vs-nosql-and-the-winner-is-#commentsAnchorThu, 28 Oct 2010 20:25:10 GMT
To me the key is it isn’t "one size fits all" any more. Many readers already use special tools for data warehousing and business intelligence. Why not add one more tool to the toolbox?

NoSQL is not a 100% replacement for relational everytime. For example, ad hoc reporting. With relational, one designs the data model (in theory) independent of the use case. Thus any query in the future, even one not envisioned, will be at least medium easy to write. NoSQL is different. With MongoDB you design your schema for your bread-and-butter use case. This means that use case, the one that happens all day long, will be super fast. But an ad hoc unanticipated query while still doable is not quite as easy.

SQL+relational give us a standard between clients and server. Want to connect to data warehouse with crystal reports or business objects? SQL+relational gives us a clean intermediary. But when writing custom applications that are new code, this is less a concern: we are likely using some object framework anyway. It really depends on the use case.

Relational can scale, but with caveats. For reporting it can scale: there we have a few very expensive queries, so easy to distribute them to all nodes; we don’t have many writes to the database, mostly reads. For online processing scaling is harder. We might need to do something like scale vertically. I have not seen good horizontal scaling on commodity hardware for online processing with RDBMS systems. Also a potential caveat would be that the most scalable products in that space do not have open source versions available.

For me, it’s not all about scale. Flexible schemas create great agility for rapid development. I see a lot of developers with only one or two servers who use MongoDB because they can code faster and make revisions faster.

So don’t throw everything out, just add one more tool. Pick a first project, do a prototype, see what happens. I think one will be pleasantly surprised!

dwight/mongodb]]>merrimanThu, 28 Oct 2010 20:25:10 GMThttp://www.sqlmag.com/article/sql-server/rdbms-vs-nosql-and-the-winner-is-#commentsAnchorhttp://www.sqlmag.com/article/sql-server/is-there-a-shortage-of-sql-server-experts-#commentsAnchorSat, 28 Aug 2010 13:06:14 GMT
What is a SQL Server "expert" and what constitutes "senior level"?

I agree with Poolet’s comments relative to how the product has been positioned.

In my geographical area I see a recurring phenomenon: Organizations need applications for business users so developers are hired to either write or install the app and the app either runs against SQL Server or SQL Server is chosen as the database of choice during develiopment.

In either case the developer, using Mgt Studio, becomes a "database expert" and says "What’s so difficult about this?" Management sees the end result and asks "So why do I need a DBA?"

If a DBA is hired, the developer is there to make claims about SQL Server that the DBA is expected to prove or disprove (rather than the one making the claim be asked to prove/disprove their claim).

I see hardware being selected on the basis of the front-end rather than the disk subsystem and I see database designs with one monster table at the center of the universe and the developer wonders why there is blocking/locking when their systems is launched and accuses the DBA of not knowing their stuff.




]]>John LangstonSat, 28 Aug 2010 13:06:14 GMThttp://www.sqlmag.com/article/sql-server/is-there-a-shortage-of-sql-server-experts-#commentsAnchorhttp://www.sqlmag.com/article/sql-server/is-there-a-shortage-of-sql-server-experts-#commentsAnchorFri, 27 Aug 2010 20:01:08 GMT
I think the question should be "Is there a shortage of SQL DBA’s who can handle multiple skill sets. How many SQL DBA’s can write SQL queries, diagnose performance problems,create DR and HA solution, write SSIS packages, take care of SharePoint administration, know the pros and cons of virtualizatio, learn Powershell... the list goes on and on. The traditional roles are gone and they need to be redefined because of Business Intelligence. Now throw in a 24/7 on call commitment, no wonder everyone wants to be a consultant]]>JohnstonFri, 27 Aug 2010 20:01:08 GMThttp://www.sqlmag.com/article/sql-server/is-there-a-shortage-of-sql-server-experts-#commentsAnchorhttp://www.sqlmag.com/article/sql-server/is-there-a-shortage-of-sql-server-experts-#commentsAnchorFri, 27 Aug 2010 09:39:31 GMT
Now thats the problem "Maybe it’s because Microsoft’s tools (SSAS, SSIS, SSRS) have only been a focus in the past five years SQL Server DBAs are required to know not only the database engine but all aspects of the server i.e. (SSAS, SSIS, SSRS). Are these pure DBA skill sets or added bonuses? So when a hiring manager asks for a DBA to have SSAS, SSIS, SSRS is he/she really asking for a developer. With each one of these skill sets you need to now HTML, VB.NET, ADO.NET and possibly AJAX. When I read job descriptions requesting for a DBA and then list those requires the first two things that come to mind is either they dont know what they want or they are looking for a Developer with some SQL Server Administration. Lets not forget IIS, Windows Sever, RAID, SANS, Cluster, Replication as well as other technologies that are required Does the term DBA mean the same thing it did 10 years ago?]]>BragdonFri, 27 Aug 2010 09:39:31 GMThttp://www.sqlmag.com/article/sql-server/is-there-a-shortage-of-sql-server-experts-#commentsAnchorhttp://www.sqlmag.com/article/sql-server/is-there-a-shortage-of-sql-server-experts-#commentsAnchorFri, 27 Aug 2010 09:32:22 GMT
Now thats the problem "Maybe it’s because Microsoft’s tools (SSAS, SSIS, SSRS) have only been a focus in the past five years SQL Server DBAs are required to know not only the database engine but all aspects of the server i.e. (SSAS, SSIS, SSRS). Are these pure DBA skill sets or added bonuses? So when a hiring manager asks for a DBA to have SSAS, SSIS, SSRS is he/she really asking for a developer. With each one of these skill sets you need to now HTML, VB.NET, ADO.NET and possibly AJAX. When I read job descriptions requesting for a DBA and then list those requires the first two things that come to mind is either they dont know what they want or they are looking for a Developer with some SQL Server Administration. Lets not forget IIS, Windows Sever, RAID, SANS, Cluster, Replication as well as other technologies that are required Does the term DBA mean the same thing it did 10 years ago?]]>BragdonFri, 27 Aug 2010 09:32:22 GMThttp://www.sqlmag.com/article/sql-server/is-there-a-shortage-of-sql-server-experts-#commentsAnchorhttp://www.sqlmag.com/article/sql-server/is-there-a-shortage-of-sql-server-experts-#commentsAnchorFri, 27 Aug 2010 08:44:42 GMT
This article is so true. I’ve interviewed dozens of SQL DBAs over the past few years for my company, and it’s apparent that there is a shortage of senior DBAs. All I have to do is ask how to do a database task by just using T-SQL. 95% of the interviewees can’t do the work when the GUI is out of play.]]>KeislerFri, 27 Aug 2010 08:44:42 GMThttp://www.sqlmag.com/article/sql-server/is-there-a-shortage-of-sql-server-experts-#commentsAnchorhttp://www.sqlmag.com/article/sql-server/is-there-a-shortage-of-sql-server-experts-#commentsAnchorFri, 27 Aug 2010 06:11:20 GMT
It really depends on how you define expert. I know alot of programmers who think they can design databases and many of them can. However, they know nothing about sizing, RAID, proper indexing, etc. Then when they run into performance issues AFTER hardware has been purchased, RAID has been set up and database coded. then they decide DBA is needed to fix it. This is a classic example I see over and over again. DBA’s, regardless of technology are perceived as ’expensive" until the system will not perform.]]>Bulldog99Fri, 27 Aug 2010 06:11:20 GMThttp://www.sqlmag.com/article/sql-server/is-there-a-shortage-of-sql-server-experts-#commentsAnchorhttp://www.sqlmag.com/article/sql-server/is-there-a-shortage-of-sql-server-experts-#commentsAnchorFri, 27 Aug 2010 05:53:48 GMT
]]>KittsFri, 27 Aug 2010 05:53:48 GMThttp://www.sqlmag.com/article/sql-server/is-there-a-shortage-of-sql-server-experts-#commentsAnchorhttp://www.sqlmag.com/article/sql-server/is-there-a-shortage-of-sql-server-experts-#commentsAnchorFri, 27 Aug 2010 04:45:54 GMT
I totally disagreed with such comments I started off with Sybase 4.x over 20 years ago and was introduced to SQL Server 4.2a until now have worked all SQL Server versions up to date. Have got over 10 years experience with the following databases, Ingres, Informix, Oracle, DB2, Sybase, & MySql, etc therefore I know what am talking about. I know people within SQL Server with over 25 years experience but no one will give them the job because they are too experienced. The same company will go for the cheap options then couple of months later they will start experiencing performance issues then they will come back to you for help. I have experienced such attitudes from the big companies for the past 12 years.

I can say that there are more people with Oracle skill-sets in the market than SQL Server for the past 8 years I have migrated more Oracle databases to SQL Server and done only 1 SQL Server migration to Oracle. Most of the big companies are always looking for the cheap options so if he has hire me for $100.00 per hour he will rather go for someone less cheaper what do you think will be the outcome? There are a lot of DIY SQL Server DBAs out there as SQL Server is new compare to Oracle therefore experience is the key. As I said I knew about 5 experienced SQL Server DBAs still looking for job as we speak when they attend interview the output is that they are too experienced which is why I disagreed with suggestion in the first place. Companies must pay for quality not quantity.
]]>WaldenFri, 27 Aug 2010 04:45:54 GMThttp://www.sqlmag.com/article/sql-server/is-there-a-shortage-of-sql-server-experts-#commentsAnchorhttp://www.sqlmag.com/article/sql-server/is-there-a-shortage-of-sql-server-experts-#commentsAnchorThu, 26 Aug 2010 17:54:48 GMT
What’s your definition of an Oracle DBA, and how would you define a SQL Server DBA? I’m not sure we’re talking about the same perception here.

In my experience, Oracle DBAs are like full-time auto mechanics, they focus on keeping the engine humming. SQL Server DBAs are often part-time, accidental, or very new to the job with no one to learn from. This is part culture and part intent -- Oracle pricing has historically been such that you wouldn’t dare let a DBA near the DBMS without months of intense training and/or a senior DBA at their shoulder. SQL Server has been marketed as "every person’s database solution", with the perception that it’s easy to manage and maintain. Maybe that has a lot to do with the perception that SQL Server doesn’t have as many senior people on task as Oracle?]]>PooletThu, 26 Aug 2010 17:54:48 GMThttp://www.sqlmag.com/article/sql-server/is-there-a-shortage-of-sql-server-experts-#commentsAnchorhttp://www.sqlmag.com/article/sql-server/is-there-a-shortage-of-sql-server-experts-#commentsAnchorThu, 26 Aug 2010 14:48:54 GMT
That might be true to a point. But I have definitely seen a large demand for Business Intelligence experience that simply doesn’t seem to be there.

Maybe it’s because Microsoft’s tools (SSAS, SSIS, SSRS) have only been a focus in the past five years.

I’ve seen job postings for BI experienced professionals stay open for months. And a recruiter friend of mine says that these roles are the hardest ones to fill.]]>SwartThu, 26 Aug 2010 14:48:54 GMThttp://www.sqlmag.com/article/sql-server/is-there-a-shortage-of-sql-server-experts-#commentsAnchorhttp://www.sqlmag.com/article/tsql3/tracking-uptime#commentsAnchorMon, 16 Aug 2010 07:35:05 GMT
Nice post. Here is a script to display server uptime in years, days, hours, minutes and seconds:

DECLARE @server_start_time DATETIME,
@seconds_diff INT,
@years_online INT,
@days_online INT,
@hours_online INT,
@minutes_online INT,
@seconds_online INT ;

SELECT @server_start_time = login_time
FROM master.sys.sysprocesses
WHERE spid = 1 ;

SELECT @seconds_diff = DATEDIFF(SECOND, @server_start_time, GETDATE()),
@years_online = @seconds_diff / 31536000,
@seconds_diff = @seconds_diff % 31536000,
@days_online = @seconds_diff / 86400,
@seconds_diff = @seconds_diff % 86400,
@hours_online = @seconds_diff / 3600,
@seconds_diff = @seconds_diff % 3600,
@minutes_online = @seconds_diff / 60,
@seconds_online = @seconds_diff % 60 ;

SELECT @server_start_time AS server_start_time,
@years_online AS years_online,
@days_online AS days_online,
@hours_online AS hours_online,
@minutes_online AS minutes_online,
@seconds_online AS seconds_online ;]]>ColamatteoMon, 16 Aug 2010 07:35:05 GMThttp://www.sqlmag.com/article/tsql3/tracking-uptime#commentsAnchorhttp://www.sqlmag.com/article/sql-server/cloud-computing-it-s-not-quite-ready-but-you-need-to-be#commentsAnchorFri, 23 Jul 2010 09:30:57 GMT
Brian, while I think I understand the point of your article, I still don’t see the difference between having your database in a cloud and having your database in a remote hosting facility. Thanks. ]]>OLOUGHLINFri, 23 Jul 2010 09:30:57 GMThttp://www.sqlmag.com/article/sql-server/cloud-computing-it-s-not-quite-ready-but-you-need-to-be#commentsAnchorhttp://www.sqlmag.com/article/sql-server/cloud-computing-it-s-not-quite-ready-but-you-need-to-be#commentsAnchorThu, 22 Jul 2010 19:30:02 GMT
Brian, you have done an excellent job painting a clear picture of where things are headed. You focus on the justification of moving to the cloud solely from a cost and overall management standpoint however. Another key reason (and equally compelling) for moving to the cloud is the computational performance it provides users. The scalability and raw compute power achievable from being able to divide the work at hand among tens, hundreds or thousands of servers will create incredibly powerful kinds of data analysis, generating new feature sets in our beloved SQL Server stack. These are things we just can’t do today because of the cost of scaling up. The cloud will make them commonplace.

This will also help level the playing field for small and mid-size companies who often can’t afford the infrastruture needed to tackle difficult data problems.

I don’t see a lot of people in the SQL Server community getting this point and I’m not sure why. We should all be really excited at the potential to work with billions and billions of rows of data, data in many forms, in a way that provides timely and insightful analysis.

Thanks,
Ben ]]>borlowitzThu, 22 Jul 2010 19:30:02 GMThttp://www.sqlmag.com/article/sql-server/cloud-computing-it-s-not-quite-ready-but-you-need-to-be#commentsAnchorhttp://www.sqlmag.com/article/sql-server/an-update-on-new-law-that-will-change-the-way-you-build-database-applications#commentsAnchorWed, 30 Jun 2010 04:23:45 GMT
The original article was very informative but this is even better! Thank you for the links to the actual Law. One thing is nor really clear though. Does this law apply outside the US borders?]]>SultanaWed, 30 Jun 2010 04:23:45 GMThttp://www.sqlmag.com/article/sql-server/an-update-on-new-law-that-will-change-the-way-you-build-database-applications#commentsAnchorhttp://www.sqlmag.com/article/sql-server/an-update-on-new-law-that-will-change-the-way-you-build-database-applications#commentsAnchorThu, 24 Jun 2010 16:39:38 GMT
You very briefly touched on it above but I think a good topic for a future article would be to get specific about HOW sql server can be used to comply with the new law. Walk us through how sql server can encrypt communication, how it can encrypt data at rest, etc. Is it possible to comply with the law using sql server as long as it is configured properly?]]>BRENEMANThu, 24 Jun 2010 16:39:38 GMThttp://www.sqlmag.com/article/sql-server/an-update-on-new-law-that-will-change-the-way-you-build-database-applications#commentsAnchorhttp://www.sqlmag.com/article/sql-server/could-unstructured-data-management-technology-replace-the-relational-database-someday-#commentsAnchorMon, 05 Apr 2010 13:32:29 GMT
What you’ve described has been around since the early 1990’s in Lotus Notes. Instead of "records," Notes has "documents" and you can stuff anything into them because they are not defined by columns. If I want to add a new field to just one document, I can do it. If I want to attach files to a document, it’s easy. It’s also quite fun because it gives RDBMS guys conniption fits - they just don’t get it.

Notes does a terrible job simulating structured data, but if structure isn’t required, then for better or worse, it gives you a lot of freedom.

]]>Charles PhillipsMon, 05 Apr 2010 13:32:29 GMThttp://www.sqlmag.com/article/sql-server/could-unstructured-data-management-technology-replace-the-relational-database-someday-#commentsAnchorhttp://www.sqlmag.com/article/sql-server/could-unstructured-data-management-technology-replace-the-relational-database-someday-#commentsAnchorMon, 05 Apr 2010 09:31:06 GMT
Great article, Ive been thinking about this same topic most recently. I’m a MCTS in SQL Server/BI and I too am looking at the cloud technology and what the means for me. More specifically, using SQL Azure vs. Windows Table storage. I may see that as moving from structured to unstructured, but Im in the novice stage of the cloud, so Im still learning. One area Im trying to learn more about is using the Entity/Properties and getting the data I need in the same Partition in the cloud, etc. It does look like I need to embrace the Borg.]]>BARKERMon, 05 Apr 2010 09:31:06 GMThttp://www.sqlmag.com/article/sql-server/could-unstructured-data-management-technology-replace-the-relational-database-someday-#commentsAnchorhttp://www.sqlmag.com/article/sql-server/what-are-your-nested-views-doing-#commentsAnchorWed, 31 Mar 2010 18:21:28 GMT
Case in Point:

(Hope this link posts)

http://www.clayarnaud.com/viewhierarchy/viewhierarchy.jpg

I wrote this utility to decipher this wacky view thing they have going on where I work. Checkout the node count on the status bar for View 436. It is a complete mess.
]]>ClayWed, 31 Mar 2010 18:21:28 GMThttp://www.sqlmag.com/article/sql-server/what-are-your-nested-views-doing-#commentsAnchorhttp://www.sqlmag.com/article/sql-server/what-are-your-nested-views-doing-#commentsAnchorWed, 31 Mar 2010 18:14:16 GMT
]]>ClayWed, 31 Mar 2010 18:14:16 GMThttp://www.sqlmag.com/article/sql-server/what-are-your-nested-views-doing-#commentsAnchorhttp://www.sqlmag.com/article/sql-server/could-unstructured-data-management-technology-replace-the-relational-database-someday-#commentsAnchorFri, 26 Mar 2010 11:28:27 GMT
While user/DBA involvement may change, relational data management will never go away because it’s based on the mathematical reality that some data elements are related to other data elements in one-to-one, one-to-many, or many-to-many relationships. Advanced management of erroneously named "unstructured" data (which is actually highly structured) is primarily based on deducing and extracting metadata that is managed in a relational manner. Data management engines may change radically, and artificially intelligent data systems may eventually make database administrators obsolete, but even then, most data with simple relationships will still be managed with relational rules.]]>John D. LambertFri, 26 Mar 2010 11:28:27 GMThttp://www.sqlmag.com/article/sql-server/could-unstructured-data-management-technology-replace-the-relational-database-someday-#commentsAnchorhttp://www.sqlmag.com/article/sql-server/could-unstructured-data-management-technology-replace-the-relational-database-someday-#commentsAnchorThu, 25 Mar 2010 16:57:38 GMT
I think of it this way: in a hospital we need the relational data structure for most of the data. But the strategic analysis is at some point unstructured. Basically, you have a lot of reports and it is the results of those reports (with some parameterization) that are the unstructured data. What are top 10 meds for AMI patients? Compare that data last quarter versus a year ago. What’s the difference in costs? in length of stay? in mortality? Any physicians stand out? Any other big change in how we treat these patients?
Now take the beginning of this list of questions - what data structure answers all of them the best? A cube? An RDBMS? Or something that has a different structure OR is largely unstructured yet ready for these kinds of questions?]]>KohlmillerThu, 25 Mar 2010 16:57:38 GMThttp://www.sqlmag.com/article/sql-server/could-unstructured-data-management-technology-replace-the-relational-database-someday-#commentsAnchorhttp://www.sqlmag.com/article/sql-server/could-unstructured-data-management-technology-replace-the-relational-database-someday-#commentsAnchorThu, 25 Mar 2010 16:54:26 GMT
Couldn’t agree more. Increasingly, corporate data is being created outside of ledgers and other similar structured formats, and in volumes which challenge traditional architectures. Can’t see RDBMS being mothballed any time soon because of the rigour within their underlying model, but their slice of the data management pie will certainly decrease.]]>Geoff AmblerThu, 25 Mar 2010 16:54:26 GMThttp://www.sqlmag.com/article/sql-server/could-unstructured-data-management-technology-replace-the-relational-database-someday-#commentsAnchorhttp://www.sqlmag.com/article/sql-server/page-life-expectancy-a-reliable-indicator-of-sql-server-memory-pressure#commentsAnchorMon, 22 Mar 2010 14:51:00 GMT
IAN, do you have a 64-bit version of SQL Server? Because if you don’t - that’s the problem.]]>Mon, 22 Mar 2010 14:51:00 GMThttp://www.sqlmag.com/article/sql-server/page-life-expectancy-a-reliable-indicator-of-sql-server-memory-pressure#commentsAnchorhttp://www.sqlmag.com/article/security/facebook-s-new-settings-raise-data-privacy-questions#commentsAnchorWed, 17 Mar 2010 12:28:25 GMT
I use Facebook to keep up with my far flung family members but I try to be as carefull as possible though I know it is risky.
One of the reasons Facebook is especially dangerous is one of your "Friends" can play one of a gazillion 3rd Party Apps that asks them to reveal private information about you and you have absolutely no way to control this--and you do not have to have a Facebook account to fall prey to this invasion of your privacy.
I have tried to warn my friends on Facebook but it is a waste of time. Since I do not use Facebook Apps not even to accept a "gift" no doubt some think I am being rude.
I am just waiting for a huge Facebook security scandal to make the news because it is surely coming. ]]>Pat WoodWed, 17 Mar 2010 12:28:25 GMThttp://www.sqlmag.com/article/security/facebook-s-new-settings-raise-data-privacy-questions#commentsAnchorhttp://www.sqlmag.com/article/business-intelligence/microsoft-s-mark-souza-lists-his-favorite-sql-server-2008-r2-features#commentsAnchorTue, 16 Mar 2010 03:26:52 GMT
I think streamInsight is a feature which has bulk of stuff to work with dlls and coding needs to be done in windows or web apps. What made to include this feature as part of SQL Server? Am I missing something significant with respect to the role that SQL Server is playing in enabling the StreamInsight functionality? Please throw some light on this.]]>SubnivisTue, 16 Mar 2010 03:26:52 GMThttp://www.sqlmag.com/article/business-intelligence/microsoft-s-mark-souza-lists-his-favorite-sql-server-2008-r2-features#commentsAnchorhttp://www.sqlmag.com/article/security/facebook-s-new-settings-raise-data-privacy-questions#commentsAnchorThu, 28 Jan 2010 15:43:50 GMT
Sadly Facebook and other social media just gives a new generation their "Boob Tube" (Television) to waste countless hours. Abraham Lincoln was quoted as saying, "Better to remain silent and be thought a fool than to speak out and remove all doubt." Facebook gives millions of people on a daily basis the ability to "remove all doubt".]]>JeffreyThu, 28 Jan 2010 15:43:50 GMThttp://www.sqlmag.com/article/security/facebook-s-new-settings-raise-data-privacy-questions#commentsAnchorhttp://www.sqlmag.com/article/sql-server/database-sizing-tools-for-sql-server-#commentsAnchorFri, 22 Jan 2010 14:55:33 GMT
Hello, Thank you for the great article! I’m in need of a sizing tool for the database. Looking for information on HEAP size be for a certain row size. Looks like hardware sizing is easier that sizing the database. But size the database properly is very important before even sizing the hardware. Thank you in advance for your time and efforts, Rudy]]>RudyFri, 22 Jan 2010 14:55:33 GMThttp://www.sqlmag.com/article/sql-server/database-sizing-tools-for-sql-server-#commentsAnchorhttp://www.sqlmag.com/article/scripting/sort-order-with-the-uniqueidentifier-data-type#commentsAnchorTue, 08 Dec 2009 13:13:50 GMT
no solution to sort correctly. (order created)]]>RexTue, 08 Dec 2009 13:13:50 GMThttp://www.sqlmag.com/article/scripting/sort-order-with-the-uniqueidentifier-data-type#commentsAnchorhttp://www.sqlmag.com/article/sql-server/page-life-expectancy-a-reliable-indicator-of-sql-server-memory-pressure#commentsAnchorFri, 30 Oct 2009 11:41:53 GMT
so how do you fix this issue if you have 8GB of memory but the instance is only using 30% of the avaliable memory and you page life expectancy is still at 50?]]>IANFri, 30 Oct 2009 11:41:53 GMThttp://www.sqlmag.com/article/sql-server/page-life-expectancy-a-reliable-indicator-of-sql-server-memory-pressure#commentsAnchorhttp://www.sqlmag.com/article/san/are-you-a-san-fan-#commentsAnchorMon, 26 Oct 2009 21:30:01 GMT
I am for all practical purposes a systems architect -- not the programming kind mind you. With assistance from our HW vendor, we’ve built & designed our SANS based on all current data we could gather from our SQL servers. We run almost all SQL servers as clusters, with RAID 10 FC disks on SAN. While our DB guys could give us rough numbers, I had to go in and start measuring disk I/O, IOPS, response times, etc... the whole gamut of OS specifics to size the number of spindles. I don’t just look at space. I look at everything in the entire system & try to get an understanding from the Developers building the application to the DBAs who managed the SQL servers. With all that and input from our HW vendor, we’ve had good reasons to use SANS. ON several DBs, the ability to support high I/O & IOPS that we couldn’t do with DASD. So far, knocking on wood, any SQL performance problems have all been at the database design level, (table structure, indexes, bad queries, etc...). The physical servers & the SAN have not been a bottlenecks.... much to the dismay of the Director over the developers. Excellent article though. Throwing HW at an application 1st should be the LAST thing you do get performance out of a system.]]>ScottMon, 26 Oct 2009 21:30:01 GMThttp://www.sqlmag.com/article/san/are-you-a-san-fan-#commentsAnchorhttp://www.sqlmag.com/article/san/are-you-a-san-fan-#commentsAnchorFri, 23 Oct 2009 17:06:51 GMT
I work for one of those SAN vendors, but I am not the SAN guy. I’m the DBA engineer helping customers to define their application requirements so the SAN guys can deliver the performance requirements. Here’s the catch: Most application guys (and DBAs included) can’t verbalize what their performance requirements are today, let alone what they will be in 12 months. The SAN technology that my company produces provides for this in many cool and cutting edge ways. Our technology can actually compensate for the lack of current knowledge and the lack of future crystal ball knowledge. Our technology also provides for a 5x9’s ++plus storage platform and a management/replication framework that can move data around within the SAN faster and more efficiently than any database layer technology can perform the same functionality. This provides the business a much more robust and agile storage platform to drive application changes faster for a more competitive advantage. You’ll notice that none of this has anything to do with I/O’s and Bandwidth, but we do excel in those areas, as well. In most cases we as database engineers like to think about the single instance running the monster database, but SANs are deployed to support dozens or hundeds of instances and hundreds and thousands of databasess, as well as that one monster database. SANs are also used to clone/snap that monster database and to back up that monster database and then to replicate that monster database to the disaster recovery site, all without DBA/user intervention. There is a place for SANs. I would like to suggest that the root of this discussion is really all about communication and cooperation and understanding between all of the smart people required to run the IT infrastructure. Let’s face the truth. We’re all nerds, and nerds aren’t known for being the best at communicating with each other. Brian, Keep up the great work, and Jimmy, I love your work too!]]>RandallFri, 23 Oct 2009 17:06:51 GMThttp://www.sqlmag.com/article/san/are-you-a-san-fan-#commentsAnchorhttp://www.sqlmag.com/article/san/are-you-a-san-fan-#commentsAnchorFri, 23 Oct 2009 04:00:28 GMT
Isn’t this problem going to be mitigated by SSD technology? Texas Storagesystem have SSD-SAN’s that can take 100.000 IOPS. There’s happing a lot in this space, a small revolution you might say?]]>HenrikFri, 23 Oct 2009 04:00:28 GMThttp://www.sqlmag.com/article/san/are-you-a-san-fan-#commentsAnchorhttp://www.sqlmag.com/article/san/are-you-a-san-fan-#commentsAnchorThu, 22 Oct 2009 17:21:22 GMT
Nicely stated, Brian. I just received an email from a customer who, for his application from which high-performance is required, got this from his outsourced SAN vendor: "4.5TB RAID 5 split across unknown number of shared spindles...should be more than 8 spindles..." The mantra I preach to my customers is: X GB capacity at Y throughput (MB/s) at Z IOPs (transfers/s) at <=10ms latency for data files, 0ms-2ms for log files A shake my cavernous calvarium in sadness at SAN admins who deal solely in capacity, which is merely the first of several vital stats required for db performance-&-scalability. Here’s a message from another customer to whom I recommended SSDs. It was essential to save his project: "...As you know, in order to get out of the SAN team’s prison, I now have non clustered servers operating with local storage which is combination of local hard drivers and FusionIO. In order to do that, I have selected servers which offer many local drives 24 Drives (HP DL 580) and SSD drives goes in PCI-E Slots..." This is an interesting combination of frank relief & practical advice. Jimmy May http://blogs.msdn.com/jimmymay]]>JimmyThu, 22 Oct 2009 17:21:22 GMThttp://www.sqlmag.com/article/san/are-you-a-san-fan-#commentsAnchorhttp://www.sqlmag.com/article/san/are-you-a-san-fan-#commentsAnchorThu, 22 Oct 2009 15:21:41 GMT
Excellent vision and guidance Thank you Husain Taiyebi]]>HusainThu, 22 Oct 2009 15:21:41 GMThttp://www.sqlmag.com/article/san/are-you-a-san-fan-#commentsAnchorhttp://www.sqlmag.com/article/san/are-you-a-san-fan-#commentsAnchorThu, 22 Oct 2009 15:09:15 GMT
I agree with all four points, and tell clients the same. Bravo for writing this.]]>JonathanThu, 22 Oct 2009 15:09:15 GMThttp://www.sqlmag.com/article/san/are-you-a-san-fan-#commentsAnchorhttp://www.sqlmag.com/article/san/are-you-a-san-fan-#commentsAnchorThu, 22 Oct 2009 15:08:01 GMT
Gee Brian -- sure wish you would take stand :))))]]>MikeThu, 22 Oct 2009 15:08:01 GMThttp://www.sqlmag.com/article/san/are-you-a-san-fan-#commentsAnchorhttp://www.sqlmag.com/article/performance/should-i-defragment-my-sql-server-data-files-#commentsAnchorWed, 21 Oct 2009 14:05:54 GMT
Is it necessary to stop SQL Server to defrag a volume that contains a database datafiles?]]>MarcosWed, 21 Oct 2009 14:05:54 GMThttp://www.sqlmag.com/article/performance/should-i-defragment-my-sql-server-data-files-#commentsAnchorhttp://www.sqlmag.com/article/sql-server/sqlserverbeta-com-provides-dbas-with-a-free-sql-server-2008-instance#commentsAnchorWed, 14 Oct 2009 14:32:31 GMT
Hello Brian, I registered with http://beta4sqlserver.com/ about 3 weeks ago. And once a week since. I haven’t received confirmation of my registration or connection information. The value of this article itself and the free SQL service offer decreases with every day I am unable to make a connection. I am not sure that e-mail to their admin was ever delivered. The mailserver said delivery was delayed. What is the next step. E-mail Dell and Maximum ASP?? Is high demand delaying e-mail containing connection information? I’ve been looking, but I haven’t seen any connection information for the Beta For SQL Server for a few weeks. I have registered several times during the past month; about once a week since the end of September 2009. Best regards, John]]>johnawman Wed, 14 Oct 2009 14:32:31 GMThttp://www.sqlmag.com/article/sql-server/sqlserverbeta-com-provides-dbas-with-a-free-sql-server-2008-instance#commentsAnchorhttp://www.sqlmag.com/article/virtualization/will-future-generations-take-virtualization-for-granted-#commentsAnchorFri, 25 Sep 2009 11:02:27 GMT
Excellent article. I am waiting to see if SQL 2008 Clusterred Env to get into Virtualization family without loosing the performance (which is the key) Thank You]]>HusainFri, 25 Sep 2009 11:02:27 GMThttp://www.sqlmag.com/article/virtualization/will-future-generations-take-virtualization-for-granted-#commentsAnchorhttp://www.sqlmag.com/article/sql-server/exploring-the-auto-update-statistics-option#commentsAnchorTue, 08 Sep 2009 13:10:59 GMT
not enough info]]>surendraTue, 08 Sep 2009 13:10:59 GMThttp://www.sqlmag.com/article/sql-server/exploring-the-auto-update-statistics-option#commentsAnchorhttp://www.sqlmag.com/article/sql-server/page-life-expectancy-a-reliable-indicator-of-sql-server-memory-pressure#commentsAnchorMon, 27 Jul 2009 23:48:43 GMT
short but sweet]]>MARKMon, 27 Jul 2009 23:48:43 GMThttp://www.sqlmag.com/article/sql-server/page-life-expectancy-a-reliable-indicator-of-sql-server-memory-pressure#commentsAnchorhttp://www.sqlmag.com/article/sql-server/cloud-computing-how-will-it-affect-corporate-it-#commentsAnchorFri, 26 Jun 2009 13:51:21 GMT
@guzmand - I agree with your thought that clouds will encourage poor db design but perhaps that will be kept in check through usage costs, if you assume a poorly designed system uses more resources which turn into higher cloud costs. I think the challenge may become similar to disk i/o planning with a SAN or virtualization, places where it’s easy to abstract things to the point you might not know all you’d like to when troubleshooting or tuning. For small/medium things it may not be a problem for the cloud, just like they might not see a performance difference between raid 5 and 10 on a local machine, bigger or more intensive apps may be different. But the only absolute is there are no absolutes - mainframes never went away and even regained attention recently for virtualization. In some aspects I see the cloud as just another name for application service providers, or hosted data centers.]]>FoxTrotFri, 26 Jun 2009 13:51:21 GMThttp://www.sqlmag.com/article/sql-server/cloud-computing-how-will-it-affect-corporate-it-#commentsAnchorhttp://www.sqlmag.com/article/sql-server/cloud-computing-how-will-it-affect-corporate-it-#commentsAnchorFri, 26 Jun 2009 03:07:44 GMT
Handling of sensitive data will be one factor limiting the use of cloud based databasess. It would at least require some stiff guarantees from the providers before this would get past the auditors. Another limiting factor is the speed of light. For time critical applications you would need to get some kind of guarantee about where in the clould you hit the physical server (e.g. on the same continent).]]>TroelsFri, 26 Jun 2009 03:07:44 GMThttp://www.sqlmag.com/article/sql-server/cloud-computing-how-will-it-affect-corporate-it-#commentsAnchorhttp://www.sqlmag.com/article/sql-server/cloud-computing-how-will-it-affect-corporate-it-#commentsAnchorThu, 25 Jun 2009 17:05:51 GMT
What I’m seeing with the growth of virtualization of Windows Servers running on ESX host is that the organization\company still has a need for an Systems Engineer that can manage the database, admin the IIS\web server and also deploy code through to integration, then QA and finally to production environment(s). He is the gatekeeper of the enviroment(s) that the applications reside and maintains a level of consistency with Change Management to ensure that Service level agreements(SLA’s) between IT and the business are met.]]>MattThu, 25 Jun 2009 17:05:51 GMThttp://www.sqlmag.com/article/sql-server/cloud-computing-how-will-it-affect-corporate-it-#commentsAnchorhttp://www.sqlmag.com/article/sql-server/cloud-computing-how-will-it-affect-corporate-it-#commentsAnchorThu, 25 Jun 2009 15:12:59 GMT
I think this is the dream of every software developer, that increased size does not equal increased knob wiggling. True scalability. I think what this will encourage is lazy database and other app design. I won’t matter if it performs poorly, the cloud team will just have to throw more resources at it. Or perhaps instead, all databases will have to be submitted to the cloud managers for review before it gets loaded, if so, that’s where all the staff would end up.]]>DanThu, 25 Jun 2009 15:12:59 GMThttp://www.sqlmag.com/article/sql-server/cloud-computing-how-will-it-affect-corporate-it-#commentsAnchorhttp://www.sqlmag.com/article/backup-recovery/are-your-backups-useless-#commentsAnchorWed, 17 Jun 2009 11:52:08 GMT
Raghu, I couldn’t have said it better. Please, do rate the articles so we don’t waste our time. Thanks.]]>MarcosWed, 17 Jun 2009 11:52:08 GMThttp://www.sqlmag.com/article/backup-recovery/are-your-backups-useless-#commentsAnchorhttp://www.sqlmag.com/article/tsql3/retrieving-sql-server-s-last-start-date-and-time#commentsAnchorFri, 22 May 2009 09:16:34 GMT
Here’s a comment 8 years later... better late than never: There is now a Microsoft Connect item on this topic. I encourage readers to vote for it. https://connect.microsoft.com/SQLServer/feedback/ViewFeedback.aspx?FeedbackID=316237]]>MichaelFri, 22 May 2009 09:16:34 GMThttp://www.sqlmag.com/article/tsql3/retrieving-sql-server-s-last-start-date-and-time#commentsAnchorhttp://www.sqlmag.com/article/performance/designing-for-your-workload-don-t-wait-until-the-last-minute#commentsAnchorThu, 21 May 2009 12:33:53 GMT
As always, most of the systems are build to deal with the average. Specialy because it is too expensive the design to extreme situations. I understand that, so the client should understand too. If not, when the client comes back complayning, show him what the bill will be to avoid that situation, if he agrees to pay, give him the perfect system he deserves! Take Care.]]>MarcosThu, 21 May 2009 12:33:53 GMThttp://www.sqlmag.com/article/performance/designing-for-your-workload-don-t-wait-until-the-last-minute#commentsAnchorhttp://www.sqlmag.com/article/data-management/understanding-the-fast-track-data-warehouse-reference-architectures#commentsAnchorThu, 14 May 2009 09:43:59 GMT
q]]>AlexisThu, 14 May 2009 09:43:59 GMThttp://www.sqlmag.com/article/data-management/understanding-the-fast-track-data-warehouse-reference-architectures#commentsAnchorhttp://www.sqlmag.com/article/tsql3/retrieving-sql-server-s-last-start-date-and-time#commentsAnchorMon, 04 May 2009 12:28:56 GMT
This is precisely the answer for which I was looking.]]>BrianMon, 04 May 2009 12:28:56 GMThttp://www.sqlmag.com/article/tsql3/retrieving-sql-server-s-last-start-date-and-time#commentsAnchorhttp://www.sqlmag.com/article/oracle/what-could-the-oracle-sun-deal-mean-for-the-database-market-#commentsAnchorMon, 27 Apr 2009 12:18:23 GMT
If Oracle wants to optimize their products for Solaris that’s great for them but most of the "bread and butter" for Oracle is the sales on Windows platforms. More that 80% of sales are on Windows OS (this from a high level Oracle sales person). I don’t think it’s a problem for Microsoft unless Oracle makes Java only work on Solaris. My biggest fear is that MySQL will be changes to MyOracle and mess up the great standings it has in the Linux world. We’ll see in the future but don’t be surprise if Oracle sells off the hardware divisions to someone else and just keep Java. Or there will be a version of Oracle that run on the Cray computer (I think Sun still owns them). Just my 2 cents worth.]]>RudyMon, 27 Apr 2009 12:18:23 GMThttp://www.sqlmag.com/article/oracle/what-could-the-oracle-sun-deal-mean-for-the-database-market-#commentsAnchorhttp://www.sqlmag.com/article/tsql3/questions-answers-tips#commentsAnchorWed, 18 Mar 2009 14:09:19 GMT
Nice information. Good to know undocumented stored procedures. Thanks.]]>MarcosWed, 18 Mar 2009 14:09:19 GMThttp://www.sqlmag.com/article/tsql3/questions-answers-tips#commentsAnchorhttp://www.sqlmag.com/article/sql-server/using-datetime-columns-as-primary-keys#commentsAnchorWed, 18 Mar 2009 10:42:34 GMT
Good tip, besides this topic has been mentioned in other articles. Looks like the magazine keeps comming back to the same subject from time to time. Is this due a lack of subject? Regards.]]>MarcosWed, 18 Mar 2009 10:42:34 GMThttp://www.sqlmag.com/article/sql-server/using-datetime-columns-as-primary-keys#commentsAnchorhttp://www.sqlmag.com/article/sql-server/the-devil-s-in-the-dewitt-clause#commentsAnchorThu, 19 Feb 2009 09:46:17 GMT
Those arguments are pretty weak. They are analogous to arguing against free speech by saying "people might lie" and "determining the truth about some things is hard."]]>EricThu, 19 Feb 2009 09:46:17 GMThttp://www.sqlmag.com/article/sql-server/the-devil-s-in-the-dewitt-clause#commentsAnchorhttp://www.sqlmag.com/article/performance/using-sql-server-profiler-to-tune-your-applications#commentsAnchorThu, 05 Feb 2009 09:24:55 GMT
Hello]]>RichardThu, 05 Feb 2009 09:24:55 GMThttp://www.sqlmag.com/article/performance/using-sql-server-profiler-to-tune-your-applications#commentsAnchorhttp://www.sqlmag.com/article/performance/designing-for-your-workload-don-t-wait-until-the-last-minute#commentsAnchorMon, 17 Nov 2008 06:06:05 GMT
Writing an artical to explaining the issues could never be a thrust of a reader. Writer should try to explain with sample solution or atleast tell the best practicies in such situation. Have a good day. Cheers!!!]]>mavoice Mon, 17 Nov 2008 06:06:05 GMThttp://www.sqlmag.com/article/performance/designing-for-your-workload-don-t-wait-until-the-last-minute#commentsAnchorhttp://www.sqlmag.com/article/data-management/life-liberty-and-the-pursuit-of-quality-databases#commentsAnchorFri, 24 Oct 2008 12:17:24 GMT
Justin is absolutely correct. You should also include links to support some of your "technical" commentary as well.]]>KENANFri, 24 Oct 2008 12:17:24 GMThttp://www.sqlmag.com/article/data-management/life-liberty-and-the-pursuit-of-quality-databases#commentsAnchorhttp://www.sqlmag.com/article/data-management/life-liberty-and-the-pursuit-of-quality-databases#commentsAnchorFri, 24 Oct 2008 06:15:51 GMT
Well Brian - what have you been reading? How about including a couple of links to some of the articles you’ve "been reading in the news"? Your point may be valid, but it would be much more useful to your readers to point us in the direction of the sources that caused you to write this diatribe. Maybe one link to meaningful information for every five inane advertising links to keywords like "database", "computer" and "bank accounts".]]>JustinFri, 24 Oct 2008 06:15:51 GMThttp://www.sqlmag.com/article/data-management/life-liberty-and-the-pursuit-of-quality-databases#commentsAnchorhttp://www.sqlmag.com/article/performance/performance-effects-of-using-guids-as-primary-keys#commentsAnchorThu, 23 Oct 2008 10:29:12 GMT
Eugene, thanks for pointing out the missing links. I’ve updated them with the correct URLS in the article. The links are: The Cost of GUIDs as Primary Keys, http://www.informit.com/articles/article.aspx?p=25862 Gert Drapers’ XPGUID: http://www.sqldev.net/xp/xpguid.htm]]>AnneThu, 23 Oct 2008 10:29:12 GMThttp://www.sqlmag.com/article/performance/performance-effects-of-using-guids-as-primary-keys#commentsAnchorhttp://www.sqlmag.com/article/performance/performance-effects-of-using-guids-as-primary-keys#commentsAnchorWed, 22 Oct 2008 13:08:57 GMT
Good, simple article that’s easy to understand. However, I’ve tried to access the two links mentioned in this article to get further information with no success.]]>eugenetsai Wed, 22 Oct 2008 13:08:57 GMThttp://www.sqlmag.com/article/performance/performance-effects-of-using-guids-as-primary-keys#commentsAnchorhttp://www.sqlmag.com/article/sql-server-2005/real-best-practices#commentsAnchorTue, 14 Oct 2008 10:35:21 GMT
I hope they do a similar exploration for SQL 2008.]]>DANTue, 14 Oct 2008 10:35:21 GMThttp://www.sqlmag.com/article/sql-server-2005/real-best-practices#commentsAnchorhttp://www.sqlmag.com/article/conferences-and-trade-shows/decisions-decisions-which-sql-server-conference-will-you-attend-#commentsAnchorFri, 26 Sep 2008 16:11:43 GMT
Hi! Do you have a list of SQL Server conferences for 2009? It’s time to make plans! Thanks]]>verenaFri, 26 Sep 2008 16:11:43 GMThttp://www.sqlmag.com/article/conferences-and-trade-shows/decisions-decisions-which-sql-server-conference-will-you-attend-#commentsAnchorhttp://www.sqlmag.com/article/benchmarks/microsoft-still-the-only-database-vendor-posting-tpc-e-scores#commentsAnchorThu, 25 Sep 2008 17:55:24 GMT
Thank you for this article which shows the full story. I blogged about this a few days ago, at which time I was amazed that the gushing Microsoft Press release about these results was all that was getting repeated. http://www.craigbailey.net/live/post/2008/09/22/SQL-SQL-Server-2008-is-a-value-leader-(but-in-what).aspx I love SQL Server, but sometimes their press department gets carried away :-) I’ll be interested to hear the IBM and Oracle story too... if it ever gets released.]]>CraigThu, 25 Sep 2008 17:55:24 GMThttp://www.sqlmag.com/article/benchmarks/microsoft-still-the-only-database-vendor-posting-tpc-e-scores#commentsAnchorhttp://www.sqlmag.com/article/performance/reducing-the-page-file-size#commentsAnchorFri, 12 Sep 2008 13:50:26 GMT
No content]]>DonFri, 12 Sep 2008 13:50:26 GMThttp://www.sqlmag.com/article/performance/reducing-the-page-file-size#commentsAnchor